Content uploaded by Haifei Shi
Author content
All content in this area was uploaded by Haifei Shi
Content may be subject to copyright.
Brain–adipose tissue cross talk
Timothy J. Bartness
*
, C. Kay Song, Haifei Shi, Robert R. Bowers and Michelle T. Foster
Department of Biology, Neurobiology & Behavior Program, Center for Behavioral Neuroscience,
Georgia State University, Atlanta, GA 30302–4010, USA
While investigating the reversible seasonal obesity of Siberian hamsters, direct sympathetic
nervous system (SNS) postganglionic innervation of white adipose tissue (WAT) has been
demonstrated using anterograde and retrograde tract tracers. The primary function of this
innervation is lipid mobilization. The brain SNS outflow to WAT has been defined using the
pseudorabies virus (PRV), a retrograde transneuronal tract tracer. These PRV-labelled SNS
outflow neurons are extensively co-localized with melanocortin-4 receptor mRNA, which,
combined with functional data, suggests their involvement in lipolysis. The SNS innervation of
WAT also regulates fat cell number, as noradrenaline inhibits and WAT denervation stimulates
fat cell proliferation in vitro and in vivo respectively. The sensory innervation of WAT has
been demonstrated by retrograde tract tracing, electrophysiological recording and labelling of
the sensory-associated neuropeptide calcitonin gene-related peptide in WAT. Local injections
of the sensory nerve neurotoxin capsaicin into WAT selectively destroy this innervation. Just as
surgical removal of WAT pads triggers compensatory increases in lipid accretion by non-
excised WAT depots, capsaicin-induced sensory denervation triggers increases in lipid
accretion of non-capsaicin-injected WAT depots, suggesting that these nerves convey
information about body fat levels to the brain. Finally, parasympathetic nervous system
innervation of WAT has been suggested, but the recent finding of no WAT immunoreactivity
for the possible parasympathetic marker vesicular acetylcholine transporter (VAChT) argues
against this claim. Collectively, these data suggest several roles for efferent and afferent neural
innervation of WAT in body fat regulation.
Sympathetic nervous system: Sensory nerves: Parasympathetic nervous system: Pseudorabies virus: Lipolysis
A critical ability of all animals is to meet energy demands
quickly with utilizable metabolic fuels. In mammals
glycogen serves as a rapidly availa ble, but extremely
limited, carbohydrate energy source. Energy is predomi-
nately stored as lipid in white adipose tissue (WAT) in the
form of triacylglycerol (Newsholme & Leech, 1983).
Access to this energy in fat can be mediated by several
means, principally by the sympathetic nervous system
(SNS) innervation of WAT. The present review starts by
discussing the SNS innervation of WAT and then
considers its sensory innervation and possible parasym pa-
thetic nervous system (PSNS) innervation. As the title of the
review suggests, the cross talk occurring between the brain
and WAT, which use these innervations as conduits, will
be highlighted. However, the provocative and interesting
cross talk between WAT and the brain that occurs via
humoral factors will not be discussed.
Adrenal medullary catecholamines, once thought to be
the principal stimulators of lipid mobilization, do not
play a major role in lipolysis
Traditionally, adrenal medullary secret ion of catechola-
mines, primarily adrenaline, is thought to be the under-
lying means by which lipid mobilization is stimulated.
Thus, signals indicating the need for lipid-derived energy
activate brain sites that, in turn, increase the SNS outflow
to the intermed ial lateral horn of the spinal cord impinging
on the sympathetic preganglionic neurons that reside there.
These neurons project separately to adrenaline- and
Abbreviations: CGRP, calcitonin gene-related peptide; FCN, fat cell number; MC4-R, melanocortin-4 receptor subtype; MEL, melatonin; PSNS,
parasympathetic nervous system; PRV, pseudorabies virus; SD, short ‘winter-like’ days; SNS, sympathetic nervous system; WAT, white adipose tissue.
*Corresponding author: Dr Timothy J. Bartness, fax +1 404 651 2509, email bartness@gsu.edu
Proceedings of the Nutrition Society (2005), 64, 53–64 DOI:10.1079/PNS2004409
g
The Authors 2005
noradrenaline-synthesizing chromaffin cells in the adrenal
gland (Hillarp & Hokfelt, 1953; Edwards et al. 1996), and
when their neuronal activity increases, there is an increase
in the release of adrenaline, and to a lesser extent nor-
adrenaline, into the circulation. Circulating adrenaline, in
turn, stimulates membrane-bound WAT adrenoceptors (for
review, see Lafontan & Berlan, 1993). Stimulation of these
b-adrenoceptors activates a cascade of lipolytic intracellular
steps resulting in increases in lipolysis (the breakdown of
triacylglycerols into glycerol and NEFA ), thereby supplying
the needed lipid-derived fuels for oxidation (Newsholme &
Leech, 1983). Despite mounting functional and histological
evidence to the contrary (for reviews, see Bartness &
Bamshad, 1998; Bartness et al. 2001), the dogma that
adrenaline is the principal stimulator of lipolysis persists.
This view probably arises because of the robust lipolytic
activity of adrenaline when added to isolated adipocytes
in vitro (for example, see White & Engel, 1958; Rizack,
1961). Evidence contrary to the primacy of adrenaline in
triggering lipolysis in vitro is the inability of adrenal deme-
dullation to block lipid mobilization triggered by several
physiological conditions in vivo, e.g. glucoprivation
(Nishizawa & Bray, 1978), electrical stimulation of the
medial hypothalamus (Takahashi & Shimazu, 1981) and
short photoperiod exposure in Siber ian hamsters (Phodo-
pus sungorus; Demas & Bartness, 2001b). As described
later (p. 54), it is now more commonly believed that the
sympathetic innervation of WAT, through the release of its
primary postganglionic neurotransmitter noradrenaline, is
the principal ini tiator of lipolysis (for review, see Bart ness
& Bamshad, 1998; Bartness et al. 2001; Dodt et al. 2003).
Naturally-occurring seasonal decreases in Siberian
hamster body fat levels have served as a useful model to
study the role of the sympathetic nervous system
innervation of white adipose tissue
Although the innervation of WAT by the SNS has received
support both neuroanatom ically and functionally for more
than 100 years (Dogiel, 1898; for review, see Bartness &
Bamshad, 1998), it has only been in the last decade that
irrefutable neuroanatomical and functional data have been
forthcoming. The scientific community turned a blind eye
toward this notion, probably because of the vigorous
lipolytic response elicite d by adrenaline when added to iso-
lated adipocytes in vitro, as discussed earlier, and because
initially only SNS innervation of blood vessels was seen in
WAT (Wirsen, 1964; Daniel & Derry, 1969; Ballantyne &
Raftery, 1974). This latter conclusion was undoubtedly
drawn because white adipocytes of ad libitum-fed animals
are filled with a single large lipid droplet that pushes the
cell membranes into tight contact with neighbouring
adipocytes, obfuscating the parenchymal space. After
fasting, however, the cells shrink, revealing catecholami-
nergic innervation of both the vasculature and white
adipocytes using histofluorescence (Diculescu & Stoica,
1970; Ballard et al. 1974; Slavin & Ballard, 1978;
Rebuffe-Scrive, 1991).
A contribution to these more compelling data has
been made through the present authors’ search for the
mechanisms underlying the reversibility of seasonal
(photoperiod-induced) obesity exhibited by Siberian ham-
sters (for reviews, see Bartness & Wade, 1985; Bartness
et al. 2002). As man y of the neuroanatomical and
functional examples of WAT innervation contained herein
are based on these findings with the Siberian hamster, this
model, which drew the authors into the ‘brain–adipose
tissue cross talk’, will be briefly described.
When housed in long ‘summer-like’ days Siberian
hamsters show a marked seasonal obesity (40–50% body
fat) that gradually and naturally develops across the first
1–2 months of life (for review, see Bartness & Wade, 1985).
Most remarkably, this obesity is completely reversible,
such that when they are exposed to short ‘winter-like’ days
(SD) they voluntarily rapidly lose body fat, with no
decrease in food intake during the period of most rapid
body mass decrease (for example, see Wade & Bartness,
1984; Bartness et al. 1989). The day length (photoperiod)
provides a reliable ‘noise-free’ environmental cue (Turek
& Campbell, 1979) that is responsible for the seasonal
changes in body fat and other annual responses (e.g. pelage
colour change, reproductive status, thermogenic capabil-
ities) in Siberian hamsters, as well as in other species
showing photoperiod-driven seasonal cycles (for review,
see Underwood & Goldman, 1987; Bartness & Goldman,
1989). This photic information is received by the retina
and transmitted via a multi-synaptic pathway to the pineal
gland, where the pinealocytes synthesize and secrete
melatonin (MEL) only at night. Thus, changes in the
duration of the nocturnal secretion of pineal MEL faith-
fully code the night length, thereby signalling the pro-
gression of the seasons (photoperiods). Pinealectomized
hamsters given exogenous peripheral MEL infusions that
mimic the natural peak duration of nocturnal MEL se-
cretion trigger the photoperiodic responses, including the
changes in body fat (Bartness & Goldman, 1988; Song &
Bartness, 1996, 1998).
Initially a hormonal intermediary for the MEL-induced
changes in body fat was sought; however, none of the
hormones that change seasonally and that also directly or
indirectly affect body fat (e.g. thyroxine, insulin, gonadal
steroids) could account for the SD-indu ced decreases in
lipid stores (for review, see Bartness & Fine, 1999;
Bartness et al. 2002). Thus, the SNS was considered as a
possible mediator for the photoperiod–MEL-induced
changes in body fat.
There is neuroanatomical and functional evidence for
the sympathetic nervous system innervation of white
adipose tissue
The inability of adrenal demedullation to block lipid
mobilization under several physiological conditions (see p.
54) led to the examination of the SNS innervation of
WAT as a possible mediator of the SD-induced decrease in
body fat shown by Siberian hamsters (Demas & Bartness,
2001b). The first direct neuroanatomical evidence for the
sympathetic postganglionic neurons was provided through
injections of a retrograde fluorescent tract tracer Fluoro-
Gold into inguinal WAT or epididymal WAT. To demon-
strate these connections bi-directionally, an anterograde
fluorescent tract tracer DiI was also injected into the
54 T. J. Bartness et al.
sympathetic chain (Youngstrom & Bartness, 1995). Fluor-
escently-labelled cell bodies were found in the sympathetic
chain as a result of retrograde labelling of the postgan-
glionic nerves and rings of fluorescence around individual
adipocytes as a result of anterograde labelling (Young-
strom & Bartness, 1995). Moreover, the distribution of
postganglionic neurons innervating these two fat pads was
found to be quite distinct, providing a likely neuroanato-
mical basis for the SD-induced differential lipid mobiliza-
tion across fat pads (for example, see Bartness et al. 1989;
Bartness, 1995, 1996), as well as for fat pad-specific
differences in lipolysis shown in other rodent species and
in human subjects. These data, along with an anecdotal
report of sympathetic nerves innervating both white
adipocytes and blood vessels, as revealed by histofluores-
cence combined with confocal microscopy (Rebuffe-
Scrive, 1991), support the view that white adipocytes are
sympathetically innervated. This phenomenon is not
species-specific because, in addition to the studies of the
SNS innervation of Siberian hamster WAT, parallel experi-
ments were conducted in laboratory rats yielding similar
findings (Youngstrom & Bartness, 1995).
These earlier studies gave no indi cation of the origins of
the SNS outflow from brain to WAT. Connections from
specific brain nuclei to WAT were, up to that time,
inferred from lesion or stimulation studies in which the
target site was destroyed or chemically or electrically
stimulated and resulting changes in WAT growth or
physiology were ascribed to disruption or activation of
the presumed SNS outflow connections to WAT (for
review, see Bartness & Bamshad, 1998). Another credible
alternative explanation for the effects of brain lesions or
stimulation on WAT would be that these manipulations
affect peripheral endocrine organ function (e.g. PSNS
innervation of the pancreas affecting insulin or glucagon
secretion) resulting in changes in hormonal secretion that,
in turn, affect WAT growth and/or physiology. In order to
determine the credibility of the post hoc explanations
ascribing the effects of brain lesion or stimulation to alter-
ations in the SNS outflow to WAT, a transneuronal tract
tracer, the Bartha’s K strain of the pseudo rabies virus
(PRV), was used. PRV had been developed by other
researchers to trace entire circuits within the same animal
and had been used successfully to show the SNS outflow to
the adrenal gland (Strack et al. 1989) among other
peripheral tissues. Neurotropic viruses, such as the PRV,
bind specifically to presynaptic neural membranes, fuse
with their axonal membrane and deliver the uncoated
capsids inside the axon. The capsids are transported by
retrograde motors (probably dynein) to the cell body,
where they replicate (Enquist & Card, 2003). The virus
only exits the infected neurons via their dendrites, only
infecting neurons that are synaptically connected to these
PRV-laden cells. This process results in the retrograde
labelling of functional chains of hierarchically-connected
neurons within an animal (Card et al. 1990; Strack &
Loewy, 1990). Initially PRV was injected into the inguinal
WAT and epididymal WAT of Siberian hamsters (Bamshad
et al. 1998). Retrogradely-infected cells were identified
using immunocytochemistry throughout the neural axis, in-
cluding the spinal cord (intermediolateral cell group,
central autonomic nucleus), the brainstem (nucleus of the
solitary tract, A5 regions and the C1/rostroventrolateral,
rostroventromedial and caudal raphe nuclei/areas), mid-
brain (periaqueductal gray) and fore brain (hypothalamic:
arcuate nucleus, dorsal, lateral, paraventricular, suprachias-
matic, nuclei and medial preoptic area; non-hypothalamic:
zona incerta, medial amygdala, septum and bed nucleus of
the stria terminalis; Bamshad et al. 1998). Many of the
virus-labelled brain sites comprising the sympathetic out-
flow to WAT had been correctly deduced as components
of this circuit using lesion or stimulation approaches (for
review, see Bartness & Bamshad, 1998), except for one
glaring omissi on, the virtual lack of PRV-labelled neurons
in the ventromedial hypothalamic nucleus, an area impli-
cated in dozens of lesion or stimulation studies (for review,
see Bartness & Bamshad, 1998). This misplaced focus
on the ventromedial hypothalamic nucleus probably oc-
curred because manipulations of the ventromedial hypo-
thalamic nucleus secondarily affect paraventricular nuclei
descending pathways that pass adjacent to the ventro-
medial hypothalamic nucleus on their way to their brainstem
and spinal cord destinations (for example, see Gold, 1973;
Gold et al. 1977; Luiten et al. 1985).
Regulators of sympathetic outflow
A multitude of virally-labelled structures across the
neuroaxis appears following the PRV injections into
WAT (Bamshad et al. 1998, 1999; Shi & Bartness, 2001;
Song & Bartness, 2001) and, moreover, within each of
these structures there are scores of PRV-infected neurons.
It seems unlikely that all these neurons in all these
structures are activated under all conditions of lipid
mobilization (e.g. with fasting, cold exposure or exercise).
Instead, it is hypothesized that subsets of these structures
and/or subpopulations of neurons within each structure are
activated by different lipolytic stimuli. This hypothesis
cannot be tested using double labelling for c-fos (the early
immediate gene product that serves as a marker of cell
activation; Hoffman et al. 1993) in these PRV-labelled
circuits because the virus itself induces c-fos during early
neuronal infection (Ozaki et al. 1996; Weiss & Chowdh-
ury, 1998). A different approach was therefore used in an
attempt to determine whi ch of the SNS outflow neurons to
WAT are involved in the SD-induced increased lipid
mobilization. To label these neurons, in situ hybridization
for MEL receptor mRNA combined with PRV labelling of
the SNS outflow to WAT was used (Song & Bartness, 2001).
As MEL does not directly aff ect lipolysis of isolated WAT
cells in vitro (Ng & Wong, 1986), it was reasoned that
there should be MEL receptors located on SNS outflow
neurons to WAT of Siberian hamsters. First the SNS
outflow neurons to WAT were labelled using PRV injected
into inguinal WAT, combined with in situ hybridization
for the MEL
1a
receptor, the subtype through which the
photoperiod causes seasonal responses in this and other
species (Song & Bartness, 2001). Double-labelled neurons
were found in several brain regions, including the supra-
chiasmatic nuclei (Fig. 1), an area previously shown to be
critical for reception of the photoperiod-encoded MEL
Biology of obesity 55
signals that trigger SD responses (Bartness et al. 1991;
Song & Ba rtness, 1998). Thus, the following scenario of
SD-induced lipid mobilization has emerged. The increases
in the duration of nocturnal MEL secretion resulting from
the increases in night length cause increases in the
stimulation of MEL
1a
receptors in some of the neurons
comprising the sympathetic outflow to WAT (e.g. supra-
chiasmatic nuclei). This, in turn, causes increases in the
sympathetic drive to WAT (seen as an increase in nor-
adrenaline turnover; Youngstrom & Bartness, 1995) there-
by increasing lipolysis. Increased lipolysis, in turn, leads to
decreases in WAT pad mass, reflected as decreased fat cell
size (Bartness, 1996; Mauer & Bartness, 1996). To the
authors’ knowledge, this condition is the only one in which
a stimulus that increases lipolysis can be traced from the
environment to the adipocyte. Although the relevance for
human obesity of photoperiod–MEL-induced changes in
body fat is unknown, this model of naturally-occurring
changes in body fat has proven useful in understanding the
SNS neuroanatomy underlying lipid mobilization.
Using this powerful methodology for co-labelling central
WAT SNS outflow neurons with expression of mRNA for
their receptors, the possibility that melanocortin-4 receptor
subtype (MC4-R) gene expression is co-localized with the
brain sympathetic circuits innervating WAT has recently
been tested. Although the melanocortins have been shown
to markedly affect food intake and thermogenesis (for
review, see Butler & Cone, 2002), their involvement in
SNS-mediated lipolysis has been implied, but not directly
tested. Chronic central administration of a synthetic MC4-
R agonist melanotan-II (Haskell-Luevano et al. 1994) de-
creases WAT pad mass by approximately 50%, even when
using pair-fed controls to account for its ability to decrease
food intake (Raposinho et al. 2003). These results indi-
cate that the melanotan-II-triggered decreases in body fat
cannot be accounted for simply by its ability to decrease
food intake. Furthermore, centrally-applied melanotan-II
decreases the RQ, suggesting that lipid-derived fuels are
being oxidized (Hwa et al. 2001). These and other data
prompted the authors to test whether MC4-R mRNA is co-
localized with PRV-labelled SNS outflow neurons to WAT
in Siberian hamsters (CK Song, D Richard and T Bartness,
unpublished results). Extensive co-localization of MC4-R
mRNA with PRV-labelled SNS outflow neurons was found
across the neural axis. All animals were found to have
large numbers of PRV + MC4-R neurons in all PRV-
labelled areas, including the paraventricular nuclei, preop tic
area, bed nucleus of the stria terminalis and amygdala in
the forebrain, periaqueductal gray in the midbrain and the
nucleus of the solitary tract, lateral paragigantocellular
Sympathetic
chain
White adipose
tissue
SCN
Label SNS
outflow to IWAT
PRV
S
S
C
C
N
N
O
O
C
C
3
3
V
V
MEL
1a
receptors are co-localized
on SNS outflow neurons to WAT
Label MEL
1a
receptor mRNA
using
in situ
hybridization
(a)
(b)
Fig. 1. Schematic representation of the double labelling of white adipose tissue (WAT)
sympathetic nervous system (SNS) outflow neurons that also express mRNA for the
functional melatonin receptor (MEL
1a
) in Siberian hamsters (Phodopus sungorus) using
the pseudorabies virus (PRV) and in situ hybridization respectively. The diagram is a
coronal section of the brain at the level of the ventral hypothalamus showing double
labelling of sympathetic outflow neurons to WAT with melanocortin-4 receptor (MC4-R)
gene expression in the suprachiasmatic nuclei (SCN). (a) Brown staining indicates PRV
labelling and the dark blue staining indicates gene expression for the MC4-R. (b) Enlarged
photomicrograph shows one PRV-labelled neuron, as indicated by the brown staining,
containing MC4-R mRNA, as indicated by the dark blue staining. 3V, third ventricle, OC,
optic chiasm. (Adapted from Song & Bartness, 2001.)
56 T. J. Bartness et al.
nucleus, lateral reticular area, rostroventrolateral medulla
and anterior giga ntocellular nucleus in the brainstem, to
name only a few of the more predominant co-localizations.
This co-localization is the most extensive that has been
seen in the work to date (Shi & Bartness, 2001; Song &
Bartness, 2001) or in studies in the literature (for brief
review, see Discussion in Shi & Bartness, 2001). These
data suggest that MC4-R may play a prominent role in the
modulation of SNS outflow neurons to WAT either through
stimulation by the endogenous melanocortin agonist a-
melanocyte-stimulating hormone and/or through inhibition
by the naturally-occurring melanocortin-3 receptor and
MC4-R antagonist agouti-related protein (Ollmann et al.
1997). Collectively, studies such as this one, and the co-
localization of MEL
1a
receptor mRNA with PRV-labelled
WAT SNS outflow neurons (Song & Bartness, 2001),
provide maps to guide future experiments for site-specific
microinjections or implants designed to turn the WAT SNS
outflow ‘on’ or ‘off’.
The sympathetic nervous system innervation of white
adipose tissue has at least three functions: lipolysis, the
regulation of fat cell number and control of some white
adipose tissue-secreted proteins
There are three recognized functions of the SNS innerva-
tion of WAT: (1) the principal initiator of lipolysis; (2)
control of fat cell number; (3) control of some WAT-
secreted proteins. The literature relating to the role of the
SNS in lipolysis has recently been reviewed (Bartness &
Bamshad, 1998; Bartness et al. 2001) and, therefore, will
only be considered briefly. Functional studies have been
conducted that take advantage of the unilateral innervation
of pairs of WAT pads. One of a pair of WAT pads can be
denervated with its contralateral counterpart serving as a
within-animal neurally-intact control, thereby keeping cir-
culating factors, age, nutritional status and the behavioural
activity the same between fat pads. Surgically-denervated
WAT show s a greatly diminished lipid mobil ization com-
pared with its neurally-intact controls across a variety of
lipolytic stimuli, e.g. fasting (Clement, 1950; Cantu &
Goodman, 1967; Bray & Nishizawa, 1978) and oestradiol
treatment of ovariectomized animals (Lazzarini & Wade,
1991). Although local surgical denervation affords anatom-
ical specificity compared with global sympathectomy
using guanethidine (Powley et al. 1983) or 6-hydroxy-
dopamine (Robidoux et al. 1995), it is not neuroanatomi-
cally selective because sympathetic and sensory nerves
cannot be distinguished visually and therefore all types of
nerves are severed. Indeed, surgical denervation decreases
the immunoreactivity of tyrosine hydroxylase and calcito-
nin gene-related peptide (CGRP) -, indi cating reduced
sympathetic and sensory innervations respectively (Shi
et al. 2005). A more selective approach than surgical
denervation is chemical SNS denervation of WAT, which
have been successfully accomplished using locally-injected
guanethidine (Demas & Bartness, 2001a,b) and, more
recently with greater reliability, usin g locally-injected
6-hydroxy-dopamine (R Bower s, CK Song, H Shi and T
Bartness, unpu blished results). Guanethidine-induced local
SNS denervation of WAT severely blunts the SD-induced
increase in lipid mobilization (Demas & Bartness, 2001b),
as does surgical denervation (Youngstrom & Bartness,
1998), although only a complete blockade is achieved when
adrenal demedullation is also added (Demas & Ba rtness,
2001b).
Recently, cross talk between white adipocytes and symp-
athetic neurons that influence lipolysis has been demon-
strated cleverly and simply in vitro by co-culturing 3T3-L1
cells and rat superior cervical ganglia postganglionic
sympathetic neurons (Turtzo et al. 2001). Characteristic
morphology associated with each cell type, as well as cell
type-specific markers, occurs in this situation, attesting to
the authenticity of each cell type (Turtzo et al. 2001).
Sympathetic neurons co-cultured with these adipocytes
markedly inhibit b-adrenoceptor-stimulated lipolysis and
leptin secretion (Turtzo et al. 2001). The effect on lipolysis
is likely to be a result of increases in the release of
neuropeptide Y by the co-cultured sympathetic neurons
(Turtzo et al. 2001); neuropeptide Y is known to inhibit
lipolysis in vitro (Castan et al. 1994) by stimulating
membrane-bound adipocyte peptide YY receptors (i.e.
neuropeptide Y receptors; Castan et al. 1993). Th e inhi-
bition of WAT leptin release also may be initiated by
stimulation of these rece ptors.
Sympathetic nervous system drive to white adipose
tissue is not necessarily uniform across fat pads
How the SNS outflow is directed across sympathetic
targets in general (for review, see Morrison, 2001; Sved
et al. 2001), and across individual WAT pads specifically,
is not well understood. It seems clear that the traditi onal
theory proposed by Cannon (1939) of an ‘all or nothing’
activation of the SNS is inadequate to account for differ-
ential sympathetic drives acro ss peripheral tissues. Thus,
recent measures of SNS drive (noradrenaline turnover and
electrophysiological activity of sympathetic nerves), as well
as viral and non-viral tract tracing experiments, suggest
that Cannon’s (1939) hypothesis may represent the excep-
tion rather than the rule for activation of the SNS (for
reviews, see Morrison, 2001; Sved et al. 2001). For
example, the SNS control of blood flow to various vascular
beds can occur independently of one another and of other
organs (e.g. kidney and brown adipose tissue; Vague et al.
1980). In addition, in some conditions ther e are coordi-
nated increases in sympathetic drives to tw o or more SNS
target tissues, whereas in other conditions there are simul-
taneous increases to one of these targets and decreases
to another. For example, with cold exposure norad renaline
turnover in brown adipose tissue (Young et al. 1982;
Garofalo et al. 1996) and WAT (Garofalo et al. 1996)
increases, but with fasting noradrenaline turnover in brown
adipose tissue decreases and in WAT it increases
(Migliorini et al. 1997). How the SNS drive to these two
tissues types is regulated is a biological mystery. An
additional puzzle is how the sympathetic drive across
WAT pads (i.e. among inguinal WAT, epididymal WAT,
retroperitoneal WAT etc.) is regulated. For example,
noradrenaline turnover differs markedly across individual
WAT pads of SD-exposed Siberian hamsters (Youngstrom
& Bartness, 1995). Another example of differential
Biology of obesity 57
sympathetic drive across WAT pads comes from a recent
preliminary study (H Shi and T Bartness, unpublished
results) in which acute glucoprivation was generated via
injection of the glucose-utilization blocker 2-deoxy-
D-
glucose. In this case, 2-deoxy-
D-glucose was found to
markedly increase noradrenaline turnover in inguinal, retro-
peritoneal and dorsosubcutaneous WAT similarly, but does
not alter epididymal WAT noradrenaline turnover (Fig. 2).
These data suggest that acute glucoprivation causes differ-
ential sympathetic drive across WAT pads and further illus-
trates the phenomenon of the separate control of sympathetic
drive across SNS target tissues. From a clinical perspective
it is of paramount importance to determine the mechan-
isms underlying the differential mobilization of lipid from
WAT, because the distribution of WAT is critical in rela-
tion to whethe r the secondary deleterious health conse-
quences of obesity are manifested (Vague et al. 1980;
Gasteyger & Tremblay, 2002). Moreover, even with modest
decreases in visceral fat these secondary health conse-
quences markedly improve (Pasanisi et al. 2001; Janssen
et al. 2002), suggesting that an ability to selectively
mobilize lipid from internal WAT depots would be of
considerable clinical importance.
Sympathetic nervous system innervation of white
adipose tissue also regulates fat cell proliferation
In addition to lipolysis the SNS innervation of WAT also
influences WAT cellularity. It has long been recognized
that WAT hypercellularity is a hallmark sign of obesity in
both man and other animals (for reviews, see Kirtland &
Gurr, 1979; Faust, 1984; Hausman et al. 2001). In ad-
dition, obesity typically is associated with decreases in
SNS activity (for reviews, see Dulloo & Miller, 1987;
Bray, 1990, 1991). It now appears that these two charac-
teristics of obesity are related to one another; indeed, it
may be hypothesized that decreases in SNS activity trigger
hypercellularity. It appears that a high SNS drive to WAT
inhibits, and a low SNS drive disinhibits (stimulates), fat
cell proliferation, which was first demonstrated in vitro;
noradrenaline added to white adipocyte precursor cells was
found to inhibit their normal proliferation (Jones et al.
1992). This effect is blocked by pretreatment wi th the
general b-adrenoceptor antagonist propranolol (Jones et al.
1992). Conversely, in Siberian hamsters (Youngstrom &
Bartness, 1998), and in laboratory rats (Cousin et al.
1993), it has been demonstrated that surgical denervation
of WAT produces pronounced (approximately 2-fold)
increases in fat cell number (FCN) with little change in
fat cell size in vivo. In addition, it has recently been
reported that the magnitude of the denervation-induced
increase in FCN varies between fat pads, with marked
increases occurring in inguinal WAT but no change in
epididymal WAT (Shi et al. 2004). This differential
increase in FCN by surgical denervation may reflect the
natural propensity for these pads to grow by hyperplasia
(inguinal WAT) v. hypertrophy (epididymal WAT;
DiGirolamo et al. 1998). Recent preliminary data suggest
that these increases in FCN probably repr esent real
increases in fat cell proliferation, rather than filling of
existing pre-adipocytes. Specifically, surgical denervation
increases the number of bromodeoxyuridine-labelled cells
(i.e. dividing cells) that also are immunoreactive for AD3,
a white adipocyte-specific membrane protein (Wright &
Hausman, 1990; Kras et al. 1999), thus showing that
they are proliferating white adipocytes (M Foster and
T Bartness, unpubl ished results). Since surgical denervation
severs all nerve types, it is possible that the lack of sensory
innervation triggers increases in FCN. It has recently been
found that WAT sensory dener vation, accomplished by
local injections of the sensory nerve neurotoxin capsaicin
(Jansco et al. 1980; Ainsworth et al. 1981), and verified by
decreased immunoreactivity for the sensory nerve-asso-
ciated neuropeptide CGRP (Skofitsch & Jacobowitz,
1985), does not increas e FCN (Shi et al. 2004). Thus,
decreases in the SNS drive to WAT appear to be a critical
stimulus for increasing fat cell proliferation or FCN.
What is the mechanism underlying the denervation-
induced increase in fat cell number?
If it were not for increases in FCN, then there would be an
upper limit to the extent of adiposity reached by human
subjects and other animals; however, hypercellularity is a
classic sign of obesity (Faust, 1984). Despite the pivotal
role played by fat cell proliferation in the obese state, little
is known about the mechanisms underlyi ng the process (for
review, see Hausman et al. 2001), especially compared
with the knowledge of factors that contribute to the process
of fat cell differentiation. One obvious potential mechan-
ism by which the SNS modulates FCN or proliferation is
via the stimulation of b-adrenoceptors on adipocyte pre-
cursor cells. Evidence that noradrenaline prevents normal
proliferation through b-adrenoceptors in vitro is the ability
of the general b-adrenoceptor blocker propranolol to
NETO (K/mg protein)
0
1000
2000
3000
4000
EWAT IWAT RWAT DWAT
*
*
*
Fig. 2. Noradrenaline turnover (NETO) is differentially increased
across white adipose tissue pads after glucoprivation induced by
peripheral injections of 2-deoxy-D-glucose (2DG; &) suggesting fat
pad-specific control of sympathetic drive in Siberian hamsters
(Phodopus sungorus). (%), Vehicle (saline; 9 g sodium chloride/l);
IWAT, inguinal white adipose tissue; RWAT, retroperitoneal white
adipose tissue; EWAT, epididymal white adipose tissue; DWAT,
dorsosubcutaneous white adipose tissue. Values are means with
their standard errors represented by vertical bars. Mean values
were significantly different from those for the vehicle: * P < 0
.
05.
(H Shi and T Bartness, unpublished results.)
58 T. J. Bartness et al.
disinhibit the noradrenaline-induced inhibition (Jones et al.
1992), as discussed earlier. Another adrenoceptor subtype,
the a
2
-adrenoceptor, has also been implicated in adipocyte
proliferation (Bouloumie et al. 1994; Valet et al. 1998).
Specifically, increases in adipocyte a
2
-adrenoceptor num-
ber precede the marked increase in FCN after surgical
denervation of WAT (Cousin et al. 1993). The exact chain
of events linking the increase in a
2
-adrenoceptor number
and stim ulation of these receptors with the increase in fat
cell proliferation is not known, but one possibility involves a
compensatory increase in adrenal medullary catecholamine
secretion that occurs after sympathetic nerve denervation
(Takahashi et al. 1993). The steps beyond the stimulation
of a-adrenoceptors resu lting in this proliferation are not
well understood, but locally released lysophosphatidic acid
(Valet et al. 1998), a glycerophospholipid (Newsholme &
Leech, 1983), may be involved. Agonists of the a
2
-adreno-
ceptor trigger a rapid and prolonged releas e of lysopho-
sphatidic in isolated WAT cells (Valet et al. 1998) and,
moreover, lysophosphatidic acid added to preadipo se cell
lines triggers fat cell proliferation (Valet et al. 1998). Final-
ly, the newly-discovered white adipocyte paracrine factor
autotoxin (a type II ecto-nucleotide pyrophosphatase
phosphodiesterase) may in turn stimulate prolifera tion via
the release of lysophosphatidic acid (Ferry et al. 2003).
Clearly, a deeper understanding of the mechani sms involved
in fat cell proliferation in general, and the modulation of
fat cell proliferation by the SNS specifically, requires
further investigation.
Sympathetic nervous system outflow from brain can
control the release of white adipose tissue-secreted
peptides
Leptin is a cytokine that is synthesized and released
primarily by white adipocytes (Maffei et al. 1995), and its
discovery (Zhang et al. 1994) led to the notion that leptin
informs the brain of body fat levels. This view, now con-
sidered dogmatic, is largely based on the frequent positive
correlation between circulating leptin concentrations and
the extent of adiposity (e.g. man (Considine et al. 1996;
Dua et al. 1996; Ostlund et al. 1996) and laboratory mice
(Frederich et al. 1995)); but leptin is also involved in other
functions (e.g. reproduction, immunology; stress; for review,
see Harris, 2000). There is an ever increasing number of
exceptions to the positive correlation between body fat
levels and circulating leptin concentration, suggesting that
leptin might not be viewed as a perfect signal of adiposity.
The SNS outflow from the brain to WAT has been
proposed as a principal controller of the secretion of leptin
(Trayhurn et al. 1998). This view stems largely from the
findings that conditions promoting increased SNS drive to
WAT, such as cold exposure (Trayhurn et al. 1995) and
fasting (Hardie et al. 1996), or direct stimulation of WAT
b-adrenoceptors by receptor agonists (Mantzoros et al.
1996; Trayhurn et al. 1996) inhibit leptin gene expression
and/or secretion. Conversely, disruption of the SNS drive
to WAT via the catecholaminergic neurotoxin 6-hydroxy-
dopamine or via a-methyl-p-tyrosine, a blocker of cate-
cholamine (noradrenaline) synthesis, increases circulating
leptin concentrations (Rayner et al. 1998; Sivi tz et al.
1999). Coll ectively, these data support a primary role for
the SNS innervation of WAT in leptin synthesis or release
(increases in sympathetic drive decrease, and decreases in
sympathetic drive incr ease, the synthesis or release of lepti n).
Does white adipose tissue have parasympathetic
nervous system innervation?
Recently, it has been reported that WAT has PSNS
innervation (Kreier et al. 2002). This view is based largely
on the presence of PRV-infected neurons in traditionally-
accepted origins of PSNS premotor neurons, such as the
dorsal vagal complex of the brainstem, after injections
of the virus into WAT (Kreier et al. 2002). Such PRV-
labelled neurons in this ‘PSNS’ area after injections into
WAT have been reported (Bamshad et al. 1998), but they
were attributed rogue SNS outflow neurons because the
notion of wholly SNS or PSNS areas or nuclei in the brain
has not withstood the test of time (for example, see Kalia
et al. 1984; Gwyn et al. 1985). In addition, it is contended
that WAT PSNS surgical denervation can be done selec-
tively at the level of the WAT pad (Kreier et al. 2002).
After this denervation, and in combination with a hyper-
insulinaemic euglycaemic clamp, insulin-mediated glucose
and NEFA uptake is reduced by approximately 30 %, with
hormone-sensitive lipase activity (involved in the hydro-
lysis of triacylglycerol) increased by approximately 50%
(Kreier et al. 2002). Reductions in catabolic responses as a
result of the presumed PSNS denervation of WAT suggest
that the function of this parasympathetic innervation is to
oppose the SNS catabolic actions in the tissue (Kreier et al.
2002), in much the same manner that these two innerva-
tions oppose one another in function in other tissues (e.g.
heart). There is no corroborating neurochemical evidence
to show that the phenotype of these neurons includes
acetylcholine, the predominant PSNS postganglionic
neurotransmitter, nor biochemical evidence for the acetyl-
cholinesterase, an enzyme important in the degradation of
acetylcholine (Ballantyne, 1968). Thus, the presence in
WAT of vesicular acetylcholine transporter, a marker of
PSNS innervation (for example, see Schafer et al. 1998),
has recently been investigated (A Giradano, K Song,
T Bartness and S Cinti, unpublis hed results). No vesicular
acetylcholine transpor ter immunoreactivity has been
found refuting possible PSNS innervation, as has been
previously suggested (Kreier et al. 2002; for comparison,
see Ballantyne, 1968; Bartness, 2002). Finally, preliminary
studies have investigated the hypothesis that if first the
SNS innervation of WAT is selectively eliminated and
then the PRV injected, PSNS innervation should be
preserved and infections of the PSNS outflow to WAT
should be unabated. Thus, WAT was locally sympatheti-
cally denervated via injections of the catecholaminergic
neurotoxin 6-hydroxy-dopamine, followed a few days
later by injections of the PRV. In 6-hydroxy-dopamine-
injected animals no PRV-infected cells were found any-
where in the brain, whereas the normal labelling of SNS
outflow neurons by the virus was seen in hamsters injected
with the 6-hydroxy-dopamine vehicle (A Giradano,
K Song, T Bartness and S Cinti, unpublished results).
Thus, these and other data discussed earlier raise some
Biology of obesity 59
doubt as to the presence of PSNS innervation of WAT.
Such innervation would be intriguing, however, and would
afford WAT the neural control possessed by most other
organs.
White adipose tissue also has sensory innervation that
probably interacts with its sympathet ic nervous system
innervation
Sensory-motor innervation of tissues is the rule, not the
exception; so it should not be surprising that there is
sensory innervation of WAT (for reviews, see Bartness &
Bamshad, 1998; Bartness et al. 2001). Sensory innervation
of WAT was initially suggested by the identification o f
substance P in WAT (Fredholm, 1985). Substance P and
CGRP are contained within, and released from, sensory
neurons and thus are considered markers of sensory inner-
vation (Hill et al. 1996). This initial biochemical analysis
has been confirmed more recently using immun ohisto-
chemistry, showing that WAT contains CGRP- and sub-
stance P-immunoreactivity (Giordano et al. 1996). Direct
evidence of the sensory innervation of WAT was first
shown neuroanatomically when crystals of the fluorescent
anterograde tract tracer True Blue were implanted in rat
WAT, resulting in labelled bipolar sensory neurons in the
dorsal root ganglia (Fishman & Dark, 1987). The functions
of this sensory innervation of WAT are not well under-
stood. One function involves the ability of leptin to
increase peripheral SNS activity, as measured electrophy-
siologically; i.e. injections of leptin into one epididymal
WAT pad increase sensory nerve firing rates from that
epididymal WAT pad, while simultaneously increasing the
sympathetic nerve activity of the contralateral epididymal
WAT pad (Niijima, 1998). These data suggest that sensory
nerves may possess leptin receptors, and that their
activation could send afferent signals from WAT to the
CNS via sensory nerves to trigger increases in the SNS
outflow to WAT, thereby increasing lipolysis.
Another possible function of the WAT sensory innerva-
tion is its involvement in a negative feedback system that
regulates lipid mobilization or accumulation through modi-
fication of the SNS drive to WAT. This view (Bartness &
Bamshad, 1998) suggests interplay between sympathetic
and sensory nerves to and from WAT respectively, and is
based on evidence for such interactions in other tissues.
For example, global sympathectomy or surgical denerva-
tion markedly increases CGRP immunoreactivity in the
mesenteric and other vascular beds, as well as in most
organs (Mione et al. 1992; Aberdeen et al . 1990). Con-
versely, a reciprocal response occurs after global sensory
denervation, resulting in increases in sympathetic activity
measured electrophysiologically (Ralevic et al. 1995).
negative
negative
SNS
outflow
Sensory
inflow
NEFA?
Glycerol?
Leptin?
WAT
Sympathetic
chain
IML
NE
Sensory
inflow
SNS
outflow
Spinal
cord
DRG
Brain
Fig. 3. Schematic diagram of hypothesized neural interplay between the sympathetic
nervous system (SNS) outflow from brain to white adipose tissue (WAT). Sensory input
(
) detects WAT substance indicative of the amount of lipid and/or its turnover,
perhaps NEFA, glycerol or leptin via dorsal root ganglion (DRG) bipolar sensory
neurons. Sensory information goes to spinal cord and then on to the brain to interact
with the SNS outflow and/or a short-feedback loop may go to the intermedial lateral
(IML) horn of the spinal cord; in either or both cases the sensory innervation performs
as a negative feedback to the SNS innervation of WAT. The SNS outflow from brain to
WAT (
) promotes lipolysis and inhibits fat cell proliferation via its principal
postganglionic neurotransmitter, noradrenaline (NE). The sensory nerves may also or
instead provide the brain with feedback about the lipid stores in WAT.
60 T. J. Bartness et al.
These and other studies (for review, see Rubino et al.
1997) suggest cross talk between the sympathetic and
sensory innervations of tissues and, based on these data, it
is speculated that the sensory nerves innervating WAT
may participate in a feedback loop to regulate the level of
its sympathetic drive, thereby regulating lipolysis (Bartness
& Bamshad, 1998).
Conclusions, speculations and future directions
WAT clearly has SNS and sensory innervation, with some
data suggesting PSNS innervation (Kreier et al. 2002). As
for the purported PSNS innervation of WAT, tract tracing
showing innervation from dorsal vagal complex neurons
to WAT that also express a PSNS neurochemical marker
(e.g. acetylcholine, NO, vasoactive intestinal peptide) are
needed, as well as functional studies such as electrical or
chemical stimulation of these neurons to elicit responses
from WAT that are the opposite of, or oppose, those of
SNS activation.
Finally, the depth of the understanding of the functi on of
the sensory afferent nerves emanating from WAT is
certainly in its infancy. As noted earlier, it is believed that
one of these mechanisms is to inform the brain of WAT
pad lipid levels as local selective sensory denervation of
WAT using capsaicin (H Shi and T Bartness, unpublished
results) results in the reparation of the lipid deficit by lipid
accretion in the other fat pads similar to that if the
capsaicin-injected fat pads had been physically removed.
Furthermore, anterograde trans-synaptic viral tract tracing
is required to determine which areas of the brain receive
this sensory innervation.
Collectively, the innervation of WAT by several types
of neurons cannot be challenged, nor can its importance be
underestimated. Additional roles for these innervations,
especially for the control of the synthesis and release of
WAT factors are extremely likely (e.g. leptin, adiponectin,
TNF-a (Orban et al. 1999), which are important in lipo-
lysis and fat cell proliferation. Understanding the roles of
the sensory innervation of WAT and its interactions with
the SNS innervation should also increase with a more
detailed neuroanatomy of the sensory innervation. The
differential activation of SNS outflow circuits to WAT and
other energy-related sympathetic targets such as brown
adipose tissue, as well as the differential SNS drives across
WAT pads, should also prove enlightening. A schematic
diagram of the cross talk between the SNS and sensory
innervations of WAT is depicted in Fig. 3.
Acknowledgements
This research was supported by National Institute of
Health R01-DK35254 to T.J.B. The authors thank Dr Ruth
Harris for her critical reading of this man uscript.
References
Aberdeen J, Corr L, Milner P, Lincoln J & Burnstock G (1990)
Marked increase in calcitonin gene-related peptide containing
nerves in the developing rat following long-term sympathect-
omy with guanethidine. Neurosciences 35, 175–184.
Ainsworth A, Halim P, Wall PD, Allt G, Mackenzie M, Gibson S
& Polak J (1981) Effects of capsaicin applied locally to adult
peripheral nerves. II. Anatomy and enzyme and peptide
chemistry of peripheral nerve and spinal cord. Pain 11,
379–388.
Ballantyne B (1968) Histochemical and biochemical aspects of
cholinesterase activity of adipose tissue. Archives Internatio-
nales de Pharmacodynamie et de Therapie 173, 343–350.
Ballantyne B & Raftery AT (1974) The intrinsic autonomic
innervation of white adipose tissue. Cytobios 10, 187–197.
Ballard K, Malmfors T & Rosell S (1974) Adrenergic innervation
and vascular patterns in canine adipose tissue. Microvascular
Research 8, 164–171.
Bamshad M, Aoki VT, Adkison MG, Warren WS & Bartness TJ
(1998) Central nervous system origins of the sympathetic
nervous system outflow to white adipose tissue. American
Journal of Physiology 275, R291–R299.
Bamshad M, Song CK & Bartness TJ (1999) CNS origins of the
sympathetic nervous system outflow to brown adipose tissue.
American Journal of Physiology 276, R1569–R1578.
Bartness TJ (1995) Short day-induced depletion of lipid stores is
fat pad- and gender-specific in Siberian hamsters. Physiology
and Behavior 58, 539–550.
Bartness TJ (1996) Photoperiod, sex, gonadal steroids and
housing density affect body fat in hamsters. Physiology and
Behavior 60, 517–529.
Bartness TJ (2002) Dual innervation of white adipose tissue:
some evidence for parasympathetic nervous system involve-
ment. Journal of Clinical Investigation 110, 1235–1237.
Bartness TJ & Bamshad M (1998) Innervation of mammalian
white adipose tissue: implications for the regulation of total
body fat. American Journal of Physiology 275, R1399–R1411.
Bartness TJ, Demas GE & Song CK (2001) Central nervous
system innervation of white adipose tissue. In Adipose Tissue,
pp. 116–130 [S Klaus, editor]. Georgetown, TX: Landes
Bioscience.
Bartness TJ, Demas GE & Song CK (2002) Seasonal changes in
adiposity: the roles of the photoperiod, melatonin and other
hormones and the sympathetic nervous system. Experimental
Biology and Medicine 227, 363–376.
Bartness TJ & Fine JB (1999) Melatonin and seasonal changes in
body fat. In Melatonin in the Promotion of Health, pp.
115–135 [RR Watson, editor]. Boca Raton, FL: CRC.
Bartness TJ & Goldman BD (1988) Peak duration of serum
melatonin and short day responses in adult Siberian hamsters.
American Journal of Physiology 255, R812–R822.
Bartness TJ & Goldman BD (1989) Mammalian pineal melato-
nin: A clock for all seasons. Experientia 45, 939–945.
Bartness TJ, Goldman BD & Bittman EL (1991) SCN lesions
block responses to systemic melatonin infusions in Siberian
hamsters. American Journal of Physiology 260, R102–R112.
Bartness TJ, Hamilton JM, Wade GN & Goldman BD (1989)
Regional differences in fat pad responses to short days in
Siberian hamsters. American Journal of Physiology 257,
R1533–R1540.
Bartness TJ & Wade GN (1985) Photoperiodic control of
seasonal body weight cycles in hamsters. Neuroscience and
Behavioral Reviews 9, 599–612.
Bouloumie A, Planat V, Devedjian JC, Valet P, Saulnier-Blache
JS, Record M & Lafontan M (1994) Alpha 2-adrenergic
stimulation promotes preadipocyte proliferation. Involvement
of mitogen-activated protein kinases. Journal of Biological
Chemistry 269, 30254–30259.
Bray GA (1990) Obesity – a state of reduced sympathetic activity
and normal or high adrenal activity (the autonomic and adrenal
hypothesis revisited). International Journal of Obesity 14,
77–91.
Biology of obesity 61
Bray GA (1991) Obesity, a disorder of nutrient partitioning: the
MONA LISA hypothesis. Journal of Nutrition 121,
1146–1162.
Bray GA & Nishizawa Y (1978) Ventromedial hypothalamus
modulates fat mobilisation during fasting. Nature 274,
900–901.
Butler AA & Cone RD (2002) The melanocortin receptors:
lessons from knockout models. Neuropeptides 36, 77–84.
Cannon WB (1939) The Wisdom of the Body. New York: Norton.
Cantu RC & Goodman HM (1967) Effects of denervation and
fasting on white adipose tissue. American Journal of Physi-
ology 212, 207–212.
Card JP, Rinaman L, Schwaber JS, Miselis RR, Whealy ME,
Robbins ME & Enquist LW (1990) Neurotropic properties of
pseudorabies virus: uptake and transneuronal passage in the rat
central nervous system. Journal of Neuroscience 10,
1974–1994.
Castan I, Valet P, Larrouy D, Voisin T, Remaury A, Daviaud D,
Laburthe M & Lafontan M (1993) Distribution of PYY
receptors in human fat cells: an antilipolytic system alongside
the alpha2-adrenergic system. American Journal of Physiology
265, E74–E80.
Castan K, Valet P, Quideau N, Voisin T, Ambid L, Laburthe M,
Lafontan M & Carpene C (1994) Antilipolytic effects of
alpha2-adrenergic agonists, neuropeptide Y, adenosine, and
PGE1 in mammal adipocytes. American Journal of Physiology
266, R1141–R1147.
Clement G (1950) La mobilisation des glycerides de reserve chez
le rat. II. Influence due systeme nerveux sympathique.
Etablissement d’un procede de mesure de I’intensite de la
mobilisation (The mobilisation of glyceride reserves in the rat.
II. Influence of the sympathetic nervous system. Establishing a
process for measuring the intensity of the mobilisation).
Archives de Sciences Physiologiques 4, 13–29.
Considine RV, Considine EL, Williams CJ, Hyde TM & Caro JF
(1996) The hypothalamic leptin receptor in humans: identifi-
cation of incidental sequence polymorphisms and absence of
the db/db mouse and fa/fa rat mutations. Diabetes 45, 992–994.
Cousin B, Casteilla L, Lafontan M, Ambid L, Langin D,
Berthault M-F & Penicaud L (1993) Local sympathetic
denervation of white adipose tissue in rats induces preadipo-
cyte proliferation without noticeable changes in metabolism.
Endocrinology 133, 2255–2262.
Daniel H & Derry DM (1969) Criteria for differentiation of
brown and white fat in the rat. Canadian Journal of Physiology
and Pharmacology 47, 941–945.
Demas GE & Bartness TJ (2001a) Novel method for localized,
functional sympathetic nervous system denervation of periph-
eral tissue using guanethidine. Journal of Neuroscience
Methods 112, 21–28.
Demas GE & Bartness TJ (2001b) Direct innervation of white fat
and adrenal medullary catecholamines mediate photoperiodic
changes in body fat. American Journal of Physiology 281,
R1499–R1505.
Diculescu I & Stoica M (1970) Fluorescence histochemical
investigations on the adrenergic innervation of the white
adipose tissue in the rat. Journal of Neuro-visceral Relations
32, 25–36.
DiGirolamo M, Fine JB, Tagra K & Rossmanith R (1998)
Qualitative regional differences in adipose tissue growth and
cellularity in male Wistar rats fed ad libitum. American
Journal of Physiology 274, R1460–R1467.
Dodt C, Lonnroth P, Wellhoner JP, Fehm HL & Elam M (2003)
Sympathetic control of white adipose tissue in lean and obese
humans. Acta Physiologica Scandinavica 177, 351–357.
Dogiel AS (1898) Die sensiblen Nervenendigungen im Herzen
und in den Blutgefassen der Saugethiere (The afferent nerve
endings in the heart and in the blood vessels of mammals).
Archiv fu
¨
r Mikroskopischen Anatomie 52, 44–70.
Dua A, Hennes MI, Hoffmann RG, Maas DL, Krakower GR,
Sonnenberg GE & Kissebah AH (1996) Leptin: a significant
indicator of total body fat but not of visceral fat and insulin
insensitivity in African-American women. Diabetes 45,
1635–1637.
Dulloo AG & Miller DS (1987) Obesity: a disorder of the
sympathetic nervous system. World Review of Nutrition and
Dietetics 50, 1–56.
Edwards SL, Anderson CR, Southwell BR & McAllen RM
(1996) Distinct preganglioinic neurons innervate noradrenaline
and adrenaline cells in the cat adrenal medulla. Neurosciences
70, 825–832.
Enquist LW & Card JP (2003) Recent advances in the use of
neurotropic viruses for circuit analysis. Current Opinion in
Neurobiology 13, 603–606.
Faust IM (1984) Role of the fat cell in energy balance
physiology. In Eating and Its Disorders, pp. 97–107 [AJ
Stunkard and E Stellar, editors]. New York: Raven Press.
Ferry G, Tellier E, Try A, Gres S, Naime I, Simon MF et al. (2003)
Autotaxin is released from adipocytes, catalyzes lysopho-
sphatidic acid synthesis, and activates preadipocyte prolifera-
tion. Up-regulated expression with adipocyte differentiation
and obesity. Journal of Biological Chemistry 278, 18162–18169.
Fishman RB & Dark J (1987) Sensory innervation of white
adipose tissue. American Journal of Physiology 253, R942–
R944.
Frederich RC, Lollmann B, Hamann A, Napolitano-Rosen A,
Kahn BB, Lowell BB & Flier JS (1995) Expression of
ob mRNA and its encoded protein in rodents. Impact of
nutrition and obesity. Journal of Clinical Investigation 96,
1658–1663.
Fredholm BB (1985) Nervous control of circulation and meta-
bolism in white adipose tissue. In New Perspectives in Adipose
Tissue: Structure, Function and Development, pp. 45–64 [A
Cryer and RLR Van, editors]. Boston, MA: Butterworth.
Garofalo MAR, Kettelhut IC, Roselino JES & Migliorini RH
(1996) Effect of acute cold exposure on norepinephrine
turnover rates in rat white adipose tissue. Journal of the
Autonomic Nervous System 60, 206–208.
Gasteyger C & Tremblay A (2002) Metabolic impact of body fat
distribution. Journal of Endocrinological Investigation 25,
876–883.
Giordano A, Morroni M, Santone G, Marchesi GF & Cinti S
(1996) Tyrosine hydroxylase, neuropeptide Y, substance P,
calcitonin gene-related peptide and vasoactive intestinal pep-
tide in nerves of rat periovarian adipose tissue: an immuno-
histochemical and ultrastructural investigation. Journal of
Neurocytology 25, 125–136.
Gold RM (1973) Hypothalamic obesity: the myth of the
ventromedial nucleus. Science 182, 488–490.
Gold RM, Jones AP & Sawchenko PE (1977) Paraventricular
area: critical focus of a longitudinal neurocircuitry mediating
food intake. Physiology and Behavior 18, 1111–1119.
Gwyn DG, Ritchie TC & Coulter JD (1985) The central
distribution of vagal catecholaminergic neurons which project
into the abdomen in the rat. Brain Research 328, 139–144.
Hardie LJ, Rayner DV, Holmes S & Trayhurn P (1996)
Circulating leptin levels are modulated by fasting, cold
exposure and insulin administration in lean but not Zucker
(fa/fa) rats as measured by ELISA. Biochemical and Bio-
physical Research Communications 223, 660–665.
Harris RBS (2000) Leptin – much more than a satiety signal.
Annual Review of Nutrition 20, 45–75.
Haskell-Luevano C, Miwa H, Dickinson C, Hruby VJ, Yamada T
& Gantz I (1994) Binding and cAMP studies of melanotropin
62 T. J. Bartness et al.
peptides with the cloned human peripheral melanocortin
receptor, hMC1R. Biochemical and Biophysical Research
Communications 204, 1137–1142.
Hausman DB, DiGirolamo M, Bartness TJ, Hausman GJ &
Martin RJ (2001) The biology of white adipocyte proliferation.
Obesity Reviews 2, 239–254.
Hill B, Ralevic V, Crowe R & Burnstock G (1996) Innervation
and nitric oxide modulation of mesenteric arteries of
the Golden hamster. European Journal of Pharmacology 317,
275–283.
Hillarp NA & Hokfelt B (1953) Evidence of adrenaline and
noradrenaline in separate adrenal medullary cells. Acta
Physiologica Scandinavica 30, 55–68.
Hoffman GE, Smith MS & Verbalis JG (1993) c-Fos and related
immediate early gene products as markers of activity in
neuroendocrine systems. Frontiers in Neuroendocrinology 14,
173–213.
Hwa JJ, Ghibaudi L, Gao J & Parker EM (2001) Central
melanocortin system modulates energy intake and expenditure
of obese and lean Zucker rats. American Journal of Physiology
281, R444–R451.
Jansco G, Kiraly E & Jansco-Gabor A (1980) Direct evidence for
an axonal site of action of capsaicin. Naunyn Schmiedeburgs
Archives of Pharmacology 31, 91–94.
Janssen I, Fortier A, Hudson R & Ross R (2002) Effects of an
energy-restrictive diet with or without exercise on abdominal
fat, intermuscular fat, and metabolic risk factors in obese
women. Diabetes Care 25, 431–438.
Jones DD, Ramsay TG, Hausman GJ & Martin RJ (1992)
Norepinephrine inhibits rat pre-adipocyte proliferation. Inter-
national Journal of Obesity 16, 349–354.
Kalia M, Fuxe K, Goldstein M, Harfstrand A, Agnati LF &
Coyle JT (1984) Evidence for the existence of putative
dopamine-, adrenaline- and noradrenaline-containing vagal
motor neurons in the brainstem of the rat. Neuroscience Letters
50, 57–62.
Kirtland J & Gurr MI (1979) Adipose tissue cellularity: a review
2. The relationship between cellularity and obesity. Interna-
tional Journal of Obesity 3, 15–55.
Kras KM, Hausman DB, Hausman GJ & Martin RJ (1999)
Adipocyte development is dependent upon stem cell recruitment
and proliferation of preadipocytes. Obesity Research 7, 491–497.
Kreier F, Fliers E, Voshol PJ, Van Eden CG, Havekes LM,
Kalsbeek A, Van Heijningen CL, Sluiter AA, Mettenleiter TC,
Romijn JA, Sauerwein HP & Buijs RM (2002) Selective
parasympathetic innervation of subcutaneous and intra-abdom-
inal fat – functional implications. Journal of Clinical Investi-
gation 110, 1243–1250.
Lafontan M & Berlan M (1993) Fat cell adrenergic receptors and
the control of white and brown fat cell function. Journal of
Lipid Research 34, 1057–1091.
Lazzarini SJ & Wade GN (1991) Role of sympathetic nerves in
effects of estradiol on rat white adipose tissue. American
Journal of Physiology 260, R47–R51.
Luiten PGM, ter Horst GJ, Karst H & Steffens AB (1985) The
course of paraventricular hypothalamic efferents to autonomic
structures in medulla and spinal cord. Brain Research 329,
374–378.
Maffei M, Fei H, Lee GH, Dani C, Leroy P, Zhang Y, Proenca R,
Negrel R, Ailhaud G & Friedman JM (1995) Increased
expression in adipocytes of ob RNA in mice with lesions of
the hypothalamus and with mutations at the db locus.
Proceedings of the National Academy of Sciences USA 92,
6957–6960.
Mantzoros CS, Qu DQ, Frederich RC, Susulic VS, Lowell BB,
Maratos-Flier E & Flier JS (1996) Activation of b
3
adrenergic
receptors suppresses leptin expression and mediates a
leptin-independent inhibition of food intake in mice. Diabetes
45, 909–914.
Mauer MM & Bartness TJ (1996) Photoperiod-dependent fat pad
mass and cellularity changes following partial lipectomy in
Siberian hamsters. American Journal of Physiology 270,
R383–R392.
Migliorini RH, Garofalo MAR & Kettelhut IC (1997) Increased
sympathetic activity in rat white adipose tissue during
prolonged fasting. American Journal of Physiology 272,
R656–R661.
Mione M, Cavanagh JFR, Kirkpatrick KA & Burnstock G (1992)
Plasticity in expression of calcitonin gene-related peptide and
substance P immunoreactivity in ganglia and fibres following
guanethidine and/or capsaicin denervation. Cell Tissue
Research 268, 491–504.
Morrison SF (2001) Differential control of sympathetic outflow.
American Journal of Physiology 281, R683–R698.
Newsholme EA & Leech AR (1983) Biochemistry for the
Medical Sciences. Chichester, West Sussex: John Wiley.
Ng TB & Wong CM (1986) Effects of pineal indoles and arginine
vasotocin on lipolysis and lipogenesis in isolated adipocytes.
Journal of Pineal Research 3, 55–66.
Niijima A (1998) Afferent signals from leptin sensors in the
white adipose tissue of the epididymis, and their reflex effect
in the rat. Journal of the Autonomic Nervous System 73, 19–25.
Nishizawa Y & Bray GA (1978) Ventromedial hypothalamic
lesions and the mobilization of fatty acids. Journal of Clinical
Investigation 61, 714–721.
Ollmann MM, Wilson BD, Yang YK, Kerns JA, Chen Y, Gantz I
& Barsh GS (1997) Antagonism of central melanocortin
receptors in vitro and in vivo by agouti-related protein. Science
278, 135–138.
Orban Z, Remaley AT, Sampson M, Trajanoski Z & Chrousos
GP (1999) The differential effect of food intake and beta-
adrenergic stimulation on adipose-derived hormones and
cytokines in man. Journal of Clinical Endocrinology and
Metabolism 84, 2126–2133.
Ostlund RE Jr, Yang JW, Klein S & Gingerich R (1996) Relation
between plasma leptin concentration and body fat, gender, diet,
age, and metabolic covariates. Journal of Clinical Endocrino-
logy and Metabolism 81, 3909–3913.
Ozaki N, Sugiura Y, Yamamoto M & Nishiyama Y (1996)
Induction of Fos protein expression in spinal cord neurons by
herpes simplex virus infections in the mouse. Neuroscience
Letters 216, 61–64.
Pasanisi F, Contaldo F, de Simone G & Mancini M (2001)
Benefits of sustained moderate weight loss in obesity. Nutrition
Metabolism and Cardiovascular Diseases 11, 401–406.
Powley TL, Walgren MC & Laughton WB (1983) Effects of
guanethidine sympathectomy on ventromedial hypothalamic
obesity. American Journal of Physiology 245, R408–R420.
Ralevic V, Karoon P & Burnstock G (1995) Long-term sensory
denervation by neonatal capsaicin treatment augments sympa-
thetic neurotransmission in rat mesenteric arteries by increas-
ing levels of norepinephrine and selectively enhancing
postjunctional actions. Journal of Pharmacology and Experi-
mental Therapeutics 274, 64–71.
Raposinho PD, White RB & Aubert ML (2003) The melanocortin
agonist melanotan-II reduces the orexigenic and adipogenic
effects of neuropeptide Y (NPY) but does not affect the NPY-
driven suppressive effects on the gonadotropic and somato-
tropic axes in the male rat. Journal of Neuroendocrinology 15,
173–181.
Rayner DV, Simon E, Duncan JS & Trayhurn P (1998)
Hyperleptinaemia in mice induced by administration of the
tyrosine hydroxylase inhibitor alpha-methyl-p-tyrosine. FEBS
Letters 429, 395–398.
Biology of obesity 63
Rebuffe-Scrive M (1991) Neuroregulation of adipose tissue:
molecular and hormonal mechanisms. International Journal of
Obesity 15, 83–86.
Rizack MA (1961) An epinephrine-sensitive lipolytic activity in
adipose tissue. Journal of Biological Chemistry 236, 657–662.
Robidoux J, Pirouzi P, Lafond J & Savard R (1995) Site-specific
effects of sympathectomy on the adrenergic control of lipolysis
in hamster fat cells. Canadian Journal of Physiology and
Pharmacology 73, 450–458.
Rubino A, Ralevic V & Burnstock G (1997) Sympathetic
neurotransmission in isolated rat atria after sensory-motor
denervation by neonatal treatment with capsaicin. Journal of
Pharmacology and Experimental Therapeutics 282, 671–675.
Schafer MK, Eiden LE & Weihe E (1998) Cholinergic neurons
and terminal fields revealed by immunohistochemistry for the
vesicular acetylcholine transporter. II. The peripheral nervous
system. Neuroscience 84, 361–376.
Shi H & Bartness TJ (2001) Neurochemical phenotype of
sympathetic nervous system outflow from brain to white fat.
Brain Research Bulletin 54, 375–385.
Shi H, Giordano A, Cinti S & Bartness TJ (2005) Sensory or
sympathetic white adipose tissue denervation differentially
affects depot growth and cellularity. American Journal of
Physiology (In the Press).
Sivitz WI, Fink BD, Morgan DA, Fox JM, Donohoue PA &
Haynes WG (1999) Sympathetic inhibition, leptin, and
uncoupling protein subtype expression in normal and fasting
rats. American Journal of Physiology 277, E668–E677.
Skofitsch G & Jacobowitz DM (1985) Calcitonin gene-related
peptide coexists with substance P in capsaicin sensitive
neurons and sensory ganglia of the rat. Peptides 6, 747–754.
Slavin BG & Ballard KW (1978) Morphological studies on the
adrenergic innervation of white adipose tissue. Anatomical
Record 191, 377–389.
Song CK & Bartness TJ (1996) The effects of anterior hypo-
thalamic lesions on short-day responses in Siberian hamsters
given timed melatonin infusions. Journal of Biological
Rhythms 11, 14–26.
Song CK & Bartness TJ (1998) Dorsocaudal SCN microknife-
cuts do not block short day responses in Siberian hamsters
given melatonin infusions. Brain Research Bulletin 45,
239–246.
Song CK & Bartness TJ (2001) CNS sympathetic outflow neurons
to white fat that express melatonin receptors may mediate
seasonal adiposity. American Journal of Physiology 281,
R666–R672.
Strack AM & Loewy AD (1990) Pseudorabies virus: a highly
specific transneuronal cell body marker in the sympathetic
nervous system. Journal of Neuroscience 10, 2139–2147.
Strack AM, Sawyer WB, Platt KB & Loewy AD (1989) CNS cell
groups regulating the sympathetic outflow to adrenal gland as
revealed by transneuronal cell body labeling with pseudorabies
virus. Brain Research 491, 274–296.
Sved AF, Cano G & Card JP (2001) Neuroanatomical specificity
of the circuits controlling sympathetic outflow to different
targets. Clinical and Experimental Pharmacology and Physio-
logy 28, 115–119.
Takahashi A, Ikarashi Y, Ishimaru H & Maruyama Y (1993)
Compensation between sympathetic nerves and adrenal medul-
lary activity: effects of adrenomedullation and chemical
sympathectomy on catecholamine turnover. Life Sciences 53,
1567–1572.
Takahashi A & Shimazu T (1981) Hypothalamic regulation of
lipid metabolism in the rat: effect of hypothalamic stimulation
on lipolysis. Journal of the Autonomic Nervous System 4,
195–205.
Trayhurn P, Duncan JS, Hoggard N & Rayner DV (1998) Regu-
lation of leptin production: a dominant role for the sympathetic
nervous system. Proceedings of the Nutrition Society 57,
413–419.
Trayhurn P, Duncan JS & Rayner DV (1995) Acute cold-induced
suppression of ob (obese) gene expression in white adipose
tissue of mice: mediation by the sympathetic nervous system.
Biochemical Journal 311, 729–733.
Trayhurn P, Duncan JS, Rayner DV & Hardie LJ (1996) Rapid
inhibition of ob gene expression and circulating leptin levels in
lean mice by the beta 3-adrenoceptor agonists BRL 35135A
and ZD2079. Biochemical and Biophysical Research Commu-
nications 228, 605–610.
Turek FW & Campbell CS (1979) Photoperiodic regulation of
neuroendocrine-gonadal activity. Biology of Reproduction 20,
32–50.
Turtzo LC, Marx R & Lane MD (2001) Cross-talk between
sympathetic neurons and adipocytes in coculture. Proceedings
of the National Academy of Sciences USA 98, 12385–12390.
Underwood H & Goldman BD (1987) Vertebrate circadian and
photoperiodic systems: role of the pineal gland and melatonin.
Journal of Biological Rhythms 2, 279–315.
Vague J, Vague P, Tramoni M, Vialettes B & Mercier P (1980)
Obesity and diabetes. Acta Diabetologica Latina 17, 87–99.
Valet P, Pages C, Jeanneton O, Daviaud D, Barbe P, Record M,
Saulnier-Blache JS & Lafontan M (1998) Alpha 2-adrenergic
receptor-mediated release of lysophosphatidic acid by adipo-
cytes. A paracrine signal for preadipocyte growth. Journal of
Clinical Investigation 101, 1431–1438.
Wade GN & Bartness TJ (1984) Effects of photoperiod and
gonadectomy on food intake, body weight and body composi-
tion in Siberian hamsters. American Journal of Physiology 246,
R26–R30.
Weiss ML & Chowdhury SI (1998) The renal afferent pathways
in the rat: a pseudorabies virus study. Brain Research 812,
227–241.
White JE & Engel FL (1958) A lipolytic action of epinephrine
and norepinephrine on rat adipose tissue in vitro. Proceedings of
the Society for Experimental Biology and Medicine 99, 375–378.
Wirsen C (1964) Adrenergic innervation of adipose tissue
examined by fluorescence microscopy. Nature 202, 913.
Wright JT & Hausman GJ (1990) Monoclonal antibodies against
cell surface antigens expressed during porcine adipocyte
differentiation. International Journal of Obesity 14, 284–291.
Young JB, Saville E, Rothwell NJ, Stock MJ & Landsberg L
(1982) Effect of diet and cold exposure on norepinephrine
turnover in brown adipose tissue of the rat. Journal of Clinical
Investigation 69, 1061–1071.
Youngstrom TG & Bartness TJ (1995) Catecholaminergic
innervation of white adipose tissue in the Siberian hamster.
American Journal of Physiology 268, R744–R751.
Youngstrom TG & Bartness TJ (1998) White adipose tissue
sympathetic nervous system denervation increases fat pad mass
and fat cell number. American Journal of Physiology 275,
R1488–R1493.
Zhang Y, Proenca R, Maffei M, Barone M, Leopold L &
Friedman JM (1994) Positional cloning of the mouse obese
gene and its human homologue. Nature 372, 425–432.
64 T. J. Bartness et al.
