Sympathetic and sensory innervation of white adipose tissue

Text-only Preview

thematic review
Thematic review series: Adipocyte Biology
Sympathetic and sensory innervation of white
adipose tissue
Timothy J. Bartness1 and C. K. Song
Department of Biology, Neurobiology and Behavior Program, Georgia State University, Atlanta,
GA 30302-4010
During our study of the reversal of seasonal obe-
year (6) to as low as ?26,000 (7), making overweight/
sity in Siberian hamsters, we found an interaction between
obesity either the number two or number seven cause of
receptors for the pineal hormone melatonin and the sym-
death in adults in the United States, respectively. Con-
pathetic nervous system (SNS) outflow from brain to white
siderable research effort has focused on determining the
Downloaded from
adipose tissue (WAT). This ultimately led us and others
to conclude that the SNS innervation of WAT is the pri-
factors involved in the development of obesity in non-
mary initiator of lipid mobilization in these as well as other
human animals; consequently, our understanding of these
animals, including humans. There is strong neurochemical
is substantial, but still far from complete. Somewhat sur-
(norepinephrine turnover), neuroanatomical (viral tract
prisingly, less effort has focused on determining the
tracing), and functional (sympathetic denervation-induced
factors involved in the reversal of obesity in nonhuman
blockade of lipolysis) evidence for the role of the SNS in
animals. We have contributed to these latter efforts by
lipid mobilization. Recent findings suggest the presence of
studying the reversal of a naturally occurring seasonal
WAT sensory innervation based on strong neuroanatomi-
obesity in Siberian hamsters (Phodopus sungorus). In the
cal (viral tract tracing, immunohistochemical markers of
process of conducting this work, we discovered the im-
by guest, on October 13, 2010
sensory nerves) and suggestive functional (capsaicin sensory
denervation-induced WAT growth) evidence, the latter im-
portance of the sympathetic nervous system (SNS) in lipid
plying a role in conveying adiposity information to the brain.
mobilization from white adipose tissue (WAT) as well as
By contrast, parasympathetic nervous system innervation of
more recently realizing the possible importance of the
WAT is characterized by largely negative neuroanatomical
sensory innervation of WAT. Several overviews of the role
evidence (viral tract tracing, immunohistochemical and bio-
of the SNS in lipid mobilization were published recently
chemical markers of parasympathetic nerves).
(8–11). To give a better understanding of how we came to
evidence (intraneural stimulation and in situ microdialysis)
the realization that lipid mobilization occurs primarily
for the role of the SNS innervation in lipid mobilization in
through the sympathetic innervation of WAT, we will de-
human WAT is convincing, with some controversy regarding
the level of sympathetic nerve activity in human obesity.—
scribe some background studies focused on the reversal
Bartness, T. J., and C. K. Song. Sympathetic and sensory
of photoperiod-induced obesity that ultimately led us to
innervation of white adipose tissue. J. Lipid Res. 2007. 48:
define and test the SNS innervation of WAT.
Supplementary key words
obesity & humans & viral tract tracing &
melanocortin & melatonin & denervation & proliferation & adipocyte &
hamster & rat & mouse & human
Abbreviations: BAT, brown adipose tissue; BrDU, bromodeoxyuri-
dine; CGRP, calcitonin gene-related peptide; CNS, central nervous
system; DRG, dorsal root ganglia; DWAT, dorsosubcutaneous white
Obesity is a disease of literally and figuratively enormous
adipose tissue; EPI, epinephrine; EWAT, epididymal white adipose
proportions (1, 2) and is an independent risk factor for
tissue; FCN, fat cell number; FCS, fat cell size; HIV, human immuno-
type II diabetes, cardiovascular disease, stroke, and some
deficiency virus; HSL, hormone-sensitive lipase; HSV, herpes simplex
virus; IBAT, interscapular brown adipose tissue; -ir, immunoreactivity;
cancers (3–5). Estimates of the mortality rate of over-
IWAT, inguinal white adipose tissue; LD, long day; MC4-R, melano-
weight/obese individuals range from ?325,000 deaths per
cortin 4-receptor; MEL, melatonin; NE, norepinephrine; NETO,
norepinephrine turnover; 6OHDA, 6-hydroxy-dopamine; PRV, pseu-
dorabies virus; PSNS, parasympathetic nervous system; PVN, paraven-
tricular nucleus; RWAT, retroperitoneal white adipose tissue; SCN,
suprachiasmatic nucleus; SD, short day; SNS, sympathetic nervous
Manuscript received 21 March 2007 and in revised form 17 April 2007.
system; TH, tyrosine hydroxylase; WAT, white adipose tissue.
Published, JLR Papers in Press, April 25, 2007.
To whom correspondence should be addressed.
DOI 10.1194/jlr.R700006-JLR200
e-mail: [email protected]
Copyright D 2007 by the American Society for Biochemistry and Molecular Biology, Inc.
This article is available online at
Journal of Lipid Research
Volume 48, 2007

retina and transmitted through a multisynaptic pathway
that includes the suprachiasmatic nucleus [SCN; the pri-
mary biological clock (32)], the hypothalamic paraven-
tricular nucleus (PVN), and the intermedial lateral horn
Animals living in temperate zones experience wide
of the spinal cord that ultimately terminates in the pineal
fluctuations in their environment, including changes in
gland (33), where this photic information is transduced
daylength, ambient temperature, rainfall, and vegetation.
into an endocrine signal via the rhythmic pattern of se-
Therefore, it is not surprising that animals indigenous
cretion of its principal hormone, melatonin (MEL) (15).
to such environments have evolved with the ability to
Specifically, MEL is only synthesized and secreted by
markedly modify their physiology and behavior in antic-
pinealocytes at night; thus, the duration of night is faith-
ipation of the approaching season. Through natural selec-
fully coded by the duration of nocturnal MEL secretion,
tion, animals that survived and reproduced were those
thereby triggering seasonal responses: long durations of
responding not to the cues of the current season but to
the circulating MEL signal short “winter-like” days (SDs),
the cues occurring within the current season that predict
and short durations of circulating MEL signal long
the forthcoming season. This makes sense in that many
“summer-like” days (LDs) (for reviews, see Refs. 15, 34).
of the seasonal physiological and behavioral modifications
The MEL receptor subtype mediating photoperiodic re-
require weeks or months to be fully manifested and could
sponses is the MEL1a receptor [or the mt1 receptor (35)].
not occur rapidly enough to be beneficial for the current
MEL itself does not directly trigger lipolysis; incubation of
season. For a wide variety of animals, including species of
isolated white adipocytes in vitro with physiological or
birds, lizards, amphibians, and mammals, the predomi-
even “industrial-strength” doses of MEL does not increase
nant environmental cue that triggers seasonally adaptive
lipolysis (36). We tested hormones that both changed sea-
responses is the rate of change in the photoperiod: that is,
sonally in Siberian hamsters and either directly or indi-
Downloaded from
how rapidly or how slowly daylength (or nightlength)
rectly affected lipolysis in these or other animals (e.g.,
lengthens or shortens (for reviews, see Refs. 12–16). If it
insulin, prolactin, glucocorticoids, gonadal steroids, thy-
was the absolute daylength, rather than the rate of progres-
roid hormones). None of these hormones, at least singly,
sion from long to short or short to long photoperiods, that
accounted for the SD-induced decrease in body fat,
triggered seasonal responses, then it would be impossible
as tested in depletion-repletion studies (for review, see
to differentiate a spring day with 14 h of light from a fall
Ref. 37).
day with 14 h of light.
When Siberian hamsters are exposed to long days
(LDs), they develop a severe obesity [?50% body fat (17–
by guest, on October 13, 2010
19)], a level as extreme as that seen in genetically inbred
strains of rats and mice [e.g., ob/ob mice (20) and Zucker
fa/fa rats (21)]. The LD-induced obesity of Siberian ham-
sters is completely reversible by exposure to short days
Because of the largely negative nature of studies testing
(SDs), with the decrease in body fat being most rapid
for the hormonal mediation of SD-induced decreases in
during the first 5–6 weeks of SDs (17, 18, 22–24). More-
adiposity in Siberian hamsters, we tested the role of ad-
over, the body fat loss during this time occurs without
renal medulla-secreted epinephrine (EPI). At that time,
a concomitant decrease in food intake but eventually
EPI was considered the principal initiator of lipolysis be-
(.6 weeks of SD exposure) is accompanied by an ?30%
cause of its ability to robustly simulate lipolysis (glycerol
decrease in food intake (17, 25, 26); thus, the initial body
release) in vitro in isolated white adipocytes (38–41). Tests
fat loss is attributable to increased energy expenditure.
of this notion in vivo, however, do not support an im-
Unlike other photoperiod-induced obesity models (27),
portant role of EPI in lipolysis. Specifically, despite adrenal
the SD-induced decrease in WAT is nonuniform, with the
demedullation removing the sole source of circulating
more internally located visceral fat pads mobilizing their
EPI, lipid mobilization is unaffected by several stimuli
lipid stores first and to a greater extent than the more
known to stimulate lipolysis [e.g., glucoprivation (42) and
externally located subcutaneous fat pads (18, 28, 29).
electrical stimulation of the medial hypothalamus (43)],
This differential lipid mobilization across WAT depots also
including SD-induced lipid mobilization in Siberian ham-
is a characteristic of body fat loss in exercising/dieting/
sters (44). Thus, EPI is not necessary for lipolysis.
fasting humans (30, 31), although the exact pattern and
Histological evidence for the SNS innervation of WAT
fat pads involved are not homologous. Nevertheless, this
has a long history beginning with Dogiel (45) in 1898,
highlights another advantage of using this model to study
when he reported the staining of nerves innervating WAT.
obesity reversal. It was this differential pattern of lipid
Confusion arose initially regarding which components of
mobilization and, as we shall see, differential sympathetic
WAT received the SNS innervation, because there were
drive to WAT, that fascinated us early in our studies and
descriptions of only the sympathetic innervation of blood
prompted us to delve into the mechanisms underlying
vessels in WAT as well as other descriptions of the sym-
this phenomenon.
pathetic innervation of the WAT parenchyma (for reviews,
How does the daylength cue get transduced into a
see Refs. 9, 11). It is now realized that the failures to detect
biological signal? The photoperiod cue is received by the
neural fibers within the parenchymal space of WAT of
Journal of Lipid Research
Volume 48, 2007

normal ad libitum-fed laboratory rats was the result of the
ing the nucleus, hijack host machinery to produce its prog-
tight packing of the fully loaded, lipid-filled adipocytes in
eny, which are then exocytosed through the dendrites
the WAT matrix. By fasting animals and thus inducing
only. Neurons making synaptic contact with the infected
lipolysis, the consequent lipid mobilization causes de-
cells then become exposed to relatively high concentra-
creases in fat cell size (FCS), thereby exposing more pa-
tions of the new virions. The virions subsequently are
renchymal space and revealing catecholaminergic (via
taken up only at synaptic contact sites, and this process
histofluorescence) innervation of both the vasculature
continues, causing an infection along the neuronal chain
and white adipocytes (46–49). The nerves innervating
from the inoculation site in WAT to higher central nervous
white adipocytes are not as boutons juxtaposed closely to
system (CNS) sites. The infected neurons can then be
the fat cells; rather, they are of the en passant variety (50).
easily visualized using standard immunohistochemistry or,
This histofluorescence labeling of catecholaminergic
because PRV is relatively easily genetically engineered,
nerve fibers within WAT is not considered direct evidence
isogenic versions of the virus have been constructed to
of SNS innervation; tract tracing of the sources of input
produce fluorescent reporters [i.e., green (58) or red (59)
to the WAT pads is the sine qua non of direct innervation
fluorescent protein]. Because the transfer of the virus
of a tissue. Therefore, we used fluorescent tract tracers
only is by a transynaptic mechanism, rather than by lateral
(DiI, FluoroGold) to demonstrate bidirectionally direct
spread to adjacent but unrelated neurons or by a non-
neuroanatomical projections of postganglionic neurons
synaptic mechanism (60), the transynaptic transfer of the
residing in the sympathetic chain to WAT (26). In addi-
PRV after injection into WAT yields a hierarchical chain of
tion, we found evidence of largely, but not completely,
functionally connected neurons from brain to WAT (for
separate postganglionic neurons within the sympathetic
review, see Ref. 61).
chain innervating inguinal white adipose tissue (IWAT)
We retrogradely labeled the SNS outflow from brain to
and epididymal white adipose tissue (EWAT) (26). Such
WAT (IWAT and EWAT) in laboratory rats and Siberian
Downloaded from
segregation of populations of postganglionic neurons in-
hamsters (62) and later to retroperitoneal white adipose
nervating individual WAT pads could provide the neuro-
tissue (RWAT) in Siberian hamsters (63), thereby identi-
anatomical underlying basis for the differential mobilization
fying CNS-SNS-WAT circuitries (for review, see Ref. 11,
of lipid across individual WAT depots, as occurs with most
64). The general patterns of infection after WAT inocula-
lipolysis-promoting stimuli [e.g., starvation (51, 52), estra-
tion with PRV prominently show that WAT (62), as well as
diol (52), and leptin (53)]. A similar conclusion was drawn
brown adipose tissue (BAT) (65), receives input from CNS
more than a decade later (54) when one of two strains of
cell groups that are part of the general SNS outflow from
the pseudorabies virus (PRV), a transneuronal tract tracer
the brain [PVN, A5 of the noradrenergic lateral tegmental
(see directly below), each possessing a distinct fluorescent
system, caudal raphe region, rostral ventrolateral medulla,
by guest, on October 13, 2010
reporter, was injected into mesenteric and subcutaneous
ventromedial medulla (66), and many other areas as well
WAT. This resulted in the identification of an additional
(67)]. Some of these include the hindbrain: area post-
segregation of neurons innervating these two WAT pads
rema, nucleus of the solitary tract, and raphe regions
but located a synapse earlier than those we found in the
(e.g., pallidus, obscurus, magnus, and dorsal raphe and the
sympathetic chain (26), in the intermediolateral horn of
raphe cap); midbrain: periaqueductal gray and pontine
the spinal cord [i.e., the sympathetic preganglionic neu-
regions; and forebrain: hypothalamic arcuate, preoptic,
rons (54)]. Although partially satisfying, the determina-
SCN, PVN and dorsomedial nuclei, and thalamic para-
tion of the preganglionic and postganglionic sympathetic
ventricular and reuniens nuclei (for a complete list, see
innervation of WAT left open to speculation the origins of
Ref. 67).
the sympathetic outflow to WAT from the brain.
Although there were some differences in the degrees
of infection (i.e., neurons participating in the circuit) at
several sites across the neural axis among these WAT pads
in Siberian hamsters, the general patterns of infection
were more similar than different for the same WAT depots
between Siberian hamsters and laboratory rats (62, 63).
This should not be misconstrued as a dismissal of “viscero-
topic” sympathetic innervation of WAT at some point(s)
We were able to define the source of the premotor
across the neuroaxis (see above for evidence of pregan-
neurons participating in the sympathetic circuits inner-
glionic and postganglionic viscerotopic patterns of sym-
vating WAT because of the pioneering work of others
pathetic nerves innervating WAT); however, unequivocal
using the Bartha’s K strain of PRV to trace the SNS outflow
demonstration of the viscerotopic organization of WAT
to numerous peripheral tissues [e.g., adrenal (55) and
circuitries requires careful studies using isogenic strains of
pineal (33)]. In brief (for reviews, see Refs. 56, 57), neuro-
the PRV, each with unique fluorescent or other reporter
tropic viruses, such as PRV, are taken up into neurons after
injected into separate WAT pads (54). From the perspec-
binding to viral attachment proteins located on the sur-
tive of shared circuits, we found some common neurons
face of neuronal membranes and the viral envelope fusing
in the sympathetic outflow circuits innervating a WAT
with the cell membrane. The resulting capsids containing
(IWAT) and a BAT depot [interscapular brown adipose
the viral core DNA enter the cytoplasm and, upon reach-
tissue (IBAT)] in a preliminary experiment (61). It would
Sympathetic and sensory innervation of white fat

not be surprising, therefore, to see some shared neurons
discussed above, the potency/efficacy of norepinephrine
at some level(s) of the neuroaxis for the sympathetic out-
(NE)-triggered lipolysis increases in a temporally and fat
flow to two distinct WAT pads, despite the differential
pad-specific manner (80). That is, during the rapid de-
sympathetic drives to WAT, differential degrees of lipid
crease in body fat occurring during the first 5–6 weeks of
mobilization, and some shared circuits for WAT and BAT
SD exposure, the potency (sensitivity/EC50) and efficacy
discussed above.
(maximal response asymptote) of NE-induced lipolysis was
increased in isolated adipocytes from IWAT and EWAT
compared with isolated adipocytes from their LD counter-
parts (80). These enhanced responses to NE were most
prominent when lipid mobilization was greater (5 vs.
10 weeks of SD exposure) and were more marked in
EWAT than in IWAT, paralleling the greater decrease in
EWAT than in IWAT mass (80). Moreover, the SD stimu-
lation of lipolysis was similar for NE and BRL 37344 (a
Given our knowledge of the SNS outflow from brain to
specific b3-adrenoceptor agonist), implying the primacy of
WAT discussed above, and the failure to identify circulat-
this receptor subtype in this response (80) (see below for a
ing factors responsible for the photoperiod-triggered in-
brief discussion of b- and a-adrenoceptors in WAT). WAT
creases in WAT lipid mobilization of Siberian hamsters
b3-adrenoceptor mRNA expression (and, thus, likely pro-
also discussed above, we thought that perhaps MEL was
tein) also increases in SDs (81), suggesting that these
interacting with the sympathetic innervation of WAT at
increases in b3-adrenoceptors could underlie the SD-
some level(s) of the neuroaxis to increase the sympathetic
induced increased NE sensitivity/efficacy. Collectively,
drive (norepinephrine turnover [NETO]) to WAT (26).
therefore, there is a coordinated set of SD-triggered sym-
Downloaded from
As noted above, MEL1a receptors receive the nocturnal
pathetic responses that promote the seemingly effortless
MEL durational signal that codes the photoperiod in
shift from the obese to the lean state in Siberian hamsters
Siberian hamsters as well as in other species (68, 69).
experiencing SDs.
Therefore, we tested whether MEL1a receptor mRNA was
To recapitulate, we have defined the likely sequence of
colocalized with brain SNS outflow neurons to WAT
events from environmental cue (change in photoperiod)
labeled after PRV injections into the tissue, using emulsion
to decreases in adipocyte size that result in the reversal of
autoradiographic in situ hybridization and immunohisto-
photoperiod-induced seasonal obesity in Siberian ham-
chemistry, respectively (70). This seminal study was fo-
sters. Thus, with increasing durations of night (SDs), the
cused on forebrain only, but we realize that MEL binding/
peak duration of nocturnal MEL secretion by the pineal
by guest, on October 13, 2010
receptors have been localized in midbrain and hindbrain
is lengthened, resulting in increases in MEL1a receptor
(e.g., median and dorsal raphe, raphe obscurus, and
stimulation, some of which are located in the circuits con-
pontine reticular nuclei) in laboratory rats (71), areas that
stituting the sympathetic outflow neurons to WAT. This
also contain neurons that are part of the sympathetic
increase in the sympathetic drive to WAT stimulates pri-
outflow to WAT (62, 67, 70, 72). Gene expression for the
marily b3-receptors that have increased efficacy/potency
MEL1a receptors is predominantly in the paraventricular
to NE, the principal sympathetic postganglionic neuro-
and reuniens nuclei of the thalamus and in the SCN (70,
transmitter, as a result of SD exposure. This, in turn, initi-
73, 74). The SCN is critical for the reception of season-
ates the lipolytic cascade that ultimately results in decreases
encoded MEL signals, because pinealectomized Siberian
in total body fat of SD-exposed Siberian hamsters.
hamsters bearing SCN lesions do not decrease their body
We do not believe, however, that MEL interacting with
or lipid mass or regress their gonads when given exoge-
the SNS in humans is an important factor controlling lipid
nously administered SD MEL signals (75–77). More im-
mobilization. This does not diminish the importance of
portantly for our topic, there is extensive colocalization
these findings, because it led us and others to realize the
of MEL1a receptor mRNA with PRV-labeled neurons in
importance of the SNS innervation of fat for lipid mobili-
some brain sites involved in the sympathetic outflow from
zation. Moreover, as will be discussed below, the sympa-
CNS to WAT, including the SCN (70). This suggests that
thetic innervation of WAT appears to be the underlying
stimulation of these receptors increases the sympathetic
mechanism that initiates lipolysis under a number of con-
drive to WAT [as shown in SDs by increases in NETO
ditions in both human and nonhuman animals.
(26)]. This, in turn, would initiate lipolysis and thereby
drive the SD-induced increases in lipid mobilization by
this species.
Unlike humans, in which increases in energy expendi-
ture and/or decreases in energy intake trigger physiolog-
ical responses to counter these attempts at reductions in
adiposity (for review, see Refs. 78, 79), Siberian hamsters
Walter Cannon (82) put forth the theory that at times of
respond to SDs with a suite of coordinated sympathetic
emergency, a general SNS discharge was triggered, pre-
responses that work together to decrease body fat. Thus, in
paring animals for “fight or flight.” This “all-or-nothing”
addition to the SD-induced increase in WAT NETO (26)
view of the sympathetic activation of peripheral tissues,
Journal of Lipid Research
Volume 48, 2007

however, still appears to predominate, especially in the
in laboratory rats, there is a trend toward increases in
interpretation of SNS electrophysiological activity to pe-
RWAT compared with EWAT NETO (95), and with acute
ripheral tissues. For example, generalizations have been
cold exposure in Siberian hamsters, IWAT NETO is
made and applied to WAT from recordings made of the
significantly greater than RWAT or EWAT NETO (no
sympathetic nerves innervating BAT, kidney, and other
change) and dorsosubcutaneous white adipose tissue
tissues. Unfortunately, the sympathetic activity to periph-
(DWAT) NETO (M. Brito, N. Brito, C. K. Song, and T. J.
eral tissues is not analogous to the current flow to the
Bartness, unpublished data). Finally, another example
electrical outlets in your home, where the current is vir-
of differential sympathetic drive across WAT pads occurs
tually identical regardless of which outlet is measured. The
with the glucoprivic stimulus 2-deoxy-D-glucose: NETO
physiological reality of sympathetic nerve activity is that
is increased to several WAT pads (IWAT, RWAT, and
it differs not only across tissues (e.g., heart vs. BAT) but
DWAT) but not in EWAT (no change) in Siberian
within a type of tissue (WAT across its varied locations, as
hamsters (9; M. Brito, N. Brito, C. K. Song, and T. J.
discussed below).
Bartness, unpublished data). It should be noted that the
There have been several studies measuring the changes
relation between increases in NETO and decreases in
in the firing rate of sympathetic nerves innervating WAT,
WAT mass, as occurred in SD-exposed Siberian hamsters
all by Niijima and associates (83–90), showing that a num-
(26), does not always hold, as third ventricularly or
ber of stimuli increase sympathetic drive to WAT, includ-
peripherally administered leptin increases WAT NETO
ing odors, tastes, histamine, leptin, and other factors. This
but is not always positively correlated with decreases in
relative paucity of electrophysiological measures of sym-
WAT mass (96). When such a disparity is seen, it may be
pathetic nerves to WAT contrasts with that for BAT (for
because WAT pad mass is not a sensitive indicator of lipid
reviews, see Refs. 91, 92), perhaps because the nerves to
mobilization when the degree of lipolysis is more sub-
WAT are more difficult to identify and are smaller, making
tle. Alternatively, receptor or post receptor events could be
Downloaded from
them more difficult to record from than the sympathetic
altered such that the sympathetic nerve activity is without
nerves innervating BAT. In these studies of sympathetic
effect. In summary, differential WAT NETO exists for
nerve activity to WAT, this activity has not been measured
several lipid-promoting stimuli, but our knowledge of the
in more than one WAT depot simultaneously (two EWAT
control of this and other trafficking of sympathetic drives
pads have been measured simultaneously, but not, for
to peripheral tissues is an important unsolved mystery that
example, EWAT and IWAT), and as we shall see, based
seems fundamental to solve for a deeper understanding
on biochemical measures of sympathetic activity (NETO),
of regulatory biology (for review, see Ref. 97).
differential sympathetic drive across WAT pads almost al-
ways occurs.
by guest, on October 13, 2010
The deficit in our knowledge of sympathetic drives
to WAT has been approached by measuring NETO as a
proxy for direct electrophysiological measures. Most
frequently, the a-methyl-para-tyrosine method has been
used, in which a-methyl-para-tyrosine, a competitive in-
hibitor of tyrosine hydroxylase (TH), the rate-limiting en-
zyme in catecholamine biosynthesis, is exploited to allow
the estimation of NETO by the rate of NE disappearance
In our previous studies, we found CNS neurons infected
from the tissues of interest (93). Using this method, there-
after PRV inoculation in WAT that also had colocalized
fore, a measure of sympathetic drive can be compared
immunoreactivity for arginine vasopressin (98), TH (98),
across multiple tissue types and multiple adipose depots.
oxytocin (98), acetylcholine transferase (72), and MEL1a
When this is done, it is clear that there is differential
receptors (70), with the vast majority of cells not showing
NETO across WAT pads in response to many lipid-
an identified neurochemical phenotype. Three possibilities
promoting stimuli. For example, as noted above, SD-
for this finding exist. First, we simply did not test enough
exposed Siberian hamsters increase WAT NETO, with the
neurochemicals or receptor types for their presence in these
more internally located WAT pads (RWAT and EWAT)
neurons (see Ref. 98, however, for a long list of antibodies
showing the greatest increases compared with more exter-
tested against neuropeptides and enzymes of synthesis that
nally located WAT pads (IWAT) (26). These SD-induced
did not colocalize with PRV-infected neurons). Second, the
increases in NETO are accompanied by proportional
viral infection inhibits the expression of neurochemicals.
decreases in WAT mass. Fasting does not significantly
Third is a combination of both possibilities. High levels
increase NETO differentially in laboratory rat RWAT or
of colocalization can be found, however (see directly be-
EWAT, although a trend for increased RWAT compared
low), suggesting that it is possible for PRV-infected neurons
with EWAT NETO exists (94). By contrast, in 48 h fasted
also to be labeled for some highly expressed substances,
Siberian hamsters, NETO is significantly increased in
specifically melanocortin 4-receptor (MC4-R) mRNA.
IWAT but not in the other fat pads (M. Brito, N. Brito,
The melanocortins have been heavily implicated in the
C. K. Song, and T. J. Bartness, unpublished data).
control of food intake and energy expenditure, largely
Similarly, although acute cold exposure does not pro-
through MC4-Rs (for review, see Refs. 99, 100). More spe-
duce significant differences in EWAT and RWAT NETO
cifically, MC4-Rs have been implicated in SNS-mediated
Sympathetic and sensory innervation of white fat

WAT lipolysis because central application of the MC3/4-R
and the other is not. Such local surgical denervation offers
agonist, melanotan-II, in laboratory rats decreases food
more neuroanatomical specificity than does global sympa-
intake and body fat (101). These decreases in body fat are
thectomy using guanethidine (104) or 6-hydroxy-dopamine
greater than can be accounted for by the decreases in food
(6OHDA) (105); however, it is not neuroanatomically
intake, as revealed by pair-feeding, suggesting increases
selective, as both sympathetic [there appears to be no or
in energy expenditure (101). We tested whether these in-
sparse parasympathetic nervous system (PSNS) innerva-
creases in lipid mobilization were attributable to stimula-
tion of WAT (72, 106), but see below for a discussion of
tion of MC4-Rs located on neurons that make up the
this issue] and sensory nerves are severed. Indeed, surgical
sympathetic outflow circuits innervating WAT and found
denervation significantly decreases TH-ir (a sympathetic
extensive colocalization of MC4-R mRNA with PRV-labeled
nerve marker) and calcitonin gene-related peptide
SNS outflow neurons across the neural axis, with a high
(CGRP)-ir (a sensory nerve maker), indicating reduced
incidence and percentage (?60% or greater) for most
sympathetic and sensory innervations, respectively (107–
brain areas showing PRV immunoreactivity (-ir) (67).
109). The universal finding of these WAT denervation
These areas include the PVN, preoptic area, bed nucleus
studies is that, regardless of the lipid-mobilizing stimulus,
of the stria terminalis, and amygdala in the forebrain,
lipid mobilization is diminished or most often blocked by
periaqueductal gray in the midbrain, and the nucleus of
surgical denervation compared with neurally intact con-
the solitary tract, lateral paragigantocellular nucleus, lat-
tralateral control pads. For example, fasting-induced
eral reticular area, rostroventrolateral medulla, and ante-
decreases in WAT mass in laboratory rats, cats, rabbits,
rior gigantocellular nucleus in the brainstem, to name a
and dogs (110–114) are blocked, by surgical denervation,
few of the more predominant sites of colocalization. This
as is estradiol-induced decreases in WAT mass of ovari-
extensive high level of colocalization (?60%) suggests that
ectomized rats (115) and the SD-induced decreases in
MC4-Rs play a prominent role in the modulation of SNS
WAT mass of Siberian hamsters (44, 116, 117). The lipid-
Downloaded from
outflow to WAT, either through stimulation by the en-
mobilizing effects of physiological doses of leptin given
dogenous melanocortin agonist a-melanocyte-stimulating
peripherally, however, are not blocked by WAT sympa-
hormone and/or through inhibition by the naturally oc-
thetic denervation in laboratory rats (53).
curring MC3/4-R inverse agonist, agouti-related protein
Complementary to the denervation studies are a few
(102). We have found similar high levels of colocalization
studies of sympathetic nerve stimulation (for review, see
of PRV with MC4-R mRNA sympathetic outflow to BAT
Ref. 11). In an ingenious yet simple study by Correll (118)
(61; C. K. Song, E. Keen-Rhinehart, C. H. Vaughan, D.
conducted almost 45 years ago, the intact nerves inner-
Richard, and T. J. Bartness, unpublished data), and while
vating EWAT were electrically stimulated after the pads
that work was in progress, a similar high level of colo-
were removed with the nerves attached from laboratory
by guest, on October 13, 2010
calization was found after PRV injections into IBAT of
rats and placed into a beaker of medium. Stimulation of
MC4-R-green fluorescent protein transgenic mice (103).
the nerves markedly increased the FFA concentration
These studies demonstrate the usefulness of PRV in de-
in the incubation medium, suggesting lipolysis (118).
lineating the neurochemical phenotype of the sympa-
This effect was blocked if the rats were sympathectomized
thetic outflow neurons to WAT, BAT, or other tissues.
4–11 days before removal and the nerves to the pads were
subsequently electrically stimulated in a similar manner
(118). Moreover, adding dibenamine, a b-adrenergic
blocker, to the incubation medium before the initiation
of electrical stimulation blocked these increases in FFAs
(119). Using the same preparation, others repeated and
extended these findings, showing that another adrenergic
receptor blocking agent [1-(2¶,4¶-dichlorophenyl)-1-hydroxyl-
The notion that adrenal medullary EPI is the principal
2-(t-butylamino)], a depletor of catecholamine stores
initiator of lipolysis was undermined because of the in-
(syrosingopine), and an inhibitor of NE release (BW
ability of adrenal demedullation (and thus no circulating
392C60) all blocked the electrical stimulation-induced
EPI) to block lipid mobilization triggered by several stim-
increase in FFA concentration in the incubation medium
uli that promote lipolysis (see above). Analogous experi-
(119). This electrical stimulation-triggered lipolysis is ex-
ments have been conducted to test whether destruction of
aggerated by pargyline, a monoamine oxidase inhibitor
the SNS innervation of WAT blocks lipid mobilization.
(thus inhibiting NE degradation), and by theophylline,
Because most of the WAT pads studied by researchers are
the phosphodiesterase inhibitor (119), the latter suggest-
bilaterally located and unilaterally innervated (for review,
ing that endogenously released catecholamines are
see Ref. 9), the “unilateral denervation model” can be
involved (119). In a relatively analogous in vivo prepara-
exploited. Specifically, in this model, one of a pair of WAT
tion in humans, Dodt et al. (120–122) intraneurally stimu-
pads is denervated, with its contralateral mate serving as a
lated the lateral femoral cutaneous nerve that innervates
within-animal neurally intact control that receives sham
subcutaneous WAT and measured local lipolysis (changes
denervation. Therefore, all other characteristics of the
in glycerol concentrations) via microdialysis of the
animal are the same: genetics, age, energy balance, and all
interstitial subcutaneous WAT area receiving the stimula-
circulating factors except that one fat pad is denervated
tion. This stimulation increases interstitial glycerol con-
Journal of Lipid Research
Volume 48, 2007

centrations in vivo in humans (120, 121). Thus, comple-
gins. Triacylglycerols in adipocytes are hydrolyzed in three
mentary electrical stimulation studies to the denervation
steps, with HSL catalyzing triacylglycerol and diacylglyc-
studies discussed above also support a role of sympathetic
erol and monoglyceride lipase completing the process
innervation in initiating lipolysis in WAT.
(for review, see Ref. 149). Perilipins interact with HSL
and are positioned on the surface of lipid droplets within
the adipocyte to block the translocation of HSL and thus
inhibit catecholamine-induced lipolysis (for reviews, see
Ref. 150, 151). With protein kinase A phosphorylation
of perilipin and HSL, however, the phosphorylated HSL
now has access to the lipid droplet and lipolysis pro-
It is beyond the scope of this review to provide a detailed
ceeds (for reviews, see Ref. 126, 148). In addition, adipo-
account of the current state of knowledge of the in-
cyte lipid binding protein (or adipocyte fatty acid binding
volvement of adrenoceptors in lipolytic responses or of
protein, A-FABP, 422 protein, aP2, and p15 protein) has
endocrine/paracrine lipolytic factors; the reader is re-
been isolated, purified, and cloned in humans and
ferred to several excellent reviews (123, 124). Although
rodents (152–155) and appears to assist in lipolysis. Adipo-
we will focus here on adrenergic effects on lipolysis, this
cyte lipid binding protein accepts FFAs and activates
is not to deny the importance of these endocrine/
HSL as well. Because adipocyte lipid binding protein
paracrine factors, such as natriuretic peptides [humans/
removes FFAs from the cytosol, the well-known end prod-
primates only (125)], tumor necrosis factor-a, glucagon,
uct inhibition of lipolysis that occurs with accumulating
and growth hormone (for reviews, see Refs. 123, 126). In
FFAs seen in vitro (156) is negated in vivo, allowing
addition to the adrenoceptor subtype a2-adrenoceptor
lipolysis to proceed.
(see directly below), other nonadrenergic factors that in-
The preponderance of the data suggests that the bal-
Downloaded from
hibit lipolysis also are important, such as insulin, the
ance between the lipolysis-promoting b-adrenoceptor
most potent of these (127, 128), as well as prostaglandin
activation and lipolysis-inhibiting a2-adrenoceptor activa-
E2, adrenomedullin, adenosine (for review, see Ref. 123),
tion dictates the degree of lipolytic activity, all other fac-
and neuropeptide Y (129–132), the latter a neuropeptide
tors being equal (for reviews, see Refs. 8, 145, 157). Thus,
that is often colocalized with NE in sympathetic nerves
when b-adrenoceptor activation predominates, lipolysis is
stimulated; conversely, when a2-adrenoceptor activation
Pioneering work of Lafontan (136–139) and others
predominates, lipolysis is inhibited (for reviews, see Refs.
(140–142) set the stage for our current knowledge of the
143–145). The affinity of the naturally occurring agonists
role of adrenoceptors (adrenergic receptors) in lipolytic
(EPI and NE) for these adrenoceptors is as follows, based
by guest, on October 13, 2010
and other cell functions. In its simplest form, four subtypes
on in vitro adipocyte membrane binding assays (145): for
of adrenoceptors have a role in catecholamine-stimulated
NE, a2 . b1 > b2 . b3; for EPI, a2 . b2 . b1 . b3. The
lipolysis; these are the b1-, b2-, and b3-adrenoceptor sub-
presence of these receptors or their levels varies consid-
types as well as the a2-adrenoceptor (for reviews, see
erably across species, with humans, laboratory rats, and
Refs. 124, 143–146). Activation of the three b-receptor
hamsters having ample a2-adrenoceptors and laboratory
subtypes stimulates lipolysis, but the participation of each
mice having none (132).
subtype in lipolysis varies according to the fat pad, species,
It has been a physiological conundrum to understand
gender, age, and degree of obesity (147). By contrast, as
how adipocytes can have receptors that stimulate lipolysis
noted above, activation of a2-adrenoceptors inhibits lipoly-
and those that inhibit lipolysis yet share the same agonists.
sis (for review, see Ref. 144). These four adrenoceptors
It has been suggested that because of the higher affinity of
coexist in adipocytes. Because activation of the b1-, b2-, and
the a2-adrenoceptors compared with the b-adrenoceptors
b3-adrenoceptors stimulates lipolysis, whereas activation of
for low naturally occurring (basal) concentrations of NE
the a2-adrenoceptor inhibits lipolysis, it is not surprising
or EPI, a2-adrenoceptors are activated, thereby inhibiting
that postreceptor events are different for each. Stimula-
lipolysis (148). This inhibition can be broken, however,
tion of the GTP binding protein Gs with b1-, b2-, and
with increases in concentrations of the agonists, especially
b3-adrenoceptor agonism triggers increases in cAMP
NE released from the postganglionic sympathetic nerve
production by adenylyl cyclase that, in turn, stimulates
terminals that impinge close to the adipocytes (148).
protein kinase A, which, among other things, phosphoryl-
Further evidence for the interplay between these receptors
ates hormone-sensitive lipase (HSL) and perilipin (for
and a role of the a2-adrenoceptor in obesity was shown in a
review, see Refs. 123, 126, 148; see below for more on
clever genetic experiment in which b3-adrenoceptor
perilipin). By contrast, stimulation of the GTP binding
knockout mice also had human a2-adrenoceptors ex-
protein Gi with a2-adrenoceptor agonism triggers de-
pressed transgenically in their WAT (158). These animals
creases in cAMP as a result of the inhibition of adenylyl
became obese, but only with both high-fat diet feeding and
cyclase and thus the lack of stimulation of protein kinase A
the presence of the a2-adrenoceptors (158).
and, therefore, the lack of phosphorylation of HSL and
The challenge for a pharmacological approach to obe-
perilipin (for reviews, see Refs. 123, 126, 148).
sity reversal through the stimulation of b3-adrenoceptors
Once this stimulation of b-adrenoceptors has begun,
has been to find a WAT-specific b3-adrenoceptor agonist
the process of hydrolyzing the stored triacylglycerol be-
(despite the low numbers of this receptor subtype in
Sympathetic and sensory innervation of white fat

human WAT) that has a long-term, nonadapting stimula-
cantly decreased CGRP-ir but not TH-ir and did not alter
tion of lipolysis without negative side effects, such as hy-
FCN (107). Collectively, these results point to the sympa-
pertension or stroke (for reviews, see Refs. 159, 160).
thetic innervation of WAT as inhibiting FCN.
Although FCN is increased by destruction of the sym-
pathetic innervation of WAT, this does not necessarily
translate into a denervation-induced increase in fat cell
proliferation. We measure cellularity using a modification
of the osmium tetroxide fixation method combined with
To this point, we have focused on the sympathetic
Coulter Counter quantification of fat cells (168). With this
control of lipid mobilization through the innervation of
method, adipocytes having diameters of ,20 mm are
WAT. A lesser known and recognized function of the
literally screened out to eliminate lipid droplets and cell
sympathetic innervation of WAT is the control of adi-
debris. Thus, denervation could cause an apparent
pocyte proliferation. Despite hypercellularity being a hall-
increase in FCN through increased accumulation of lipid
mark of obesity (for review, see Ref. 161), relatively little
in small adipocytes by the elimination of basal lipolysis,
research on fat cell proliferation has and is occurring rela-
thereby allowing them to be counted. To test for bona fide
tive to research on fat cell differentiation (for review, see
proliferation by sympathetic denervation, we recently
Ref. 162), making the control of fat cell number (FCN)
(109) injected 6OHDA or capsaicin locally in WAT to
a clinically and physiologically significant unknown. Its
selectively destroy the sympathetic innervation and spare
importance in obesity is obvious, as increases in adiposity
the sensory innervation or to selectively destroy the
would be quite restrained, instead of being apparently
sensory innervation and spare the sympathetic inner-
limitless, if existing adipocytes merely filled to capacity
vation, respectively. Surgical denervation was included as
with lipid. Although there are a host of circulating and
a positive control. To label proliferating adipocytes, we
Downloaded from
paracrine factors that affect adipocyte proliferation (for
also injected bromodeoxyuridine (BrDU; a nonradioactive
review, see Ref. 162), there also is growing support for SNS
method of identifying dividing cells), and to determine
innervation of WAT playing a highly significant part in this
whether the BrDU-labeled cells were adipocytes, we dou-
process. To our knowledge, the first unambiguous demon-
ble labeled the tissue for AD3-ir, a white adipocyte-specific
stration of the likely role of the sympathetics in fat pro-
membrane protein (169, 170). Surgical denervation sig-
liferation was the inhibition of preadipocyte proliferation
nificantly decreased TH-ir and CGRP-ir, whereas 6OHDA
in vitro by physiological concentrations of NE (163). This
treatment only significantly decreased TH-ir, but both
effect is blocked by pretreatment of the preadipocytes
treatments were associated with ?3- to 4-fold increases in
with the b-adrenoceptor antagonist propranolol (163) and
BrDU1AD3-ir cells. By contrast, capsaicin treatment sig-
by guest, on October 13, 2010
was possible because adipocyte precursor cells possess
nificantly decreased CGRP-ir but not TH-ir and was not
b-adrenoceptors (164). Therefore, if NE inhibits fat cell
associated with an increase in BrDU1AD3-ir (109). These
proliferation in vitro, then sympathetic denervation of
results suggest that destruction of the sympathetic in-
WAT in vivo should stimulate fat cell proliferation. Indeed,
nervation of WAT stimulates preadipocyte proliferation
surgical denervation significantly increases FCN in labo-
(109). Collectively, it appears that increases in the sym-
ratory rat WAT (165).
pathetic stimulation of b-adrenoceptors inhibits fat cell
We have since replicated this surgical denervation-
proliferation, whereas decreases in sympathetic stimulation
induced increase in FCN several times in Siberian hamster
of b-adrenoceptors promotes it.
WAT using the unilateral denervation model (63, 107,
Interestingly, the magnitude of the SNS denervation-
109). A consistent finding of these studies is a denervation-
induced increases in FCN (proliferation) appears to re-
induced increase in FCN but not FCS (63, 107). To more
flect the particular propensity of each WAT depot to grow
precisely determine whether it was the surgical denerva-
by increasing FCN. Specifically, surgically denervated
tion of WAT via the destruction of its sympathetic in-
Siberian hamster IWAT achieves a greater percentage
nervation or of its sensory innervation (see below), we
increase in FCN than similarly denervated RWAT (63),
surgically denervated WAT (positive control) or locally
whereas EWAT surgical denervation does not affect
injected 6OHDA to kill only the sympathetic nerves and
FCN (107). This lack of denervation-induced increases
spare the sensory nerves or locally injected capsaicin to kill
in FCN by EWAT is completely consistent with the ten-
the sensory nerves and spare the sympathetics (107).
dency of EWAT to increase its mass by increasing FCS, and
Capsaicin is the pungent part of red chili peppers and also
the denervation-induced increases in FCN by RWAT and
is a selective neurotoxin for unmyelinated sensory nerves
IWAT reflect the tendency of these pads to increase their
(166, 167). We verified the destruction of the sympathetics
masses primarily by increasing FCN, at least in the obesity
by 6OHDA using immunohistochemistry for TH and the
associated with aging in Wistar rats (171). These effects of
destruction of sensory innervation by capsaicin using im-
denervation also are consistent with the greater in vivo
munohistochemistry for CGRP (107). Surgical denervation
incorporation of radiolabeled thymidine into DNA in
of WAT significantly decreased both TH-ir and CGRP-ir,
RWAT (a radioactive method of identifying dividing cells)
as expected, and 6OHDA treatment significantly de-
than in EWAT from high-fat diet-fed laboratory rats (172).
creased TH-ir but not CGRP-ir, and both triggered an
Therefore, it is easy to envision that a decrease in sym-
?3-fold increase in FCN. By contrast, capsaicin signifi-
pathetic drive to some of these WAT pads could promote
Journal of Lipid Research
Volume 48, 2007

fat cell proliferation and thereby make significant con-
theless, PRV was then injected into the presumed sympa-
tributions to increases in adiposity.
thetically denervated WAT pad to label the origins of the
The underlying mechanisms triggering the sympathetic
hypothesized remaining parasympathetic nerves. This re-
denervation-induced increase in fat cell proliferation are
sulted in extensive bilateral infection of the dorsal motor
unknown beyond the decreases in the activation of adipo-
nucleus of the vagus, thereby laying the claim of exten-
cyte b-adrenoceptors (163). Mature adipocytes and their
sive WAT PSNS innervation (176). Unfortunately, these
precursor cells have a2-adrenoceptors, the function of
remaining nerves were not labeled with any known neuro-
which in mature fat cells is to inhibit lipolysis when stimu-
chemical markers of the PSNS, nor were the parasympa-
lated (see above). The a2-adrenoceptors, in turn, have
thetic ganglia that always accompany PSNS innervation
been implicated in adipocyte proliferation (164, 173),
of tissues/glands identified. Furthermore, the extensive
because the surgical denervation of laboratory rat WAT is
bilateral dorsal motor nucleus of the vagus infection is
associated with increases in FCN that are preceded by
questioned, as this area innervates most peripheral tissues
increases in adipocyte a2-adrenoceptor number (165).
unilaterally in rodents (179–182).
Stimulation of these receptors in denervated WAT could
For these and other reasons (72, 106, 177, 178), we
occur by any remaining sympathetic nerves or by adrenal
sought PSNS nerve markers in WAT (72). We used three
medullary catecholamines (primarily EPI). Either of these
types of animals, a standard mouse strain (C57BL mice), a
alternatives seems possible because a2-adrenoceptors
genetically obese mouse strain (ob/ob mice), and a stan-
show special affinity for low concentrations of EPI and to
dard laboratory rat strain (Sprague-Dawley rats), exam-
a lesser degree NE (see above). In neurally intact WAT,
ined three WAT pads (IWAT, RWAT, and EWAT), and
decreases in the sympathetic drive to WAT and conse-
tested for three previously proven neurochemical markers
quent stimulation of fat cell proliferation seem to be
of PSNS innervation in other tissues (vesicular acetylcho-
associated with decreases in b-adrenoceptor number and
line transporter, vasoactive intestinal peptide, and neuro-
Downloaded from
increases in a2-adrenoceptor number as the WAT pad
nal nitric oxide synthase). There was only one result: no
mass expands (174). One way that a2-adrenoceptor stimu-
labeling in any animal in any WAT pad for any established
lation might trigger fat cell proliferation is through the
PSNS marker (72). Next, we selectively and locally sym-
local release of lysophosphatidic acid that is elicited by
pathetically denervated WAT with 6OHDA, as shown by
WAT a2-adrenoceptor activation (173). Because lysophos-
significantly decreased WAT NE content and TH-ir, with
phatidic acid added to preadipose cell lines triggers fat cell
no effect on sensory innervation (CGRP-ir), thereby spar-
proliferation (173), one could imagine such an event oc-
ing the putative PSNS innervation (72). PRV was then
curring in vivo as well. This hypothesized role for a2-
injected into the 6OHDA- or vehicle-injected WAT several
adrenoceptors in fat cell proliferation is bolstered by the
days later. PRV did not infect the sympathetic chain, spinal
by guest, on October 13, 2010
genetic addition of these receptors in b3-adrenoceptor
cord, or brain in any animal with 6OHDA-treated WAT,
knockout mice fed a high-fat diet, in which their obesity is
but vehicle-injected WAT inoculated with PRV had typical
solely attributable to increases in FCN (158). Finally,
sympathetic chain, spinal cord, and brain viral infection
decreases in sympathetic drive may stimulate the release of
patterns (72). These data showing a lack of significant or
the recently discovered white adipocyte paracrine factor
no WAT PSNS innervation are supported by older bio-
autotoxin, a type II ectonucleotide pyrophosphatase phos-
chemical data showing no acetylcholinesterase, an enzyme
phodiesterase, that in turn could stimulate proliferation
important in the degradation of acetylcholine (183). Col-
by its ability to release lysophosphatidic acid (175). Col-
lectively, these findings suggest that PSNS innervation
lectively, this scenario provides a possible mechanism by
either does not exist or is relatively minor in its extent. If,
which fat cell proliferation is stimulated with decreases or
however, convincing data supporting PSNS innervation
abolition of its sympathetic drive.
emerge in the future, this would be interesting, as it would
afford WAT the fine control of metabolism that other
tissues/glands have with dual autonomic nervous sys-
tem innervation.
Most but not all (e.g., peripheral blood vessels and sweat
glands) tissues receive dual innervation by the autonomic
nervous system. There is an initial report of WAT PSNS
innervation (176), and a thorough discussion of those find-
ings and the possibility of PSNS innervation of WAT were
presented recently (72, 106, 177, 178). In that initial re-
Because sensory innervation of tissues is the rule, not
port (176), selective surgical denervation of the sympa-
the exception, it should not be surprising that WAT has
thetic innervation of WAT was thought to be achieved,
significant sensory innervation of WAT (for reviews, see
thereby sparing the PSNS innervation. Although it is clear
Refs. 11, 64). The first study showing direct evidence of
that separate neural provisions to WAT were indeed either
sensory innervation (i.e., tract tracing) was from Fishman
severed or spared based on the researchers’ intentions
and Dark (184); the anterograde tract tracer, True Blue,
(54), irrefutable evidence of their respective associations
was applied to laboratory rat IWAT or DWAT, resulting in
with the SNS or PSNS remains to be demonstrated. Never-
labeled neurons in the dorsal root ganglia (DRG), the
Sympathetic and sensory innervation of white fat

home of pseudounipolar (“bipolar”) sensory neurons.
unpublished data). Others reported that injection of
These data are convincingly corroborated by histology
cholera toxin B, a retrograde tracer, into retroperitoneal
performed at the level of the WAT pad, showing substance
WAT (RWAT) labeled approximately four to seven cells in
P-ir (185) and CGRP-ir in laboratory rat WAT (186). Sub-
the nucleus gracilis of the brainstem but found no labeling
stance P and CGRP are contained within, and released
in the dorsal horn of the spinal cord (they did not look at
from, sensory neurons and thus are considered markers of
the DRG); thus, it appears that they may have labeled
sensory innervation (187). We extended these findings to
visceral sensory inputs but not spinal sensory nerves,
Siberian hamsters, in which CGRP-ir occurred in IWAT
although the design of the study was not strictly to define
and EWAT (107–109). Moreover, coculture of DRG cells
the origins of the sensory innervation of WAT.
with adipocytes from the 3T3-L1 cell line results in the
The functional role of WAT sensory innervation is still
stimulation of neurite outgrowth from DRG cells and is
largely unknown. Although leptin, the largely adipose-
associated with increases in angiopoietin-1 and trkA, the
derived cytokine thought to convey adiposity information
former, a ligand for the endothelial cell-specific Tie2 re-
to the brain (191), is thought to exert many of its effects on
ceptor and the latter, a high-affinity receptor for nerve
energy balance by penetrating the brain and binding to
growth factor (188). Although electron microscopy did
receptors there (for reviews, see Refs. 192, 193), another
not indicate sensory nerve-adipocyte synapses, en passant-
periphery to brain conduit exists via sensory nerves. First,
type sensory nerve-adipocyte interactions were observed
leptin receptors have been localized in nodose ganglion
(188). Collectively, these in vitro data further support the
neurons in laboratory rats and humans (194), cell bodies
sensory innervation of WAT.
of vagal afferents, and on the vagal trunk itself in labora-
Although the pioneering study of Fishman and Dark
tory rats (195). In addition, single vagal afferent activities
(184) showed dorsal root ganglion pseudounipolar sen-
from intestinal mechanoreceptors are electrophysio-
sory neurons innervating WAT using traditional tract-
logically responsive to leptin in cats (196, 197). Moreover,
Downloaded from
tracing methodology, the reception sites in the brain for
injections of leptin into one laboratory rat EWAT pad
this sensory innervation are not labeled by this technique.
triggers a dose-dependent increase in its sensory nerve af-
We addressed this issue using a transneuronal viral tract
ferent activity (83, 87) and elicits increases in sympathetic
tracer, the H129 strain of herpes simplex virus-1 [(HSV)
nerve activity to the contralateral EWAT pad, suggestive of
generously donated by Dr. Richard Dix, Georgia State
a reflex arc (87). Given that peripheral and central leptin
University], borrowing the approach that Rinaman and
can increase the sympathetic drive to WAT, albeit non-
Schwartz (189) used to define the sensory inputs to the
uniformly (96), and that leptin can decrease adiposity
brain from the stomach. Unlike PRV, H129 initially seems
independent of food intake (198, 199), the electrophys-
to travel in a retrograde direction from the site of in-
iological data from WAT sensory nerves after local leptin
by guest, on October 13, 2010
oculation, in our case from IWAT, to the soma of the
injection suggest receip