Energy balance results from the coordination of multiple pathways affecting energy expenditure and food intake. Candidate neuropeptides involved in energy balance are the melanocortins. Several species, including Siberian hamsters studied here, decrease and increase food intake in response to stimulation and blockade of the melanocortin 4-receptor (MC4-R). In addition, central application of the MC3/4-R agonist melanotan-II decreases body fat (increases lipolysis) beyond that accounted for by its ability to decrease food intake. Because an increase in the sympathetic nervous system drive to white adipose tissue (WAT) is the principal initiator of lipolysis, we tested whether the sympathetic outflow circuitry from brain to WAT contained MC4-R mRNA expressing cells. This was accomplished by labeling the sympathetic outflow to inguinal WAT using the pseudorabies virus (PRV), a transneuronal retrograde viral tract tracer, and then processing the brain for colocalization of PRV immunoreactivity with MC4-R mRNA, the latter assessed by in situ hybridization. MC4-R mRNA was impressively colocalized in PRV-labeled cells (approximately greater than 60%) in many brain areas across the neuroaxis, including those typically implicated in lipid mobilization (e.g., hypothalamic paraventricular, suprachiasmatic, arcuate and dorsomedial nuclei, lateral hypothalamic area), as well as those not traditionally identified with lipolysis (e.g., preoptic area, subzona incerta of the lateral hypothalamus, periaqueductal gray, solitary nucleus). These data provide compelling neuroanatomical evidence that could underlie a direct central modulation of the sympathetic outflow to WAT by the melanocortins through the MC4-Rs resulting in changes in lipid mobilization and adiposity.
- pseudorabies virus
- Siberian hamster
the ability to mobilize lipid energy stores from white adipose tissue (WAT) is critical to provide energy when the incessant energy demands of the central nervous system (CNS) and peripheral tissues cannot be met. The search for a neurochemical means to trigger WAT lipid mobilization in obese humans is one of the two pharmacological holy grails for obesity treatment, the other being an effective, long-lasting appetite suppressant. One potential target for the stimulation of lipid mobilization from WAT is via the sympathetic nervous system (SNS) innervation of this tissue (for a review, see Refs. 5 and 6). Convincing evidence of the SNS innervation of WAT has arisen from studies using traditional tract tracers through which the postganglionic innervation of WAT was demonstrated bidirectionally (53). Moreover, the CNS origins of the SNS outflow from brain to and through these postganglionic neurons terminating in WAT have been defined using a transneuronal, retrograde viral tract tracer, the pseudorabies virus (PRV; Refs. 2, 10, 42, 44). Because the initial infection by PRV allows neurons to continue to produce their peptides, enzymes of synthesis, and receptors, albeit likely dampened (for a review, see Ref. 45), this affords the opportunity to neurochemically phenotype some of these neurons. For example, PRV labeling of the WAT SNS outflow neurons can be combined with in situ hybridization for the mRNA of candidate receptors. Thus we identified the functional melatonin receptor for photoperiodic responses, the MEL1a receptor, in Siberian hamsters (Phodopus sungorus) on sympathetic outflow neurons to WAT (44). Colocalization of PRV with neurotransmitter receptors is highly suggestive of a biological role for that neurotransmitter in the labeled circuit and provides targets for subsequent functional tests in which microinjections of receptor agonists or antagonists can be made to turn the SNS outflow to WAT on or off, thereby affecting lipolysis and thus ultimately adiposity. Candidates for modulation of the SNS drive to WAT are the melanocortin receptors (for a review, see Ref. 14). The melanocortin-3 and -4 receptors (MC3-R and MC4-R, respectively) appear predominantly in brain and spinal cord (26, 32), but MC4-Rs are also found in peripheral tissues such as the penis and pelvic ganglion (49). Furthermore, although early studies indicated only MC2- and MC5-Rs on white adipocytes (9), more recently functional and molecular evidence points to the presence of the MC4-R as well (22). The endogenous agonist for these receptors is α-melanocyte stimulating hormone (α-MSH), a cleaved product of the polypeptide precursor proopiomelanocortin (POMC) processed in the arcuate nucleus (Arc) of the hypothalamus and the nucleus of the solitary tract (sol) in the brainstem (11, 36). In addition to the agonist, a natural antagonist also exists: agouti-related protein (33). The role of the melanocortins and its two central receptor subtypes in the control of food intake seems undeniable (for a review, see: Ref. 50), including in Siberian hamsters (15, 40, 41), with fewer data supporting its role in energy expenditure (e.g., 51, 52).
Much less is known about the effects of the melanocortins on lipid mobilization or accretion in WAT that is independent of their effects on food intake or thermogenesis. Although stimulation of MC4-Rs expressed in differentiated cultured adipocytes modulate leptin gene expression and peptide secretion in vitro (22), suggesting a direct peripheral effect of melanocortins on fat, there also are strong implications for the ability of central melanocortins to affect peripheral WAT depots. For example, chronic central administration of melanotan II (MTII), a synthetic analog of α-MSH (20), decreases WAT pad mass by ∼50% compared with pair-fed rats, suggesting that the mobilization of lipid is not accounted for by the decreases in food intake alone (39). Similar effects are seen when MTII is administered peripherally (e.g., Ref. 13). Finally, centrally administered MTII decreases the respiratory quotient, suggesting the oxidation of lipid-derived fuels (23). Collectively, these data point toward a possible role of stimulation of central MC3/4-Rs in lipid mobilization from WAT. Given the growing evidence of the involvement of the SNS innervation of WAT in lipolysis, rather than adrenal medullary-released catecholamines (for reviews, see Refs. 5 and 7), we hypothesize that centrally activated MC3/4-Rs, and especially the MC4-Rs that have already been implicated in increasing SNS nerve activity when stimulated centrally (21, 38), may increase the activity of the SNS outflow circuits to WAT and thereby promote lipid mobilization. For this to be a physiological reality, however, the SNS outflow neurons to WAT should possess MC4-Rs. Therefore, we asked: Do the SNS outflow neurons from brain to WAT possess MC4-Rs? This was accomplished by labeling the SNS outflow to WAT using the PRV transneuronal tract tracing method for defining circuits within the same animal (2, 10, 42, 44), followed by processing the brain sections for combined immunohistochemistry and in situ hybridization for PRV and MC4-R mRNA, respectively.
All procedures were approved by the Georgia State University Institutional Animal Care and Use Committee and are in accordance with Public Health Service and the U.S. Department of Agriculture guidelines. Adult male Siberian hamsters ∼3.5 mo old (n = 4) were obtained from our breeding colony, the lineage of which has been described recently (10). Hamsters were group-housed and exposed to a long-day photoperiod (16:8-h light-dark cycle; lights on at 0200) from birth. Room temperature was maintained at 21 ± 2°C. Hamsters were fed PMI Rodent Diet no. 5001 (Purina, St. Louis, MO) throughout the experiment. Hamsters were single housed 1 wk before PRV injections.
Animals were anesthetized with pentobarbital sodium (50 mg/kg), and the target incision area over the rear haunch was shaved and wiped with 50% ethanol. An incision was made at the dorsal hind limb of the animal and lateral to the spinal column that continued rostrally and then ventrally to the ventral hind limb. Care was taken with the depth of the incision so as to not damage the underlying fat pad and vasculature. The application of the PRV was performed essentially as described previously (2). Briefly, once the IWAT pad was exposed, a series of injections of PRV 152 (generous gift of Lynn Enquist, Princeton, NJ) was made using a 1.0-μl Hamilton syringe at five loci within the fat pad (1.5 × 108 pfu/ml; 150 nl/loci) to evenly distribute the virus. The incision was closed with sterile wound clips, and nitrofurazone powder was applied to minimize infection. The animals were then transferred to clean cages and survived for 6 days, the optimal postinoculation survival time for infection to reach the rostral forebrain from this fat pad (2).
Six days postinjection, the animals were perfused transcardially with heparinized (0.02%) saline and phosphate-buffered (0.1 M; pH 7.4) paraformaldehyde (4% wt/vol). The brains and spinal cords were extracted and postfixed in the same fixative overnight at 4°C and sunk in sucrose (30% wt vol; with 0.1% sodium azide). The brains were sliced at 25 μm using a cryostat, and the sections were immersed in a cryoprotectant at −20°C until further processed. The spinal cords were embedded in gelatin (20%) before slicing on a freezing stage sliding microtome at 50 μm in the horizontal plane. The spinal cord sections were then mounted on to gelatin-coated slides and coverslipped with Vectashield (Vector Laboratories, Burlingame, CA). The thoracic and lumbar spinal cord regions were scanned for the presence of lysed cells or bilateral infection, both signs of nonspecific infections, by examining the sections for the PRV152-induced enhanced green fluorescent protein reporter using an Olympus BX41 microscope with a green fluorochrome filter.
Cloning of MC4-R cDNA.
MC4-R cDNA was cloned into pCR II vector. In brief, total RNA was isolated from midbrain of laboratory rat (Rattus norvegicus) using Trizol (Invitrogen, Carlsbad, CA) and reverse-transcribed using primers (5′-TGAGACATGAAGCACACGCAG-3′) based on the published MC4-R cDNA sequence (U 67863.1). Aliquots of the first-strand cDNA were used as templates for specific amplification of MC4-R partial cDNA (forward primer, 5′-ACTCGGACAGCAGCGCTGTC-3′, and reverse primer, 5′-TGAGACATGAAGCACACGCAG-3′). The PCR reaction contained 1× RT buffer, 2 mM MgCl2, 10 μl of first-round RT reaction, 100 μM dNTPs, 500 nM primers, 2.5 U of Taq polymerase in 100 μl total volume. The MC4-R cDNA was amplified for 35 cycles using the following conditions: 94°C for 40 s, 60°C for 1 min, and 72°C for 1 min. An aliquot of the PCR product was subjected to agarose gel electrophoresis and revealed the expected 289 bp product for MC4-R, corresponding to nucleotides 707–995 of the published sequence of rat MC4-R cDNA (U 67863.1). The cDNA was cloned into pCR II (Invitrogen) which contains SP6 and T7 ribonuclease polymerase promoters flanking the cloning site in the forward direction with respect to the lac gene. Specificity of the probe was tested in a ribonuclease protection assay (RPA). Briefly, the c-DNA-plasmid was linearized with XhoI and used as templates for run-off cRNA transcripts with SP6 polymerase in the presence of biotin-UTP. The RPA was carried out according to manufacturer's directions using an RPA II Ribonuclease Protection Assay Kit (Ambion). For each sample, 30 μg total RNA extracted from rat hypothalamus was hybridized overnight with 600 pg probe at 42°C. Single-strand RNA was digested with ribonuclease (RNase) A/RNase T1 at 37°C for 30 min. The hybrids were precipitated, resuspended, and then electrophoresed on a denaturing polyacrylamide gel. The fragments were transferred onto a nylon membrane (Hybond N, Amersham, Piscataway, NJ) and cross-linked with UV exposure. The hybrids were then detected using chemiluminescence (BrightStar BioDetect Nonisotopic Detection, Ambion) and exposed to X-ray film. β-actin was included as an internal control to verify that equal amounts of RNA were used per sample. The specificity of the MC4-R probe was demonstrated by the presence of a single hybrid of expected size (Fig. 1).
Combined in situ hybridization/immunohistochemistry.
Every third brain section was processed for combined in situ hybridization and immunohistochemistry for MC4-R mRNA and PRV-ir, respectively.
To generate sense and antisense probes, the plasmids containing the 289 bp MC4-R cDNA were linearized with HindIII and XhoI restriction enzymes (Promega, Madison, WI) and transcribed in vitro using T7 and SP6 RNA polymerases (Stratagene, La Jolla, CA ), respectively, in the presence of 35S-UTP (PerkinElmer, Boston, MA).
In situ hybridization for MC4-R.
Brain sections were rinsed in 2× SSC (1× = 0.15 M sodium chloride and 0.015 M sodium citrate), dehydrated in ascending concentrations of ethanol, delipidated with chloroform, rehydrated in descending concentrations of ethanol, deaminated in acetic anhydride (0.25%)-triethanolamine (0.1 M in saline) and then incubated in formamide (50%; with 2× SSC, 0.1% sodium dodecyl sulfate, 25 mM dithiothreitol). The sections were then hybridized with 35S-labeled MC4-R cRNA or sense control probes (30,000 cpm/μl) at 57°C overnight. The following day, the sections were cooled to room temperature, rinsed in 1× SSC, 2× SSC-formamide (1:1, 30 min at 53°C), incubated in RNase A (100 mg/ml), washed in 2× SSC-formamide (1:1, 90 min at 57°C), rinsed in 1× SSC and in PBS (0.1M; pH 7.4) for immunohistochemical processing to detect PRV in the same sections.
Immunohistochemistry for PRV.
Sections were incubated sequentially in the primary antibody (Rb132, 1:10,000; a generous gift of Lynn Enquist, Princeton, NJ) overnight, secondary antibody (goat anti-rabbit; 1:500; Vector Laboratories) for 2 h and in avidin-biotin complex (1.75 μl/ml of each avidin and of biotin) for 1 h. The specific labels were detected using diaminobenzidine (0.2 mg/ml; Sigma, St. Louis, MO) as the chromogen in the presence of peroxide (0.0025%). All steps in the immunohistochemistry procedure were performed at 22°C. The sections were mounted onto gelatin-coated slides, rinsed in deionized water and ethanol and then air-dried. Isotopic labels were detected autoradiographically by dipping sections into NTB3 liquid emulsion (Eastman Kodak, Rochester, NY). Slides were exposed for 5 days in 4°C, developed, and fixed in D-19 and Fixer (Eastman Kodak), respectively, counterstained with cresyl violet and coverslipped.
Data analysis and imaging.
All brain sections were analyzed (Olympus BX41 microscope) for single- and double-labeling in all nuclei where PRV infection was found from the level of the preoptic area (POA) rostrally and through the brainstem caudally. PRV-infected neurons were considered double-labeled if the granular deposition above the cells was at least 7 times that of background levels in adjacent areas within the same sections and conformed to the shape of the cells. Images were captured digitally with an Olympus DP70 camera and acquired using Adobe Photoshop (v6.0, San Jose, CA). A mouse brain atlas (37) was used to identify brain regions, as no atlas exists for Siberian hamsters. The mouse atlas is anatomically closer to the brains of Siberian hamsters than a Syrian hamster atlas in our opinion. PRV-infected neurons and the subpopulation of PRV cells that also were labeled for MC4-R mRNA were counted. Data were collapsed across each bilateral nucleus/region within each animal and then averaged across the animals.
None of the animals exhibited overt signs of illness for the 6 days postinoculation period. No cases of lysis or bilateral infection were observed in the spinal cord sections; therefore, tissues from all four animals were processed and analyzed. The CNS infection pattern seen in this study using PRV 152 injected into IWAT is virtually identical with that found previously using its parental Bartha's PRV strain (2) injected into the same tissue. We present the infection data in greater detail in this paper (Table 1) than in our previous report (2). We analyzed only those brain regions that had both PRV infections and that also had at least 20 colocalized cells.
Unilateral PRV infections ipsilateral to the inoculation site were found in the thoracic and lumbar portions of the spinal cord (Fig. 2). In the brain stem, PRV infection was found bilaterally, most notably in the sol, area postrema (AP), lateral paragigantocellular nucleus, raphe regions (raphe pallidus, raphe obscurus, raphe magnus), and reticular regions, especially the parvicellular lateral reticular nucleus and rostroventrolateral reticular nucleus/adrenaline cells (RVL/C1). Midbrain PRV infections were most prominent in the periaqueductal gray (PAG), mostly in the lateral PAG (LPAG) and in the pedunculopontine tegmental nuclei.
Heavy PRV infections also were found in the forebrain, including the hypothalamic paraventricular nucleus (PVN), dorsomedial nucleus (DM), Arc, medial preoptic areas (medial/central/lateral) and lateral hypothalamus (LH).
MC4-R expression in PRV-infected neurons.
As expected, hybridization with sense control probes caused no specific autoradiographic signals (data not shown).
In general, the distribution of MC4-R mRNA was very similar to the pattern of PRV infection (Fig. 3). Moreover, there were PRV-labeled cells that also were labeled for MC4-R mRNA (MC4-R + PRV) in many regions along the neuroaxis (Table 1). The ratio of colocalized cells, [(MC4-R+PRV)÷PRV]%, ranged from 29.3 ± 14.8% to 87.5 ± 4.4%, in the locus coeruleus (LC) and linear nucleus (Li), respectively (Table 1).
In the forebrain, colocalized cells were found in many areas, including the POA, suprachiasmatic nucleus (SCh; Fig. 4), PVN (Fig. 5), LH (Fig. 6), DM (Fig. 7), ventromedial hypothalamus (VMH; Fig. 7) and Arc (Fig. 7). The density of colocalized cells within the PVN varied across the subregions, with the highest absolute number of colocalized cells found in the medial parvocellular PVN (PaMP; 92 ± 24), followed by the posterior PVN (PaPo; 87 ± 51). Despite having the lowest total number of PRV-infected cells within the PVN, the rostral PVN (PaV; 24.3 ± 12.6) possessed a high percentage of these cells (∼80%) that also were MC4-R mRNA-positive. The VMH also had double-labeled cells, but to a more modest degree than the PVN (means ± SE: 62.8 ± 19.2; percent: 49.1 ± 6.6%). Similar to what we have noted before, injections of PRV into IWAT cause only sparse infection in the VMH predominantly along the outer edges of the nucleus (2, 10, 42, 44) (Fig. 7). Furthermore, MC4-R mRNA within this region was not impressive. MC4-R + PRV cells also were found in both subregions of the SCh (Fig. 4), at approximately equivalent ratios (SChdm: 50.8 ± 3.6%; SChvl: 47.6 ± 0.3%). Colocalized MC4-R + PRV cells were sparse within the thalamic areas but were noticeably concentrated in the ventral reunions nucleus (Table 1). In addition, we found double-labeled cells highly concentrated within an area ventral to the zona incerta (ZI) and lateral to the PVN and dorsolateral to the LH proper. This area has not been explicitly defined in any rodent brain atlas to our knowledge, and thus, for lack of a better description, we termed this the sub-zona incerta (subZI) region of the LH (Table 1; Fig. 6).
MC4-R + PRV neurons were found in the midbrain predominantly in the PAG and at its most caudal end (near the rostral brainstem) in the parabrachial regions (Fig. 8). Most of the ∼800 MC4-R + PRV-labeled cells in the PAG were found within the LPAG (∼366) and dorsomedial PAG (∼144). The highest percent colocalization in the midbrain was found in the nucleus of the fields of Forel (84.4 ± 5.8%) and in the Edinger Westphal nucleus (78.2 ± 8.9%), despite the scant infection in these areas (26.5 ± 7.9 and 23.0 ± 10.9, respectively).
Several areas in the brainstem showed high levels of colocalization, particularly the sol (65.5 ± 0.8%; Fig. 9), RVL/C1 (61.7 ± 1.6%; Fig. 9), gigantocellular reticular nucleus (62.1 ± 1.5%), and AP (58.6 ± 5.7%), but also collectively in the reticular parvicellular reticular nucleus (69.7 ± 4.4%), intermediate reticular nucleus (64.1 ± 2.6%), paramedian reticular nucleus (63.1 ± 4.3%), and raphe regions (RC) (71.6 ± 1.9%), rostral linear/caudal linear nucleus of the raphe (66.7 ± 11.8%), and dorsal raphe (66.7 ± 11.5%). The dorsomotor nucleus of the vagus nerve (DMV) had very few MC4-R + PRV cells (18.5 ± 4.3); however, this number comprised 60.0 ± 2.9% of PRV-infected cells in this nucleus (Fig. 9).
The results of the present study show that the central origins of the SNS outflow neurons to IWAT, labeled using the transneuronal tract tracer PRV, were highly colocalized with MC4-R mRNA across the neuroaxis. Such a predominant colocalization suggests an important role of the melanocortins in the functions of the SNS innervation of WAT, including the principal initiation of lipid mobilization from WAT (for reviews, see Refs. 5 and 7). These data support the suggestive physiological evidence for increases in lipid mobilization triggered by central melanocortin receptor stimulation (e.g., 23, 39) and add to the important role of the melanocortins in energy balance beyond that known for the control of ingestive behaviors (15, 40, 41) and thermogenesis (51, 52). Although the present work focuses on the possible role of central melanocortins stimulating lipid mobilization via the SNS innervation of WAT, we are not dismissing the direct stimulation of melanocortin receptors possessed by white adipocytes, especially the MC4-R, in lipolysis (22). The physiological significance of peripheral melanocortin-stimulated lipolysis, however, is unknown.
The extensive histological analyses presented here show the colocalization of MC4-R mRNA with the sympathetic outflow from brain to IWAT; however, because we are assaying gene expression, these transcription data do not necessarily indicate translation into MC4-R protein for these cells, as is always the caveat when mRNA, rather than protein, is measured. Furthermore, these data are specific for IWAT sympathetic circuitry but do not necessarily generalize to the SNS innervation of other WAT pads. We previously have labeled the SNS outflow circuits from brain to epididymal WAT (EWAT), as well as to retroperitoneal WAT (RWAT) in Siberian hamsters (2, 10, 42, 44) and IWAT and EWAT in laboratory rats (2). Superficially, there are more similarities than differences in the labeled SNS outflow circuits among WAT pads and between species (for a review, see Ref. 7); however, this does not necessarily mean that similar locations and levels of colocalization of MC4-R mRNA with the SNS circuitry innervating these fat pads exist. We recently found in a preliminary study that a single third ventricular injection of MTII significantly increased the sympathetic drive (i.e., norepinephrine turnover) to IWAT, dorsosubcutaneous WAT, and RWAT but not to EWAT in Siberian hamsters (Brito N, Brito M, Rahman W, and Bartness, T, unpublished observations). These data suggest the likelihood of colocalization of MC4-Rs with neurons composing the sympathetic outflow to several WAT pads.
In addition to the MC4-Rs produced by adipocytes, MC4-Rs are extensively distributed throughout the CNS in laboratory rats (26) and mice (29). Because our focus was to test whether MC4-R mRNA is expressed by the central origins of the sympathetic outflow from brain to IWAT, rather than the distribution of brain MC4-R mRNA per se, we did not quantify the single-labeling of the latter here. Because hybridization with sense control probes did not result in autoradiographic signals (data not shown), we are confident that this pervasive nature of the MC4-R mRNA in Siberian hamster brain represents bona fide labeling of the gene expression for this melanocortin receptor subtype and the ribonuclease protection assay data (Fig. 1) suggest the same.
There were large numbers of cells labeled for both MC4-R mRNA and PRV (MC4-R + PRV) across the neuroaxis (Table 1) in a pattern similar to that for MC4-R mRNA alone in laboratory rats and mice (26, 29). Specifically, percentages of MC4-R + PRV neurons were remarkably high [between ∼29% (LC) ∼ 87% (Li)]. Such consistently high levels of colocalization are relatively rare for PRV-labeled SNS outflow to a wide range of peripheral tissues [i.e., SNS outflow from brain to the adrenal gland (47), stellate ganglion (heart; Ref. 46), pancreas (24), or wall of the tail artery (43)], where neurochemical phenotyping was a goal (see discussion in Ref. 42 for an overview of this issue). For example, although percentages of colocalized PRV-infected neurons can reach high values [e.g., PRV injected into pancreas shows ∼60% PRV + tyrosine hydroxylase colocalization in the subceruleus (SubC) (30) and PRV injected into the tail artery wall shows ∼95% PRV + tyrosine hydroxylase colocalization in the SubC (43)], 80/104 or 77% of brain structures had MC4-R+PRV colocalization percentages >58% in the present study (Table 1).
Ideally, because the present results provide a map of MC4-R + PRV colocalization, these data could be integrated with the results of physiological studies showing the stimulation or inhibition of lipid mobilization after parenchymal injections of MC4-R receptor agonists or antagonists, respectively. To date, however, we are not aware of such data. Nevertheless, some of the sites showing colocalization of MC4-R + PRV are implicated in the stimulation of lipid mobilization in experiments in which a proxy of lipolysis, such as circulating free fatty acid (FFA), is measured after brain stimulation. For example, in the forebrain, electrical stimulation of the mammillary-premammillary area in laboratory rats (4) and the ZI in dogs (34) and cats (3) result in increases in plasma FFA concentrations. The MC4-R + PRV-labeled neurons in the ZI and surround garnered our attention because of the high number and percentage of colocalized labeled neurons (means ± SE: 111.0 ± 30.3 and 68.5 ± 3.2%, respectively), as well as in the subZI region mentioned above (means ± SE: 278.0 ± 73.6% and 73.6 ± 3.6%, respectively). Electrical stimulation of the SCh, another area with MC4-R + PRV labeling, leads to decreases in the respiratory quotient, demonstrating utilization of lipid-derived fuel (12). Furthermore, knife cuts caudal to the SCh block lipid mobilization due to cold exposure, forced exercise, insulin-induced hypoglycemia and 24-h fasts (18).
Not all the electrical stimulation data, however, are consistent with the colocalizations seen here or with sites of PRV infection alone identified in our previous studies of the sympathetic outflow from brain to WAT in Siberian hamsters (2, 10, 42, 44) or laboratory rats (2). For example, electrical stimulation of the VMH, but not the LH in rabbits (28) and electrical stimulation of the VMH in rats (48) increases circulating FFAs. We have only seen sparse PRV infections in the VMH here and previously (2, 10, 42, 44), and this is confined to a region just lateral to the nucleus. The lack of double-labeling in the VMH in the present study, or even of MC4-R mRNA alone (data not shown) is consistent with the modest MC4-R labeling in mice expressing green fluorescent protein under the control of a MC4-R promoter (29) and in rats using in situ hybridization for the MC4-R mRNA (26). The positive effects of VMH electrical stimulation on lipid mobilization might be due to ancillary stimulation of the descending PVN projections coursing just laterally to the VMH (31) that are part of the SNS outflow to WAT (for discussion of this issue, see Refs. 2 and 5). In the present study, as in our previous studies (2, 10, 42, 44), the LH is intensively infected, but support for the involvement of this area in lipid mobilization is missing or has not been tested. When rabbit LH is electrically stimulated, lipid mobilization is not increased but does occur with VMH stimulation in these animals (28). The lack of LH stimulation-induced lipolysis may be because the LH is a large heterologous area, and it therefore is easy to imagine that stimulation of one subarea might not trigger lipolysis, whereas stimulation of another area might do so. Finally, there are no studies of midbrain or brainstem sites identified as part of the SNS outflow to WAT using PRV, where lipid mobilization has been examined to date.
The hypothalamic PVN had extensive MC4-R + PRV colocalizations in the present study and is well recognized as a key component in energy balance, including its role in the autonomic functioning of most peripheral organs (for a review, see Ref. 35). Specifically in the present study, the PVN subregion with the highest ratio of MC4R + PRV-labeled cells was the PaV (∼80%), with the highest absolute number of colocalized cells within the PVN being found in the parvocellular subdivisions (i.e., PaMP, PaPo). These areas compose the parvocellular monosynaptic connections to spinal preganglionic SNS cells that significantly contribute to the PVN control of autonomic functions in the periphery. The abundant PVN colocalization data here, as well as the acknowledged role of the PVN in energy balance in general, make it a prime site for future functional studies of lipid mobilization by the melanocortins.
The DMV has traditionally been seen as a source of parasympathetic nervous system (PSNS) preganglionic neurons, and neurons infected there after WAT PRV injections have been used as evidence for the PSNS innervation of this tissue (27). In our seminal study of the SNS outflow from brain to IWAT (2), we found PRV-labeled DMV neurons, results consistent with our findings in the present study. In both laboratory rats (26) and mice (29), there are double-labeled DMV neurons showing MC4-R mRNA or expressing GFP under the control of a MC4-R promoter, respectively, with immunohistochemistry for choline acetyltransferase (ChAT), a marker of preganglionic PSNS neurons. In these studies (26, 29), however, not all MC4-R neurons were double-labeled for ChAT, suggesting that not all DMV neurons are PSNS preganglionics. Moreover, it is becoming increasingly clear that traditional characterizations of nuclei as strictly PSNS or SNS are not definitive (e.g., 19, 25). In addition, there is no histological or biochemical evidence for acetylcholine-related factors in WAT of laboratory rats (1), nor have we seen any vesicular acetylcholine transporter immunoreactivity in laboratory rat or mouse WAT (Giordano A, Song CK, Bowers R, Ehlen JC, Bartness T and Cinti S, unpublished observation), casting further doubt as to the PSNS innervation of WAT. Indeed, chemical sympathectomy of the SNS innervation of WAT, accomplished by local injections of the catecholaminergic neurotoxin 6-hydroxy-dopamine, blocks all infections in the spinal cord and brain in Siberian hamsters subsequently injected with PRV (Giordano A, Song CK, Bowers R, Ehlen JC, Bartness T and Cinti S, unpublished observation). Thus we attribute the DMV double-labeling here and single PRV-labeling in our earlier work (2) as rogue sympathetic neurons in an area otherwise traditionally recognized as a principle origin of PSNS outflow to the periphery.
Finally, across the neural axis, the sol showed the highest absolute number of colocalized MC4-R + PRV cells. The sol has been associated with the control of ingestive behavior and body weight (for a review, see Ref. 8) and more specifically, melanocortin involvement in the control of food intake has been suggested for this region (e.g., 17, 54). The melanocortins and the brainstem, including the sol, also are implicated in the control of brown fat thermogenesis via its sympathetic innervation (e.g., 51). Because conditions such as cold exposure trigger increases in sympathetic drive (norepinephrine turnover) to both brown fat (to increase thermogenesis) and WAT (to mobilize lipid fuels for thermogenesis) (16), it may be that the MC4-R + PRV cells labeled are involved in the latter.
Collectively, the data of the present study show an extensive colocalization of MC4-R + PRV in the sympathetic outflow circuits from the brain to WAT. Moreover, this colocalization occurred across the neuroaxis strongly implicating the melanocortins and MC4-Rs specifically as important components of the sympathetic control of lipid mobilization. Because the present data provide a map of sites with high MC4-R + PRV concentrations, future studies can be conducted testing the ability of site-specific melanocortin injections to alter lipid mobilization.
This research was supported by National Institute of Health Research Grant R01 DK-35254 and Georgia State University Research Enhancement Fund to T. J. Bartness and to the STC Program of the National Science Foundation under Agreement No. IBN-9876754 for the Viral Tract-Tracing Core.
The authors thank David Marshall for providing excellent animal care. The authors also thank Dr. Lynn Enquist (Princeton University) for his continued technical and conceptual guidance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society