A R T I C L E
The hypothalamic arcuate nucleus: A key site for mediating
leptin’s effects on glucose homeostasis and locomotor activity
Roberto Coppari,1,6Masumi Ichinose,1,5,6Charlotte E. Lee,1Abigail E. Pullen,1Christopher D. Kenny,1
Robert A. McGovern,1Vinsee Tang,1Shun M. Liu,3Thomas Ludwig,4Streamson C. Chua, Jr.,3,7
Bradford B. Lowell,1,7,* and Joel K. Elmquist1,2,7,*
1Department of Medicine, Division of Endocrinology
2Department of Neurology, Program in Neuroscience, Beth Israel Deaconess Medical Center, Harvard Medical School, 99 Brookline Avenue,
Boston, Massachusetts 02215
3Department of Pediatrics
4Department of Anatomy and Cell Biology, Columbia University, New York, New York 10032
5Department of Anatomy, Shiga University of Medical Science, Seta, Otsu, Shiga 520-2192, Japan
6These authors contributed equally to this work.
7These authors contributed equally to this work.
*Correspondence: firstname.lastname@example.org (B.B.L.); email@example.com (J.K.E.)
Leptin is required for normal energy and glucose homeostasis. The hypothalamic arcuate nucleus (ARH) has been pro-
posed as an important site of leptin action. To assess the physiological significance of leptin signaling in the ARH, we
used mice homozygous for a FLPe-reactivatable, leptin receptor null allele (Leprneo/neomice). Similar to Leprdb/dbmice,
these mice are obese, hyperglycemic, hyperinsulinemic, infertile, and hypoactive. To selectively restore leptin signaling in
the ARH, we generated an adeno-associated virus expressing FLPe-recombinase, which was delivered unilaterally into the
hypothalamus using stereotaxic injections. We found that unilateral restoration of leptin signaling in the ARH of Leprneo/neo
mice leads to a modest decrease in body weight and food intake. In contrast, unilateral reactivation markedly improved
hyperinsulinemia and normalized blood glucose levels and locomotor activity. These data demonstrate that leptin signaling
in the ARH is sufficient for mediating leptin’s effects on glucose homeostasis and locomotor activity.
these, the ARH has been proposed as an important site for
mediating leptin’s effect on energy homeostasis (Cowley et al.,
2003; Schwartz et al., 2003; Zigman and Elmquist, 2003). In-
deed, several reports support this view: Takeda et al. (2002)
demonstrated that icv leptin infusion failed to reduce body
weight in ARH-lesioned Lepob/obmice. Moreover, ARH-specific
LEPR-B gene therapy in rats lacking functional leptin receptor
results in an amelioration of the obese phenotype (Morton et
al., 2003). The ARH contains two populations of first-order, lep-
tin-responsive neurons: The orexigenic NPY/AGRP and the
anorexigenic CART/POMC neurons (Spiegelman and Flier,
2001; Saper et al., 2002). NPY/AGRP neurons are directly in-
hibited by leptin (van den Top et al., 2004), whereas CART/
POMC neurons are directly activated by leptin (Cowley et al.,
2001; Elias et al., 1999). Consistent with this, Lepob/obmice
have increased hypothalamic Agrp and Npy mRNA levels (Mi-
zuno and Mobbs, 1999; Schwartz et al., 1996; Ahima et al.,
1996; Stephens et al., 1995) and reduced Pomc mRNA levels
(Schwartz et al., 1997; Thornton et al., 1997). Further support-
ing the importance of the ARH in controlling leptin actions on
energy homeostasis, we have recently shown that mice lacking
LEPRs only in POMC neurons are mildly obese (Balthasar et
In addition to the well-documented effects on body weight,
leptin signaling is required for normal glucose homeostasis as
demonstrated by the fact that both Lepob/oband Leprdb/db
Leptin is secreted by adipocytes and signals to the brain the
status of the body’s energy content (Spiegelman and Flier,
2001; Friedman, 2004). Mice lacking leptin (Lepob/obmice) or
leptin receptor signaling (Leprdb/dbmice) are obese, diabetic,
infertile, and hypoactive (Chen et al., 1996; Chua et al., 1996;
Lee et al., 1996; Tartaglia et al., 1995; Zhang et al., 1994; Cole-
man, 1978). Recently, it has also been shown that leptin plays
a critical role in neuronal plasticity (Pinto et al., 2004; Bouret et
al., 2004). Substantial evidence suggests that the brain medi-
ates the majority of leptin’s action on energy homeostasis. For
example, deletion of leptin receptors (LEPRs) in neurons in-
duces obesity (Cohen et al., 2001), whereas expression of
LEPRs in neurons of Leprdb/dbmice leads to an amelioration
of their obesity (Kowalski et al., 2001). Moreover, intracerebro-
ventricular (icv) administration of leptin in Lepob/obmice causes
reduction of body weight and food intake (Campfield et al.,
Among the five splice variants described in mice (Lee et al.,
1996), the long form of the leptin receptor (LEPR-B) is required
for normal body weight homeostasis (Chen et al., 1996; Lee et
al., 1996). Within the brain, abundant expression of LEPR-B
has been found in several sites including hypothalamic groups
such as the arcuate (ARH), the ventromedial (VMH), the dor-
somedial (DMH), and the ventral premammillary (PMV) nuclei
(Mercer et al., 1996; Elmquist et al., 1998; Fei et al., 1997;
Schwartz et al., 1996; Thornton et al., 1997). Prominent among
CELL METABOLISM : JANUARY 2005 · VOL. 1 · COPYRIGHT © 2005 ELSEVIER INC. DOI:10.1016/j.cmet.2004.12.00463
A R T I C L E
mice have impaired insulin/glucose homeostasis (Spiegelman
and Flier, 2001; Pelleymounter et al., 1995; Coleman, 1978).
However, it is still unclear whether leptin regulates glucose ho-
meostasis directly or indirectly through its action on body
weight. Several studies support the view that leptin does have
an effect on glucose homeostasis independent of its effect on
body weight regulation. For example, Pelleymounter et al.
(1995) reported that daily leptin administration in Lepob/ob
mice, at doses that did not have an effect on body weight,
normalized serum glucose level. Moreover, Schwartz et al.
(1996) showed that leptin-treated Lepob/obmice had 40%
greater reduction in glucose level compared with pair-fed
Lepob/obcontrol mice. Furthermore, Shimomura et al. (1999)
described that adipose-deficient, leptin-deficient, lipodystro-
phic mice are insulin resistant and that insulin sensitivity can
be restored in these mice by leptin infusion. This effect was
also independent of leptin’s action on body weight. To date, it
is unclear whether leptin’s effects on insulin-target tissues are
mediated indirectly by the brain or directly by LEPRs in these
tissues (Kamohara et al., 1997; Minokoshi et al., 2002).
As previously stated, leptin directly acts on NPY/AGRP and
CART/POMC neurons. Thus, the NPY pathway and the mela-
nocortin pathway might be involved in leptin-mediated control
of glucose homeostasis. Indeed, Lepob/obmice lacking the Npy
gene (Lepob/ob; Npy−/−mice) have almost normal serum glu-
cose levels and 50% reduced insulinemia compared to
Lepob/obmice (Erickson et al., 1996). Leptin activates the mela-
nocortin pathway by stimulating melanocortinergic POMC neu-
rons and by inhibiting AGRP neurons (AGRP is the natural an-
tagonist at the melanocortin receptors) (Cowley et al., 2001;
Elias et al., 1999; Elmquist et al., 1999; Schwartz et al., 1996;
Roseberry et al., 2004). Recently, it has also been shown that
central melanocortin signaling regulates insulin action. For ex-
ample, icv infusion of either the natural agonist (α-melanocyte
stimulating hormone [α-MSH]) or the synthetic antagonist
(SHU9119) of the melanocortin receptors 3 and 4 (MC3R and
MC4R) in rats has opposite effects. α-MSH-treated rats have
enhanced insulin action, whereas SHU9119-treated rats have
diminished insulin action (Obici et al., 2001).
Leptin signaling has also been shown to be a critical regula-
tor of reproductive function. Indeed both Lepob/oband Leprdb/db
mice are infertile (Spiegelman and Flier, 2001; Bates et al.,
2003; Coleman, 1978). The NPY pathway has been proposed
to mediate leptin’s effect on reproductive function. Consistent
with this hypothesis, Lepob/ob; Npy−/−mice have improved fer-
tility (Erickson et al., 1996). Also, in agreement with this, mice
lacking leptin-mediated STAT3 activation, which are obese but
have normal hypothalamic Npy gene expression, are fertile
(Bates et al., 2003).
Finally, leptin exerts also a positive action on locomotor ac-
tivity as suggested by the fact that Lepob/obmice are hypo-
active and that their locomotor activity can be normalized by
leptin treatment (Pelleymounter et al., 1995). To examine whether
leptin signaling only in ARH neurons is sufficient to prevent obe-
sity, diabetes, infertility, and hypoactivity, we re-expressed LEPRs
under the control of the endogenous leptin receptor promoter in
neurons in the ARH of mice otherwise deficient in leptin recep-
Restoring leptin receptor expression
in the arcuate nucleus
In order to selectively express LEPRs in ARH neurons, we em-
ployed Leprneo/neomice (McMinn et al., 2004). These mice are
homozygous for a FLPe-reactivatable, Lepr-null allele (Figure
1B) and, as a result, are similar to Leprdb/dbmice (McMinn et
al., 2004). FLPe-mediated deletion of the FRT-flanked Neo cas-
sette creates a normally functioning Lepr allele. Indeed, mice
homozygous for the FLPe-reactivated, Lepr allele (Leprflox/flox
mice, Figure 1C) had body weights indistinguishable from wild-
type littermates (McMinn et al., 2004; Balthasar et al., 2004).
Site-specific reactivation of the Lepr allele was achieved by
stereotaxic delivery of FLPe-recombinase in Leprneo/neomale
mice. Due to the unavailability of antibodies for FLPe-recombi-
nase and in order to visualize FLPe-expressing cells in stereo-
taxically injected Leprneo/neomice, we engineered an adeno-
associated viral vector that would also express enhanced
green fluorescent protein (eGFP) (Figure 1A). Therefore, we
were able to visualize FLPe-expressing cells by performing
immunohistochemistry for eGFP.
The AAV-FLPe-IRES-eGFP vector was stereotaxically in-
jected into the hypothalamus with coordinates focused on the
ARH of Leprneo/neomice. This procedure inherently resulted in
a high percentage of injection sites that were centered outside
of the ARH as well as several injections that were centered in
the ARH. However, these missed injections served as impor-
tant anatomic controls. Thus, we categorized the injections as
either ARH-misses (control group) or ARH-hits first by neuro-
anatomic inspection. In addition, as described in Experimental
procedures, we also categorized the injections by counting
eGFP-positive cells within the hypothalamus. Briefly, brains
that had >100 eGFP-positive cells in the ARH (25 ?m sections
containing the ARH in a 1:5 series) were scored as ARH-hit.
Those cases with <100 eGFP-positive cells in the ARH were
grouped as ARH-missed. Importantly, we also obtained cases
that had >100 eGFP-positive cells in the ARH plus in other
nuclei which are known to contain Lepr-b mRNA (e.g., the DMH
and the VMH). These cases with ARH rescue plus the addi-
tional sites were not categorized in the ARH-hit group. Figures
2A and 2B show representative photomicrographs of im-
munohistochemistry for eGFP in ARH-missed and ARH-hit
Leprneo/neomice, respectively. The eGFP immunohistochem-
istry was used as an index to categorize the center of the injec-
tion sites and, thus, the site containing the majority of FLPe-
expressing cells. We also determined whether the delivery of
FLPe successfully restored leptin signaling in transduced neu-
rons. To accomplish this, we assessed the rapid phosphoryla-
tion and nuclear translocation of signal transducer and activa-
tor of transcription 3 (STAT3) in response to leptin (Munzberg
et al., 2003; Li and Friedman, 1999; Baumann et al., 1996;
Hosoi et al., 2002). Therefore, 45 min before perfusion, mice
were injected intraperitoneally with leptin, and phospho-STAT3
(P-STAT3) immunohistochemistry was performed. As can be
seen in Figure 2C, ARH-missed cases displayed very modest
leptin-induced P-STAT3 immunoreactivity in the ARH. In con-
trast, prominent P-STAT3 immunoreactivity was characteristi-
cally observed in mice with ARH-specific restoration of leptin
receptor signaling (Figure 2D). The hypothalamic distribution
of leptin-induced P-STAT3-positive neurons in these mice is
CELL METABOLISM : JANUARY 2005
Leptin action in the hypothalamic arcuate nucleus
Figure 1. Schematic representation of adeno-asso-
ciated viral (AAV) vector and modified Lepr alleles
A) AAV-FLPe-IRES-eGFP was obtained by cloning
the FLPe gene into a transfer vector (pAAV-M2-
IGFP) such that FLPe and eGFP are driven by CMV
B) The FRT-Neo-FRT cassette is located upstream
of exon 17 of the Lepr allele. This allele was bred to
homozygosity to generate Leprneo/neomice.
C) The FLPe-mediated removal of the FRT-Neo-FRT
cassette produces the FRT-modified and loxP-
flanked Lepr allele which functions as a wild-type
allele. Leprflox/floxmice are homozygous for this
D) The CRE-deleted Lepr allele lacks exon 17 and
is a null allele. Lepr?/?mice are homozygous for
outlined in Table 1. These data demonstrate that leptin signal-
ing was established in ARH-neurons of Leprneo/neomice.
We also attempted to rule out the possibility that the viral
injections and/or the expression of FLPe and eGFP by neurons
in the ARH per se had effects on body weight and glucose
homeostasis in obese and diabetic mice. We performed the
following experiment. The AAV-FLPe-IRES-eGFP vector was
stereotaxically injected into the hypothalamus with coordinates
focused on the ARH of mice homozygous for a Lepr-null allele
that cannot be reactivated by FLPe (LeprD/Dmice, Figure 2D).
LeprD/Dmice were categorized as ARH-hit and ARH-missed as
described above and in Experimental procedures. Figures 3A
and 3B show representative photomicrographs of immuno-
histochemistry for eGFP in ARH-missed and ARH-hit LeprD/D
mice, respectively. Both, ARH-missed and ARH-hit LeprD/D
mice displayed no leptin-induced P-STAT3 immunoreactivity
(Figures 3C and 3D, respectively). Importantly, ARH-missed
and ARH-hit LeprD/Dmice had indistinguishable body weight
(12-week-old LeprD/Dmice: ARH-missed = 50.02 [g] ± 1.85 [n =
6]; ARH-hit = 51.12 [g] ± 1.6 [n = 3], Figure 4C). These mice
also had indistinguishable serum insulin (12-week-old LeprD/D
mice: ARH-missed = 89.22 ng/ml ± 18.21 [n = 4]; ARH-hit = 80.88
ng/ml ± 27.58 [n = 3], Figure 5B) and glucose levels (12-week-
old LeprD/Dmice: ARH-missed = 492.5 mg/dl ± 34.17 [n = 4];
ARH-hit = 562 mg/dl ± 81.6 [n = 3], Figure 5D). These data
demonstrate that the viral injections and/or the expression of
FLPe and eGFP in neurons in the ARH had no effects on body
weight and glucose homeostasis in obese and diabetic mice.
gated the possibility that re-establishment, unilaterally, of leptin
signaling in ARH neurons would be sufficient to restore normal
energy homeostasis in Leprneo/neomice. As shown in Figure
4A, ARH-hit Leprneo/neomice had significantly reduced body
weight (starting at 7 weeks of age) when compared to ARH-
missed Leprneo/neomice. We found that the difference in body
weight between 12-week-old ARH-hit Leprneo/neomice and
ARH-missed Leprneo/neomice was 5.2 g. This represented ap-
proximately 22% of the total body weight difference between
mice with normal leptin signaling (not surgically treated Lepr+/+
mice) and ARH-missed Leprneo/neomice (Figure 4B). Body
composition analysis revealed that the reduction in body
weight observed in ARH-hit Leprneo/neomice was due to a re-
duction in fat mass (Figure 4D). Since ARH-hit Leprneo/neomice
had improved energy homeostasis, they must have either re-
duced food intake or increased energy expenditure or both.
ARH-hit Leprneo/neomice had reduced cumulative food intake
(Figure 4E). However, energy expenditure was not elevated in
these mice (Figure 4F). These data suggest that leptin signaling
in one side of the ARH is sufficient to mediate w20% of leptin’s
action on body weight homeostasis.
Restoration of leptin receptors in the arcuate nucleus
dramatically improves glucose homeostasis
As noted, leptin signaling is required for normal blood glucose
and insulin levels. In fact, Lepob/obmice and Leprdb/dbmice
develop overt diabetes (Spiegelman and Flier, 2001; Coleman,
1978). We found that unilateral restoration of leptin signaling in
the ARH was sufficient to remarkably improve glucose homeo-
stasis in Leprneo/neomice. Indeed, insulinemia was greatly re-
duced at both 4 and 8 weeks after FLPe-recombinase was de-
livered into the ARH of Leprneo/neomice (Figure 5A). Notably, 4
weeks after surgery, the blood glucose levels in ARH-hit
Leptin action in the arcuate nucleus reduces body weight
Mice lacking leptin signaling have excessive body and fat
mass, increased food intake, and reduced energy expenditure
(Spiegelman and Flier, 2001; Friedman, 2004). Thus, we investi-
CELL METABOLISM : JANUARY 200565
A R T I C L E
Figure 2. Unilateral Lepr gene reactivation in the
ARH of Leprneo/neomice
Photomicrograph of immunohistochemistry for eGFP
in ARH-missed (A; Case 17 in Table 1) and ARH-hit
(B; Case 9 in Table 1) Leprneo/neomice. Photomicro-
graph of immunohistochemistry for leptin-induced
phospho-STAT3 in ARH-missed (C; Case 17) and
ARH-hit (D; Case 9) Leprneo/neomice. Median emi-
nence (ME), third ventricle (3V), hypothalamic arcu-
ate nucleus (ARH). Scale bar = 100 ?m.
Leprneo/neomice were not statistically reduced compared to
that of ARH-missed Leprneo/neomice (although a trend toward
lower glycemia was seen). However, at 8 weeks after surgery,
blood glucose levels were normalized. Indeed, we found that
12-week-old ARH-hit Leprneo/neomice had blood glucose
levels indistinguishable to age-matched Lepr+/+mice (Figure
5C). These data strongly suggest that insulin action was greatly
improved in ARH-hit Leprneo/neomice.
are able to mediate leptin’s effect on locomotor activity, we
recorded ambulatory movements of ARH-hit Leprneo/neoand
ARH-missed Leprneo/neomice. As shown in Figure 6A, ARH-hit
Leprneo/neomice had significantly increased 24 hr locomotor
activity when compared to ARH-missed Leprneo/neomice. No-
tably, the majority of the increase in activity occurred during the
dark cycle (Figures 6B and 6C). Since the ambulatory activity of
ARH-hit Leprneo/neomice was not statistically different to that
of wild-type control mice (Figure 6A), we conclude that
leptin-sensitive, ARH neurons are sufficient for mediating the
majority, if not all of leptin’s action on locomotor activity.
Arcuate nucleus leptin receptors and reproduction
Leptin is also known to affect fertility. In fact, mice lacking lep-
tin signaling are infertile (Spiegelman and Flier, 2001; Coleman,
1978; Bates et al., 2003). Therefore, we tested whether ARH-hit
Leprneo/neomale mice were able to produce offspring by breed-
ing these male mice for one week with adult females. The
one-week breeding period was chosen because in similar
housing conditions, adult wild-type male mice were able to im-
pregnate adult wild-type female mice in this window of time.
All male mice (n = 4) that were bred with female mice for 6–7
days generated offspring. In contrast, both ARH-hit and
ARH-missed Leprneo/neomice were unable to produce off-
spring. Indeed, none of the females were found to be pregnant
after the one-week breeding period. These data indicate that
unilateral leptin signaling in the ARH is insufficient to restore
the capacity of Leprneo/neomale mice to generate offspring.
The incidence of obesity and diabetes continues to rise in in-
dustrialized countries (Flier, 2004; Friedman, 2004; Barsh et al.,
2000). In order to prevent and/or treat these conditions it is
critical to understand the cellular and neuroanatomic pathways
that control energy and glucose homeostasis. The hormone
leptin is required for normal body weight and glucose homeo-
stasis and is key in governing these biological programs (Flier,
2004; Friedman, 2004; Barsh et al., 2000). During the last de-
cade it has become evident that leptin’s primary site of action
is the central nervous system (CNS). However, leptin receptors
are expressed in several CNS sites and relatively little is known
about which neuronal groups mediate each of the specific ac-
tions of leptin. The arcuate nucleus in the hypothalamus has
been proposed as one important site for mediating leptin’s ef-
fect on energy homeostasis (Cowley et al., 2003; Schwartz et
al., 2003; Zigman and Elmquist, 2003). Supporting this view,
we have shown that deletion of LEPRs only in POMC cells
leads to mild obesity (Balthasar et al., 2004). In addition,
Morton et al. (2003) performed ARH-specific Lepr-b gene ther-
Leptin action in the arcuate nucleus
and locomotor activity
Leptin positively regulates locomotor activity as supported by
the fact that Lepob/obmice are hypoactive and their total activ-
ity can be normalized by leptin administration (Pelleymounter
et al., 1995). To date, it is unknown which sites in the brain
mediate this effect of leptin. In order to assess if ARH neurons
CELL METABOLISM : JANUARY 2005
Leptin action in the hypothalamic arcuate nucleus
Table 1. Hypothalamic distribution of leptin-induced P-STAT3 positive neurons in ARH-missed and ARH-hit Leprneo/neomice
LtRt RtRtRt RtRt
P-STAT3-positive neurons were estimated in hypothalamic nuclei using a camera lucida device (all sections; 1:5 series). Arcuate nucleus (ARH); retrochiasmatic area
(RCA); ventromedial nucleus (VMH); dorsomedial nucleus (DMH); lateral hypothalamic area (LH); suprachiasmatic nucleus (Sch). Lt = left side; Rt = right side.
apy in leptin receptor-deficient rats and showed that this ame-
liorated obesity. This study used adenoviral vectors to trans-
genically express LEPR-B under the control of CMV regulatory
elements. In contrast, our approach used the delivery of
FLPe-recombinase in young FLPe-reactivatalbe, Lepr-null mice
(Leprneo/neomice) (McMinn et al., 2004). This approach allowed
us to express endogenous LEPRs at physiological levels only
in neurons that would normally express LEPRs.
Collectively, our data suggest that restoring physiological
leptin signaling in ARH neurons is sufficient to prevent the full
obesity syndrome seen in leptin receptor-deficient mice. How-
ever, the reduction in body weight is relatively modest (w20%)
Figure 3. Delivery of AAV-FLPe-IRES-eGFP in the
ARH of Lepr?/?mice does not lead to Lepr gene re-
Photomicrograph of immunohistochemistry for eGFP
in ARH-missed (A) and ARH-hit (B) Lepr?/?mice.
Photomicrograph of immunohistochemistry for lep-
tin-induced phospho-STAT3 in ARH-missed (C) and
ARH-hit (D) Lepr?/?mice. Median eminence (ME),
third ventricle (3V), hypothalamic arcuate nucleus
(ARH). Scale bar = 100 ?m.
CELL METABOLISM : JANUARY 2005 67
A R T I C L E
Figure 4. Unilateral reactivation of the Lepr gene in the hypothalamic arcuate nucleus of Leprneo/neomice leads to reduced body weight, fat mass, and food intake
A) Body weight curves of ARH-missed (n = 9) and ARH-hit (n = 9) Leprneo/neomice.
B) Body weight of 12-week-old ARH-missed (n = 9), ARH-hit (n = 9) Leprneo/neoand Lepr+/+(n = 10) mice.
C) Body weight of 12-week-old ARH-missed (n = 6), ARH-hit (n = 3) Lepr?/?mice. Please note that the body weights are displayed separately from the mice in (B)
since the mice have dissimilar genetic backgrounds (C57Bl6/J and 129 in the case of Leprneo/neomice and C57Bl6/J, 129 and FVB in the case of Lepr?/?mice).
D) Fat and lean mass in 12-week-old ARH-missed (n = 9), ARH-hit (n = 6) Leprneo/neoand Lepr+/+(n = 4) mice were measured by DEXA.
E) Cumulative food intake in ARH-missed (n = 9) and ARH-hit (n = 9) Leprneo/neomice was collected between ages 5 and 12 weeks.
F) Oxygen consumption was measured with CLAMS in 13-week-old ARH-missed (n = 9) and ARH-hit (n = 5) Leprneo/neomice. *p < 0.05; **p < 0.01; ***p < 0.001 versus
compared to mice with normal leptin signaling. Given the fact
that leptin signaling was restored only in one side of the ARH,
it remains unknown whether bilateral reactivation would lead
to normal body weight homeostasis. However, it is notable that
hypothalamic lesions (including the ARH) in rodents cause sig-
nificant obesity only when performed bilaterally (Elmquist et al.,
1999). The fact that unilateral lesions fail to produce obesity
suggests that unilateral ARH function should be sufficient for
normal body weight regulation.
Interestingly, despite the modest reduction in body weight,
unilateral reactivation markedly improved hyperinsulinemia and
normalized blood glucose levels. Although our study demon-
strates that LEPRs expression by ARH neurons is sufficient to
mediate leptin’s effects on glucose homeostasis, the down-
stream pathways mediating these effects are still unknown. We
propose that melanocortin receptor- and NPY receptor-express-
ing neurons are downstream targets of leptin-responsive ARH
neurons in the leptin-mediated control of glucose homeostasis.
Consistent with this hypothesis, it has been shown that activa-
tion of the central melanocortin pathway leads to increased insu-
lin action (Obici et al., 2001) and decreased insulin levels (Fan et
al., 2000). Since the ARH contains first-order, leptin-responsive
neurons that secrete either the agonist (α-MSH) or the antago-
nist (AGRP) of the melanocortin receptors (Schwartz et al.,
1996, 1997; Cowley et al., 2001; van den Top et al., 2004),
mice with restored leptin signaling in ARH neurons would be
expected to have improved melanocortin signaling and per-
haps insulin action. Moreover, it has been shown that NPY is
required for the development of the full diabetes syndrome in
leptin-deficient mice (Erickson et al., 1996), suggesting that
elevated NPY (as found in Lepob/ob) positively contributes to
their diabetes. Since leptin inhibits NPY-secreting ARH neurons
(van den Top et al., 2004; Roseberry et al., 2004), restoration
of leptin signaling in the ARH is expected to reduce NPY and
therefore to ameliorate glucose homeostasis. Alternatively, it is
possible that the improvement of glucose/insulin homeostasis
is secondary to the reduction in body weight. However, it is
unlikely that this is the sole mechanism underlying the improve-
ment in glucose homeostasis. First, blood glucose levels, while
normalized at 8 weeks after FLPe-recombinase was delivered
into the ARH of Leprneo/neomice, were not normalized at 4
CELL METABOLISM : JANUARY 2005
Leptin action in the hypothalamic arcuate nucleus
Figure 5. Unilateral reactivation of the Lepr gene in
the hypothalamic arcuate nucleus of Leprneo/neo
mice leads to improved glucose homeostasis
(A) Serum insulin and (C) blood glucose levels in fed
ARH-missed (n = 9), ARH-hit (n = 6) Leprneo/neoand
Lepr+/+(n = 3–9) mice. Note that the AAV injections
do not affect glucose homeostasis. Also note that
panels (B) and (D) are separated from (A) and (C)
since the mice have dissimilar genetic backgrounds
(as noted in the legend for Figure 4). (B) Serum
insulin and (D) serum glucose levels in fed ARH-
missed (n = 4), ARH-hit (n = 3) Lepr?/?mice. *p <
0.05; **p < 0.01; ***p < 0.001 versus ARH-missed
weeks after surgery. Since body weight was similarly reduced
at both the 4 and 8 week points, it is unlikely to be the cause
of the normal glycemia at the 8 week point. Moreover, food
restriction experiments in obese mice, which led to a similar
reduction of body weight as that seen in ARH-hit Leprneo/neo
mice, were not able to normalize blood glucose levels
(Schwartz et al., 1996; Yamamoto et al., 2000). In addition, as
detailed below, reactivation of LEPRs in both the VMH and the
lateral hypothalamic area (LH) of Leprneo/neomice resulted in a
reduction in body weight similar to ARH-hit Leprneo/neomice.
However, this body weight reduction was not associated with
improved glucose homeostasis. Thus, we conclude that the
improvements in glucose homeostasis are likely to be indepen-
dent of a body weight reduction. This suggests that leptin sig-
naling in ARH neurons exerts direct control over insulin/glucose
homeostasis. This concept would be consistent with other
claims that hypothalamic neurons regulate glucose homeosta-
sis (Obici et al., 2001; Fan et al., 2000).
Leptin also exerts a positive action on locomotor activity as
suggested by the fact that Lepob/obmice are hypoactive and
that their locomotor activity can be normalized by leptin treat-
ment (Pelleymounter et al., 1995). However, the neuronal
groups that mediate this effect of leptin are unknown. Our
study shows that restoring leptin signaling in ARH neurons is
sufficient to normalize locomotor activity. Therefore, this finding
establishes the ARH as a key site for mediating leptin’s effect
on locomotor activity. We suggest the existence of a novel,
yet undefined, neuronal pathway connecting ARH neurons to
cortical and/or subcortical areas regulating voluntary locomo-
tion. Additional studies will be required to reveal the impor-
Figure 6. Unilateral reactivation of the Lepr gene in the hypothalamic arcuate nucleus of Leprneo/neomice leads to increased locomotor activity
(A) 24 hr (B) nocturnal and (C) diurnal locomotor activity were measured with CLAMS in 13-week-old ARH-missed (n = 9), ARH-hit (n = 5) Leprneo/neoand Lepr+/+(n =
5) mice. *p < 0.05; **p < 0.01; ***p < 0.001 versus ARH-missed Leprneo/neomice.
CELL METABOLISM : JANUARY 2005 69
A R T I C L E
293A cells and purified by heparin column. The eluted virus was dialyzed
against PBS and the titer was assessed by dot blot hybridization. All these
procedures were performed by the Harvard Gene Therapy Initiative core
Four-week-old Leprneo/neoand LeprD/Dmale mice were stereotaxically in-
jected with AAV-FLPe-IRES-eGFP into the ARH with a glass micropipette
and air pressure injector system (Chamberlin et al., 1998).
tance of this hypothesized neurocircuit mediating leptin’s effect
on ambulatory movements.
Further supporting our conclusions that leptin signaling
specifically in the ARH is a major feeding-independent regula-
tor of glucose homeostasis and locomotor activity were physi-
ological data on four Leprneo/neomice that had brains with
>100 eGFP-positive cells in both the LH and the VMH and
<100 eGFP-positive cells in any other nucleus known to con-
tain Lepr-b mRNA. Like ARH-hit Leprneo/neomice, these LH +
VMH-hit Leprneo/neomice had w20% reduction in body weight
(12-week-old Leprneo/neomice: ARH-missed = 47.44 [g] ± 0.56
[n = 9]; ARH-hit = **42.23 [g] ± 1.6 [n = 9]; LH + VMH-hit =
*43.74 [g] ± 2.09 [n = 4]; *p < 0.05, **p < 0.01 versus
ARH-missed). Importantly, despite the reduced body weight,
blood glucose levels and locomotor activity were not normal-
ized in LH + VMH-hit Leprneo/neomice (blood glucose levels in
12-week-old Leprneo/neomice: ARH-missed = 300 [mg/dl] ± 48
; LH + VMH-hit = 231 [mg/dl] ± 76 [n = 4]; locomotor activity
in 12-week-old Leprneo/neomice: ARH-missed = 4447 [counts/
day] ± 631 [n=9]; LH+VMH-hit = 4646 [counts/day] ± 923 [n =
4]). Further analysis, which is in progress in our laboratories, is
needed to definitely assess the physiological importance of
leptin signaling specifically in the LH or the VMH. However,
we predict that leptin’s effects on glucose homeostasis and
locomotor activity are not mediated by LEPRs on neurons con-
tained within these two hypothalamic areas.
In summary, restoration of leptin signaling in ARH neurons
leads to improved, but not normal energy homeostasis. Thus,
other leptin-responsive neuronal groups are likely also to be
important in mediating leptin’s effect on food intake and body
weight. In contrast, LEPRs expression by ARH neurons is suffi-
cient to mediate the majority of leptin’s effects on glucose ho-
meostasis and locomotor activity. Delivery of FLPe in a
neuron-specific fashion by the generation of neuron-specific
FLPe-transgenic Leprneo/neomice (for example, Npy-FLPe,
Pomc-FLPe, or Agrp-FLPe transgenic mice) will be critical to
reveal the relative contribution of leptin signaling in specific
populations on neurons in the ARH in mediating the varied ef-
fects of leptin.
Body and blood composition
Tail vein blood was collected at noon ± 2 hr from fed 8- and 12-week-old
mice. Blood was assayed for glucose level (Fisher Scientific, Morrison
Plains, New Jersey) and successively serum was collected by centrifugation
and assayed for insulin levels using commercially available kits (Crystal
Chem. Inc., Downers Grove, Illinois). Serum was assayed for glucose levels
using an enzymatic glucose oxidase method (Thermo Electron, Victoria,
Australia). After blood was collected, mice were ketamine anesthetized for
dual-energy X-ray absorptiometry (MEC Lunar Corp., Minster, Ohio)
Oxygen consumption, locomotor activity, and fertility test
Metabolic rate and physical activity were measured in 13-week-old mice
using a comprehensive lab animal monitoring system (CLAMS, Columbia
Instruments, Columbus, Ohio). Mice were acclimated in the monitoring
chambers for 2 days then data were collected for 3 days. Data analysis was
performed only in mice that did not lose weight during the experiment. After
CLAMS analysis, every mouse was housed with two 7- to 12-week-old FVB
female mice for a period of 6–7 days. Female mice were monitored for the
following 4 weeks and the presence of the offspring was counted as event
of pregnancy. Female control mice were tested for their fertility by breeding
with wild-type male mice. All female mice had events of pregnancy when
bred with wild-type male mice. Food and water were provided ad libitum
during the entire period.
eGFP and P-STAT3 immunohistochemistry
Fed 14-week-old male mice were injected intraperitoneally with 100 ?g of
recombinant mouse leptin (A.F. Parlow, National Hormone and Peptide Pro-
gram) and perfused with 10% formalin 45 min later. Either eGFP or P-STAT3
immunohistochemistry was performed on microtome cut 25 ?m brain sec-
tions as described earlier (Liu et al., 2003; Elias et al., 1999; Munzberg et
Categorization of ARH-hit and ARH-missed Leprneo/neo
Brain sections stained against eGFP by immunohistochemistry were used
for this categorization. The borders of the nuclei containing eGFP-positive
cells were drawn on a paper sheet using a camera lucida device using
darkfield optics and an atlas of the mouse brain (Franklin and Paxinos,
1997). The eGFP-positive cells were plotted and counted such that the
ARH-hit and the ARH-missed were grouped as follows. Brains that had
>100 eGFP-positive cells in the arcuate nucleus of the hypothalamus (all
sections in a 1:5 series) and <100 eGFP-positive cells in any other nucleus
known to contain Lepr-b mRNA (all sections in a 1:5 series) were scored as
ARH-hit. Brains that had <100 eGFP-positive cells in any nucleus known to
contain Lepr-b mRNA (all sections in a 1:5 series) were scored as
ARH-missed. These criteria were established a priori. Also, an individual
blinded to the neuroanatomic categorizations and physiological responses
of each case performed this analysis.
Care of all mice was within the Institutional Animal Care and Use Committee
(IACUC) guidelines, and all the procedures were approved by the Beth Israel
Deaconess Medical Center IACUC. Mice were housed individually at 22°C–
24°C using a 14 hr light/10 hr dark cycle with chow food (Tekland F6 Rodent
Diet8664, Harlan Tekland, Madison, Wisconsisn) and water provided ad li-
Leprneo/+mice were provided by Dr. S. Chua, Jr. (McMinn et al., 2004) . The
genetic background of Leprneo/+mice is an admixture of C57Bl6/J and 129.
Leprneo/neoand Lepr+/+male mice were obtained by mating Leprneo/+mice
with Leprneo/+mice. Mice were genotyped by PCR with primers (1 and 2)
across the loxP site: 1, 5#-AAT GAA AAA GTT GTT TTG GGA CGA-3# and
2, 5#-CAG GCT TGA GAA CAT GAA CAC AAC AAC-3#. LeprD/Dmice are
homozygous for a null LEPR allele lacking exon 17 and were generated as
described previously (Balthasar et al., 2004). The genetic background of
LeprD/Dmice is an admixture of C57Bl6/J, 129, and FVB.
Data sets were analyzed for statistical significance using PRISM 3.0
(GraphPad, San Diego, California) for a two-tailed unpaired Student’s t test.
Statistical comparisons shown in Figures 4B, 4D, 5A, 5C, 6A, 6B, and 6C
were made by using one-way ANOVA (Turkey’s post test). All parameters
are expressed as mean ± SEM.
AAV-FLPe-IRES-eGFP generation and microinjection
The FLPe gene was cloned into the transfer vector pAAV-M2-IGFP such
that the AAV vector plasmid called here pAAV-FLPe-M2-IGFP was gener-
ated. The virus was generated by tripartite transfection (AAV-rep/cap ex-
pression plasmid, adenovirus miniplasmid, and pAAV-FLPe-M2-IGFP) into
The authors would like to thank Drs. Mineko Fujimiya, Clifford Saper and
Christian Bjorbaek for helpful advice. This work was supported by grants
CELL METABOLISM : JANUARY 2005
Leptin action in the hypothalamic arcuate nucleus
Elmquist, J.K., Elias, C.F., and Saper, C.B. (1999). From lesions to leptin:
hypothalamic control of food intake and body weight. Neuron 22, 221–232.
from the NIH (PO1 DK56116 to B.B.L. and J.K.E., DK57621 and DK26687
to S.C.C., Jr., MH61583 and DK53301 to J.K.E.) and by Takeda Chemical
Industries, Ltd., Japan. We thank NIDDK’s National Hormone & Peptide Pro-
gram and A.F. Parlow for providing recombinant mouse leptin.
Erickson, J.C., Hollopeter, G., and Palmiter, R.D. (1996). Attenuation of the
obesity syndrome of ob/ob mice by the loss of neuropeptide Y. Science
Fan, W., Dinulescu, D.M., Butler, A.A., Zhou, J., Marks, D.L., and Cone, R.D.
(2000). The central melanocortin system can directly regulate serum insulin
levels. Endocrinology 141, 3072–3079.
Received: November 10, 2004
Revised: December 22, 2004
Accepted: December 23, 2004
Published: January 19, 2005
Fei, H., Okano, H.J., Li, C., Lee, G.H., Zhao, C., Darnell, R., and Friedman,
J.M. (1997). Anatomic localization of alternatively spliced leptin receptors
(Ob-R) in mouse brain and other tissues. Proc. Natl. Acad. Sci. USA 94,
Ahima, R.S., Prabakaran, D., Mantzoros, C., Qu, D., Lowell, B., Mara-
tos-Flier, E., and Flier, J.S. (1996). Role of leptin in the neuroendocrine re-
sponse to fasting. Nature 382, 250–252.
Flier, J.S. (2004). Obesity wars: molecular progress confronts an expanding
epidemic. Cell 116, 337–350.
Franklin, K.B.J., and Paxinos, G. (1997). The Mouse Brain in Sterotaxic Co-
ordinates (San Diego: Academic Press).
Balthasar, N., Coppari, R., McMinn, J., Liu, S.M., Lee, C.E., Tang, V., Kenny,
C.D., McGovern, R.A., Chua, S.C., Jr., Elmquist, J.K., and Lowell, B.B.
(2004). Leptin receptor signaling in POMC neurons is required for normal
body weight homeostasis. Neuron 42, 983–991.
Friedman, J.M. (2004). Modern science versus the stigma of obesity. Nat.
Med. 10, 563–569.
Hosoi, T., Kawagishi, T., Okuma, Y., Tanaka, J., and Nomura, Y. (2002). Brain
stem is a direct target for leptin's action in the central nervous system.
Endocrinology 143, 3498–3504.
Barsh, G.S., Farooqi, I.S., and O'Rahilly, S. (2000). Genetics of body-weight
regulation. Nature 404, 644–651.
Bates, S.H., Stearns, W.H., Dundon, T.A., Schubert, M., Tso, A.W., Wang, Y.,
Banks, A.S., Lavery, H.J., Haq, A.K., Maratos-Flier, E., et al. (2003). STAT3
signalling is required for leptin regulation of energy balance but not repro-
duction. Nature 421, 856–859.
Kamohara, S., Burcelin, R., Halaas, J.L., Friedman, J.M., and Charron, M.J.
(1997). Acute stimulation of glucose metabolism in mice by leptin treatment.
Nature 389, 374–377.
Kowalski, T.J., Liu, S.M., Leibel, R.L., and Chua, S.C., Jr. (2001). Transgenic
complementation of leptin-receptor deficiency. I. Rescue of the obesity/
diabetes phenotype of LEPR-null mice expressing a LEPR-B transgene. Di-
abetes 50, 425–435.
Baumann, H., Morella, K.K., White, D.W., Dembski, M., Bailon, P.S., Kim,
H., Lai, C.F., and Tartaglia, L.A. (1996). The full-length leptin receptor has
signaling capabilities of interleukin 6-type cytokine receptors. Proc. Natl.
Acad. Sci. USA 93, 8374–8378.
Bouret, S.G., Draper, S.J., and Simerly, R.B. (2004). Trophic action of leptin
on hypothalamic neurons that regulate feeding. Science 304, 108–110.
Lee, G.H., Proenca, R., Montez, J.M., Carroll, K.M., Darvishzadeh, J.G.,
Lee, J.I., and Friedman, J.M. (1996). Abnormal splicing of the leptin receptor
in diabetic mice. Nature 379, 632–635.
Campfield, L.A., Smith, F.J., Guisez, Y., Devos, R., and Burn, P. (1995). Re-
combinant mouse OB protein: evidence for a peripheral signal linking adi-
posity and central neural networks. Science 269, 546–549.
Li, C., and Friedman, J.M. (1999). Leptin receptor activation of SH2 domain
containing protein tyrosine phosphatase 2 modulates Ob receptor signal
transduction. Proc. Natl. Acad. Sci. USA 96, 9677–9682.
Chamberlin, N.L., Du, B., de Lacalle, S., and Saper, C.B. (1998). Recombi-
nant adeno-associated virus vector: use for transgene expression and ante-
rograde tract tracing in the CNS. Brain Res. 793, 169–175.
Liu, H., Kishi, T., Roseberry, A.G., Cai, X., Lee, C.E., Montez, J.M., Friedman,
J.M., and Elmquist, J.K. (2003). Transgenic mice expressing green fluores-
cent protein under the control of the melanocortin-4 receptor promoter. J.
Neurosci. 23, 7143–7154.
Chen, H., Charlat, O., Tartaglia, L.A., Woolf, E.A., Weng, X., Ellis, S.J., Lakey,
N.D., Culpepper, J., Moore, K.J., Breitbart, R.E., et al. (1996). Evidence that
the diabetes gene encodes the leptin receptor: identification of a mutation
in the leptin receptor gene in db/db mice. Cell 84, 491–495.
McMinn, J.E., Liu, S.M., Dragatsis, I., Dietrich, P., Ludwig, T., Eiden, S., and
Chua, S.C., Jr. (2004). An allelic series for the leptin receptor gene gener-
ated by CRE and FLP recombinase. Mamm. Genome 15, 677–685.
Chua, S.C., Jr., Chung, W.K., Wu-Peng, X.S., Zhang, Y., Liu, S.M., Tartaglia,
L., and Leibel, R.L. (1996). Phenotypes of mouse diabetes and rat fatty due
to mutations in the OB (leptin) receptor. Science 271, 994–996.
Mercer, J.G., Hoggard, N., Williams, L.M., Lawrence, C.B., Hannah, L.T.,
and Trayhurn, P. (1996). Localization of leptin receptor mRNA and the long
form splice variant (Ob-Rb) in mouse hypothalamus and adjacent brain re-
gions by in situ hybridization. FEBS Lett. 387, 113–116.
Cohen, P., Zhao, C., Cai, X., Montez, J.M., Rohani, S.C., Feinstein, P., Mom-
baerts, P., and Friedman, J.M. (2001). Selective deletion of leptin receptor
in neurons leads to obesity. J. Clin. Invest. 108, 1113–1121.
Minokoshi, Y., Kim, Y.B., Peroni, O.D., Fryer, L.G., Muller, C., Carling, D.,
and Kahn, B.B. (2002). Leptin stimulates fatty-acid oxidation by activating
AMP-activated protein kinase. Nature 415, 339–343.
Coleman, D.L. (1978). Obese and diabetes: two mutant genes causing
diabetes-obesity syndromes in mice. Diabetologia 14, 141–148.
Mizuno, T.M., and Mobbs, C.V. (1999). Hypothalamic agouti-related protein
messenger ribonucleic acid is inhibited by leptin and stimulated by fasting.
Endocrinology 140, 814–817.
Cowley, M.A., Smart, J.L., Rubinstein, M., Cerdan, M.G., Diano, S., Horvath,
T.L., Cone, R.D., and Low, M.J. (2001). Leptin activates anorexigenic POMC
neurons through a neural network in the arcuate nucleus. Nature 411,
Morton, G.J., Niswender, K.D., Rhodes, C.J., Myers, M.G., Jr., Blevins, J.E.,
Baskin, D.G., and Schwartz, M.W. (2003). Arcuate nucleus-specific leptin
receptor gene therapy attenuates the obesity phenotype of Koletsky
(fa(k)/fa(k)) rats. Endocrinology 144, 2016–2024.
Cowley, M.A., Cone, R.D., Enriori, P., Louiselle, I., Williams, S.M., and Evans,
A.E. (2003). Electrophysiological actions of peripheral hormones on melano-
cortin neurons. Ann. N Y Acad. Sci. 994, 175–186.
Munzberg, H., Huo, L., Nillni, E.A., Hollenberg, A.N., and Bjorbaek, C.
(2003). Role of signal transducer and activator of transcription 3 in regula-
tion of hypothalamic proopiomelanocortin gene expression by leptin. En-
docrinology 144, 2121–2131.
Elias, C.F., Aschkenasi, C., Lee, C., Kelly, J., Ahima, R.S., Bjorbaek, C., Flier,
J.S., Saper, C.B., and Elmquist, J.K. (1999). Leptin differentially regulates
NPY and POMC neurons projecting to the lateral hypothalamic area. Neu-
ron 23, 775–786.
Elmquist, J.K., Bjorbaek, C., Ahima, R.S., Flier, J.S., and Saper, C.B. (1998).
Distributions of leptin receptor mRNA isoforms in the rat brain. J. Comp.
Neurol. 395, 535–547.
Obici, S., Feng, Z., Tan, J., Liu, L., Karkanias, G., and Rossetti, L. (2001).
Central melanocortin receptors regulate insulin action. J. Clin. Invest. 108,
CELL METABOLISM : JANUARY 200571
A R T I C L E
Pelleymounter, M.A., Cullen, M.J., Baker, M.B., Hecht, R., Winters, D.,
Boone, T., and Collins, F. (1995). Effects of the obese gene product on body
weight regulation in ob/ob mice. Science 269, 540–543.
Spiegelman, B.M., and Flier, J.S. (2001). Obesity and the regulation of en-
ergy balance. Cell 104, 531–543.
Stephens, T.W., Basinski, M., Bristow, P.K., Bue-Valleskey, J.M., Burgett,
S.G., Craft, L., Hale, J., Hoffmann, J., Hsiung, H.M., Kriauciunas, A., et al.
(1995). The role of neuropeptide Y in the antiobesity action of the obese
gene product. Nature 377, 530–532.
Pinto, S., Roseberry, A.G., Liu, H., Diano, S., Shanabrough, M., Cai, X.,
Friedman, J.M., and Horvath, T.L. (2004). Rapid rewiring of arcuate nucleus
feeding circuits by leptin. Science 304, 110–115.
Takeda, S., Elefteriou, F., Levasseur, R., Liu, X., Zhao, L., Parker, K.L., Arm-
strong, D., Ducy, P., and Karsenty, G. (2002). Leptin regulates bone forma-
tion via the sympathetic nervous system. Cell 111, 305–317.
Roseberry, A.G., Liu, H., Jackson, A.C., Cai, X., and Friedman, J.M. (2004).
Neuropeptide Y-mediated inhibition of proopiomelanocortin neurons in the
arcuate nucleus shows enhanced desensitization in ob/ob mice. Neuron 41,
Tartaglia, L.A., Dembski, M., Weng, X., Deng, N., Culpepper, J., Devos, R.,
Richards, G.J., Campfield, L.A., Clark, F.T., Deeds, J., et al. (1995). Identifi-
cation and expression cloning of a leptin receptor, OB-R. Cell 83, 1263–
Saper, C.B., Chou, T.C., and Elmquist, J.K. (2002). The need to feed: ho-
meostatic and hedonic control of eating. Neuron 36, 199–211.
Schwartz, M.W., Baskin, D.G., Bukowski, T.R., Kuijper, J.L., Foster, D., Las-
ser, G., Prunkard, D.E., Porte, D., Jr., Woods, S.C., Seeley, R.J., and Weigle,
D.S. (1996). Specificity of leptin action on elevated blood glucose levels and
hypothalamic neuropeptide Y gene expression in ob/ob mice. Diabetes 45,
Thornton, J.E., Cheung, C.C., Clifton, D.K., and Steiner, R.A. (1997). Regula-
tion of hypothalamic proopiomelanocortin mRNA by leptin in ob/ob mice.
Endocrinology 138, 5063–5066.
van den Top, M., Lee, K., Whyment, A.D., Blanks, A.M., and Spanswick, D.
(2004). Orexigen-sensitive NPY/AgRP pacemaker neurons in the hypothala-
mic arcuate nucleus. Nat. Neurosci. 7, 493–494.
Schwartz, M.W., Seeley, R.J., Woods, S.C., Weigle, D.S., Campfield, L.A.,
Burn, P., and Baskin, D.G. (1997). Leptin increases hypothalamic
pro-opiomelanocortin mRNA expression in the rostral arcuate nucleus. Dia-
betes 46, 2119–2123.
Yamamoto, Y., Ueta, Y., Serino, R., Nomura, M., Shibuya, I., and Yamashita,
H. (2000). Effects of food restriction on the hypothalamic prepro-orexin
gene expression in genetically obese mice. Brain Res. Bull. 51, 515–521.
Schwartz, M.W., Woods, S.C., Seeley, R.J., Barsh, G.S., Baskin, D.G., and
Leibel, R.L. (2003). Is the energy homeostasis system inherently biased to-
ward weight gain? Diabetes 52, 232–238.
Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold, L., and Friedman,
J.M. (1994). Positional cloning of the mouse obese gene and its human
homologue. Nature 372, 425–432.
Shimomura, I., Hammer, R.E., Ikemoto, S., Brown, M.S., and Goldstein, J.L.
(1999). Leptin reverses insulin resistance and diabetes mellitus in mice with
congenital lipodystrophy. Nature 401, 73–76.
Zigman, J.M., and Elmquist, J.K. (2003). Minireview: From anorexia to obe-
sity–the yin and yang of body weight control. Endocrinology 144, 3749–
CELL METABOLISM : JANUARY 2005