Published as doi: 10.1096/fj.06-5989fje.
(The FASEB Journal. 2006;20:2654-2656.)
© 2006 FASEB
Leptin signaling in neurotensin neurons involves STAT, MAP kinases ERK1/2, and p38 through c-Fos and ATF1
Hong Cui*,
Fang Cai* and
Denise D. Belsham*,
,1
Departments of
* Physiology,
Obstetrics and Gynaecology, and Medicine, University of Toronto, Toronto, Ontario, Canada; and Division of Cellular and Molecular Biology, Toronto General Hospital Research Institute, University Health Network, Toronto, Ontario, Canada
1Correspondence: Department of Physiology, University of Toronto, Medical Sciences Bldg. 3247A, 1 King’s College Cir., Toronto, ON M5S 1A8, Canada. E-mail: d.belsham{at}utoronto.ca
SPECIFIC AIMS
The adipokine leptin signals energy status to the hypothalamus, which triggers a network of neuropeptide responses. Each hypothalamic cell type expresses a unique complement of neuropeptides, receptors, and second messengers; thus each likely responds specifically to peripheral hormones. The aims of this study were to 1) define leptin signaling in a clonal population of mouse neurotensin (NT)-expressing hypothalamic neurons, N-39, and 2) link these specific signaling events directly to leptin-mediated NT transcriptional regulation.
PRINCIPAL FINDINGS
1. Leptin induces the STAT3 and MAPK ERK1/2 pathways and c-fos protein synthesis, but not the PI3-K/Akt pathway
In a previous study we observed that 10–11 M (0.01 nM) and 10–7 M (100 nM) leptin significantly induced NT gene expression in clonal hypothalamic N-39 cells that endogenously express the long form of the leptin receptor, ObRb. To define which pathways are activated by leptin, we determined by Western blot analysis the status of STAT3, ERK1/2, and AKT phosphorylation on leptin stimulation of ObRb. Leptin (10–11 M or 10–7 M) significantly induced phosphorylation of STAT3 and ERK 1/2. However, there was no significant increase in Akt phosphorylation with leptin treatment. We also found that c-Fos protein was induced by leptin.
2. p38 MAPK and ATF-1 are activated by leptin
To determine whether leptin regulates p38 MAPK in NT-expressing hypothalamic neurons, we examined the level of p38 MAPK phosphorylation. Exposure of N-39 neurons to leptin (10–11 M or 10–7 M) resulted in a significant increase in the level of p38 MAPK phosphorylation (Fig. 1
A). Leptin (10–11 M or 10–7 M) significantly induced phosphorylation of ATF-1 but failed to affect phosphorylation of CREB (Fig. 1B
), effectors of p38 MAPK. ATF-1 is a CREB-related transcription factor characterized by a b-ZIP DNA binding domain. The CREB antibody (Ab) produces two bands and recognizes CREB (upper band) and ATF-1 (lower band).
View larger version (52K):
[in this window]
[in a new window]
|
Figure 1. Leptin induces phosphorylation of p38 MAPK and ATF-1, but not CREB. N-39 neurons were stimulated with or without leptin for the indicated concentrations and time. Cell lysates were examined by Western blot with phosphospecific and total antibodies for each respective protein, A) p38 and B) CREB/ATF-1.
|
|
3. Leptin activates ATF-1 via p38 MAPK, while induction of NT gene expression may be linked to a number of signaling pathways
Although it is known that CREB/ATF-1 are downstream substrates of p38 MAPK, other signaling pathways have been shown to phosphorylate CREB/ATF-1. To determine whether p38 MAPK activation results in ATF-1 phosphorylation and plays a role in cAMP response element (CRE) -dependent gene expression, we used the p38 MAPK inhibitor, SB 203580. Leptin-stimulated ATF-1 phosphorylation was blocked by SB 203580 (Fig. 2
A) but CREB phosphorylation was unaffected. To determine whether the induction of NT gene expression by leptin was mediated by p38/ATF-1 phosphorylation, N-39 neurons were treated with or without leptin in the presence or absence of three pharmacological inhibitors of p38: SB 203580, SB 202190, and SB 239063. Leptin induced neurotensin gene expression and the JAK2 inhibitor AG490 blocked leptin-mediated induction of NT gene expression in N-39 neurons (Fig. 2B
). SB 203580 alone induced NT gene expression by 2.2-fold (Fig. 2B
), indicating it may have nonspecific effects on the regulation of basal NT transcription. SB 203580 blocked induction of NT at both concentrations of leptin; SB 202190 and SB 239063 both blocked leptin-mediated induction of NT transcription with no change in basal NT gene expression (Fig. 2B
). U0126, a MEK1/2 inhibitor, blocked downstream effects of ERK1/2 and induction of NT, but only at 10–7 M leptin (Fig. 2B
). Although we found no increased phosphorylation of Akt on leptin treatment, the inhibitor of PI3K LY294002 appears to inhibit the leptin-mediated increase in NT gene expression (Fig. 2B
).
4. Leptin induces enhanced binding of protein to the CRE/activating protein-1 response element of the NT/N promoter
To study NT gene regulation by leptin at the transcriptional level and whether ATF-1 or c-Fos binding were involved, we focused on a region of the 5' flanking region of the mouse NT gene that contains a CRE/activating protein-1 site. In this study, the CRE/activating protein-1-like motif (TGACATCA) at –60 to –37 was of particular interest due to the induction of c-Fos protein and ATF-1 phosphorylation by leptin and its putative involvement in NT gene regulation. Using EMSA with a CRE/activating protein-1 oligonucleotide from –60 to –37 of the mouse NT/N gene, we detect four protein complexes with N-39 nuclear extracts, two of which appear to be regulated by leptin exposure. We demonstrate that nuclear extracts from N-39 neurons treated with leptin (10–11 M or 10–7 M; 15 min) increased complex formation compared with nuclear extracts from untreated cells.
Ab supershift assays using leptin-treated nuclear extracts from N-39 cells indicate that ATF-1 and c-Fos antibodies are capable of producing a slower mobility complex or a supershift. Chromatin immunoprecipitation (ChIP) was used to demonstrate whether the CRE/activating protein-1 region of the NT/N promoter was active in cell culture and if leptin would affect in vivo binding of the transcription factors to these elements. ChIP analysis indicates that leptin induces enhanced binding of ATF-1 or c-Fos to the –252 to –36 bp region of the NT/N gene containing the CRE/activating protein-1 element, as detected by PCR. These results suggest that ATF-1 and c-Fos may be involved in the leptin-mediated induction of NT gene expression at the consensus CRE/activating protein-1 site (Fig. 3
).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 3. Schematic diagram of signal transduction pathways induced by leptin in the clonal, hypothalamic NT cell model, N-39. In N-39, we provide evidence for activation of three specific signal transduction pathways by leptin, all dependent on JAK2 signaling. ERK1/2 and p38 are induced by leptin receptor activation. STAT3 is also involved in leptin-mediated regulation of NT/N gene expression; we describe its binding to the mouse NT/N promoter region. We describe the involvement of downstream activators c-Fos and ATF-1, and speculate that JAK/STAT and p38 MAPK pathways are predominant at physiological leptin levels whereas the ERK1/2 MAPK pathway may be utilized more prominently at higher leptin concentrations.
|
|
CONCLUSIONS
Leptin induces its effects through the ObRb, expressed primarily in the hypothalamus, and it is well documented in vivo that the janus-activated kinase (JAK)-STAT pathway is involved in this process. The mechanisms involved in alternative leptin signaling in the hypothalamus are poorly understood, although other important pathways are necessary for the diverse functions ascribed to leptin. The hypothalamus itself is a heterogeneous population of cell types, so it is difficult to study signal transduction in individual neurons from the hypothalamus. To circumvent this, we generated clonal hypothalamic cell models representing many unique hypothalamic cell types, including the NT-expressing cell model N-39. These neurons represent the first central cell models to study endogenous leptin receptor signal transduction and allow analysis of differential leptin-mediated signaling events in divergent cell types. Analysis of leptin receptor activation in peripheral tissues paints a diverse picture of overlapping signal transduction pathways that are likely accountable for cell-specific responses.
Hypothalamic NT neurons are responsive to leptin. Evidence from in vivo studies in rats indicates that NT may modulate the central effects of leptin on feeding behavior. We have reported that leptin directly induces NT gene expression in clonal NT-expressing hypothalamic cell models, N-39 and N-36/1, and that this induction requires STAT3 binding at the level of the NT/N promoter region. We therefore sought to define the leptin-mediated signal transduction events in these cells. Results demonstrate that in the N-39 neurons, leptin significantly induces p38 MAPK activity, causing phosphorylation of its downstream substrate, ATF-1, a CREB-related transcription factor. The increase in pATF-1 activity was completely inhibited by SB203580, indicating that the p38 pathway is an upstream mediator of ATF-1-specific phosphorylation in these cells. In contrast, leptin did not alter the level of pCREB or binding activation, indicating that ATF-1 plays a major role in what appears to be cell-specific leptin signaling.
In N-39 neurons, which endogenously express ObRb, we find induction of ERK1/2 phosphorylation and c-Fos protein expression by leptin. ERK1/2 signaling is necessary for the leptin-mediated induction of NT gene expression in N-39 neurons as demonstrated by inhibitors of MEK1/2, an upstream effector of ERK1/2. Although CREB is known as a PKA effector, it is also a major downstream substrate of ERK1/2 activation. It has been shown that CREB/ATF-1 binding at the CRE activates c-Fos gene expression. There is evidence that activation of ERK1/2 induces c-Fos gene expression in some cell types; this gene regulation can be CREB/ATF-1 dependent or independent. We speculate in our N-39 model that acute ERK1/2 phosphorylation by leptin induces c-Fos gene expression at 30–60 min and is a permissive step in regulating NT transcription by leptin.
Leptin can achieve highly divergent and apparently cell-specific regulation of signal transduction pathways. We have linked leptin-mediated p38 activation to ATF-1 phosphorylation, but it has no effect on CREB. This is the first description of p38/ATF-1 activation by leptin in hypothalamic neurons. We propose that regulation of NT gene expression by leptin involves synergistic interactions between the JAK-STAT, p38, and ERK1/2 signal transduction pathways, resulting in downstream activation of some transcription factors that likely include STAT3, ATF-1, and c-Fos (Fig. 3)
. These proteins bind to the mouse NT/N promoter region at the CRE/activating protein-1 and STAT binding sites to direct transcription of the NT gene. There appears to be differential binding to the CRE/activating protein-1 site depending on the concentration of leptin used. c-Fos binds more prevalently at a higher leptin concentration (10–7 M) and ATF-1 binds predominantly at a more physiological leptin concentration. STAT3 binding to the NT/N promoter was also more prevalent at a more physiological level of leptin (10–11 M). The MEK inhibitor results indicate that the ERK1/2 pathway is predominantly active at the 10–7 M leptin concentration, as blocking this pathway with U0126 at the 10–11 M leptin concentration had no effect on induction of NT mRNA levels. We again propose that leptin may induce differential signaling pathways and effector molecules at physiological vs. supraphysiological (high) leptin levels, perhaps providing a potential mechanism by which leptin resistance develops in the obese. These studies offer further evidence that NT neurons of the hypothalamus may indeed be first-order neurons responsive to leptin, which in turn regulates NT-responsive neuronal cell types. Since NT has been described as a potent anorexigenic compound and decreases feeding, we suggest this may be a mechanism by which NT neurons from the hypothalamus contribute to the regulation of energy homeostasis.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.06-5989fje
This article has been cited by other articles:
|
|
|
|
 
J. Wauman, A.-S. De Smet, D. Catteeuw, D. Belsham, and J. Tavernier
Insulin Receptor Substrate 4 Couples the Leptin Receptor to Multiple Signaling Pathways
Mol. Endocrinol.,
April 1, 2008;
22(4):
965 - 977.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Z. He, J. Jiang, M. Kokkinaki, N. Golestaneh, M.-C. Hofmann, and M. Dym
Gdnf Upregulates c-Fos Transcription via the Ras/Erk1/2 Pathway to Promote Mouse Spermatogonial Stem Cell Proliferation
Stem Cells,
January 1, 2008;
26(1):
266 - 278.
[Abstract]
[Full Text]
[PDF]
|
|
|
Copyright © 2006 by The Federation of American Societies for Experimental Biology.
80% confluence, then serum-starved overnight. After pretreatment with or without 10 µM SB 203580 for 1 h, N-39 neurons were stimulated with leptin (10–11 or 10–7 M, respectively) for 15 min. Cell lysates were examined by Western blot with phosphospecific and total CREB antibodies. A summary of all experiments performed (n
3) is presented in the graph as mean ± SEM; *significantly different (P<0.05) from untreated control. Immunoblot is representative of an experiment performed at least 3 times. B) Cells were plated in 60 mm plates to 
-actin mRNA levels (mean±SE, n=3). *Significantly different (P<0.05) compared with untreated control.
