Abstract
Immune activation results in adaptive neuroendocrine responses, including activation of the hypothalamic–pituitary–adrenal axis, which are dependent on the integrity of medullary catecholaminergic (CA) systems. In contrast, although specific roles of pontine, midbrain, and hypothalamic CA systems in neuroendocrine function have been described, the functional roles of these CA systems in modulating neuroendocrine function during immune responses have not been investigated. We have, therefore, investigated the effects of immune activation on the various CA systems of the central nervous system (CNS) and explored this relationship with changes in plasma corticosterone and plasma prolactin. Male BALB/c mice were injected with lipopolysaccharide (LPS, 500 μg/kg i.p.) and 2 h later cardiac blood was taken and mice were perfused with fixative. Immunostaining procedures were performed using antibodies raised against c-Fos and tyrosine hydroxylase, a marker of CA neurons, and detailed topographical analysis of the CA systems within the CNS was performed. LPS-injected mice had increased concentrations of plasma corticosterone and decreased concentrations of plasma prolactin compared with vehicle-injected controls. LPS-injected mice had increased numbers of c-Fos-positive CA neurons within the medullary (A1, A2, C1, C2), pontine (A6) and midbrain (A10) cell groups when compared with vehicle-injected controls. Among hypothalamic CA cell groups, LPS had differential effects on the numbers of c-Fos-positive CA neurons in topographically organised subdivisions of the arcuate nucleus (A12). Changes in plasma prolactin concentrations correlated with the numbers of c-Fos-positive CA neurons within the area postrema, the medullary CA cell groups, the medial posterior division of the arcuate, and the zona incerta. The present study identifies topographically organised, anatomically distinct CA systems that are likely to modulate some of the neuroendocrine responses to immune activation, and may provide novel targets for the relief of symptoms associated with illness and disease.
Introduction
The changes in neuroendocrine activity in response to peripheral immune system activation are an important adaptive response. Neuroendocrine responses to immune activation include increases in plasma concentrations of corticosterone and either increases or decreases in plasma concentrations of prolactin (Rettori et al. 1994, Rivest 2001, De Laurentiis et al. 2002). These hormones, in turn, influence homeostatic functions and provide feedback to regulate the immune system (Buckingham et al. 1996, Freeman et al. 2000). Although these changes are adaptive during acute states of immune activation, chronic sickness and immune-related disease can lead to decreases in quality of life and increases in both morbidity and mortality (Curtis & Patrick 2003). A better understanding of the neural systems regulating neuroendocrine responses to peripheral immune activation may lead to novel therapeutic strategies for the treatment of chronic illness and immune-related disease.
Previous studies have implicated brainstem catecholaminergic (CA) systems in the neuroendocrine responses to peripheral immune activation. In particular, brainstem CA systems are integral to the increased hypothalamic–pituitary–adrenal (HPA) activity. Noradrenaline within the paraventricular nucleus of the hypothalamus (Pa) stimulates corticotrophin-releasing factor gene expression, leading to activation of the HPA axis (Itoi et al. 1994), while depletion of noradrenaline and adrenaline can block increases in plasma corticosterone concentrations following immune activation (Chuluyan et al. 1992, Swiergiel et al. 1996). The brainstem CA systems within the nucleus of the solitary tract (nTS) and ventral lateral medulla (VLM), via direct noradrenergic and adrenergic fibre projections to the Pa, influence the HPA axis response to peripheral immune activation (Sawchenko & Swanson 1981, Cunningham et al. 1990, Ericsson et al. 1994). Chemical or surgical ablation of the medullary CA cell bodies or disruption of their hypothalamic projection systems attenuates the increased HPA activity associated with immune system activation (Chuluyan et al. 1992, Ericsson et al. 1994).
Other CA systems within the midbrain and hypothalamus may be involved in neuroendocrine responses to immune system activation. Immune activation results in increased extracellular concentrations of dopamine within various regions of the brain that may be involved in neuroendocrine regulation, including the arcuate nucleus (Arc), Pa and hippocampus (MohanKumar et al. 1999). These regions are known to be innervated by a variety of potential sources of dopamine, including the ventral tegmental area (VTA, A10) and zona incerta (ZI, A13) (Swanson 1982, Wagner et al. 1995). The secretion of some neuroendocrine hormones such as prolactin is regulated by dopamine (Reymond & Porter 1985), but the effects of immune activation on prolactin are controversial (Rettori et al. 1994, De Laurentiis et al. 2002) and deserve further attention.
The aim of the present study was to characterise the responses of topographically organised subpopulations of CA neurons within the central nervous system (CNS) to i.p. administration of an acute peripheral immune stimulus, lipopolysaccharide (LPS; cell-wall component of Gram-negative bacteria), by using double immunostaining for the protein product of the immediate-early gene c-fos and tyrosine hydroxylase (TH), a marker of CA neurones. The second aim of the study was to correlate changes in c-Fos expression within medullary, pontine, midbrain and hypothalamic CA cell groups with neuroendocrine responses as measured by changes in plasma corticosterone and plasma prolactin, in order to identify CA cell groups that are likely to be involved in neuroendocrine function during states of sickness or disease.
Materials and Methods
Animals
Adult male BALB/c mice (6–8 weeks, 21–23 g) were housed in a group of 12 in a temperature-controlled room (23±2 °C) on a 14 h:10 h light:darkness cycle (lights on at 0500 h), with food and water available ad libitum. Mice were allowed to acclimate under these conditions for one week before the experiment. Small cardboard ‘huts’ were kept in the home cage at all times prior to and throughout the experiment to enrich the environment and allow mice to seek darkness during the light phase. All animal procedures were approved by the University of Bristol Ethical Review Group and were conducted in accordance with Home Office guidelines and the UK Animals (Scientific Procedures) Act 1986. In addition, all studies were consistent with the NIH Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85A23) and were covered by Animal Welfare Assurance #A5057-01.
LPS administration and tissue processing
At 15-min intervals between 1200 h and 1330 h, time-matched pairs of mice were given i.p. injections (100 μl) of either sterile 0.05 M sodium phosphate-buffered saline (PBS), pH 7.4 (PBS; n = 5), or 12.5 μg LPS dissolved in PBS (n = 7; Escherichia coli LPS, 026:B6, Sigma, UK). After injections, animals were put back into their home cage. Two hours post-injection mice were deeply anaesthetised with 150 μl Sagatal (sodium pentobarbital, 60 mg/ml i.p.) and thoracic cavities were exposed. Approximately 0.5 ml blood was taken by cardiac puncture, mixed with 50 μl 0.5 M ethylenediamine tetraacetic acid (EDTA), and kept on ice until centrifugation at 10 000 r.p.m. for 10 min; plasma was stored at −20 °C until used for radioimmunoassay. Immediately after cardiac puncture mice were perfused through the ascending aorta of the heart with 150 ml PBS at 4 °C followed by 250 ml 4% paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.4 (PB), at 4 °C. Brains were post-fixed in the same paraformaldehyde solution overnight at 4 °C, followed by two 12 h washes in PB at 4 °C. Brains were then placed into PB containing 30% sucrose at 4 °C for 3 days. Brains were blocked using a mouse brain matrix (RBM-2000C, ASI Instruments, Warren, MI, USA) to ensure a consistent coronal plane in each brain. Brains were then frozen on dry ice and stored at −80 °C until sectioning. Brains were sectioned in the coronal plane at a thickness of 30 μm and collected in a series of six consecutive wells containing cryoprotectant (0.05 M phosphate buffer, 30% ethylene glycol w/w, 20% glycerol w/w, pH 7.4) and stored at −20 °C until immunostaining.
Immunohistochemistry
Every sixth section of the brain (and to obtain higher resolution every third section for the hypothalamus) was used for double immunostaining using an antibody directed against the protein product of the immediate-early gene, c-fos (rabbit anti-c-Fos polyclonal antibody, PC-38 (Ab-5), 1:12 000; Merck Biosciences, Nottingham, Notts, UK), and an antibody directed against TH (rabbit anti-TH polyclonal antibody, AB152, 1:4000; Chemicon, Chandlers Ford, Hants, UK). Free-floating tissue was incubated in 24-well tissue culture plates and washed in plastic tubs using mesh wells (Corning Costar, Sunderland, Tyne & Wear, UK), and gently shaken on an orbital shaker throughout double immunostaining. Tissue was first washed in PBS for 15 min, then incubated in 1% hydrogen peroxide in PBS for 15 min, washed again for 15 min in PBS, pre-incubated in PBS containing 0.3% Triton X-100 (PBST) for 15 min, and then incubated for 12–16 h with rabbit anti-c-Fos antibody in PBST. Tissue was then washed twice for 15 min in PBST followed by incubation with a biotinylated swine anti-rabbit IgG polyclonal antibody (E0353, 1:200; DAKO, Ely, Cambridgeshire, UK) in PBST for 90 min. Tissue was again washed twice for 15 min in PBST followed by incubation with Elite ABC reagent (PK-6100, 1:200; Vector, Orton Southgate, Peterborough, UK) in PBST for 90 min. Finally, tissue was washed for 15 min in PBST, 15 min in PBS, and then incubated in SG substrate (Vector, UK; diluted as recommended by Vector) in PBS for 10 min. Tissue was immediately washed in PBS for 15 min, incubated in 1% hydrogen peroxide in PBS for 15 min, washed in PBS for 15 min followed by PBST for 15 min, then incubated with rabbit anti-TH antibody in PBST for 12–16 h. All subsequent steps were identical to those described above for the immunoperoxidase localisation of c-Fos-immunoreactivity, except for the substrate reaction. Sections were incubated in a solution containing 0.01% 3,3′-diaminobenzidine tetrahydrochloride (DAB) and 0.0015% hydrogen peroxide in PBS for 17 min, and then washed in PBST for 15 min and PBS for 15 min. Brain slices were then mounted on gelatine-coated glass slides, dehydrated and cleared with xylene, then coverslipped using DPX mounting medium (R A Lamb, London, UK). The colour reaction product of the TH immunostaining was a golden-brown colour and localised to the cytoplasm while the colour reaction product of the c-Fos immuno-staining was a dark blue-black colour and localised to the nucleus. Immunohistochemical controls in which the primary antibodies were omitted resulted in lack of positive staining on a random sample of experimental tissue.
Cell counting
Cell counting was performed in a blind manner with respect to treatment groups after randomisation of slides. Cell counts were made to determine (1) the number of c-Fos-positive CA (TH-positive) neurons, (2) the total number of CA neurons and (3) the number of c-Fos-positive non-catecholaminergic (non-CA, TH-negative) nuclei. The catecholaminergic cell groups, brain regions, and bregma levels analysed are listed in Table 1.
Corticosterone and prolactin measurements
For measurement of corticosterone, plasma was thawed immediately before use and diluted 1:100 in a 0.025 M citrate buffer, pH 3.0 (B buffer; to denature the binding globulin) and aliquoted into three 100 μl volumes. Rabbit anti-corticosterone polyclonal antiserum (50 μl, 1:124 final dilution; kindly supplied by Prof. G Makara, Institute of Experimental Medicine, Budapest, Hungary) in B buffer was added to each tube followed by 50 μl 125I-corticosterone tracer with a specific activity of 2–3 mCi/μg (MP Biomedicals, Irvine, CA, USA). Tubes were mixed well and left to incubate overnight at 4 °C. The B buffer (0.5 ml) containing 0.5% charcoal and 0.05% dextran T70 (Sigma) was then added to each tube, mixed well, and centrifuged at 4000 r.p.m. at 4 °C for 15 min; the supernatant was siphoned off and the total radioactivity in the remaining pellet was determined using a Cobra II gamma counter (Packard Bioscience, UK). Plasma prolactin concentrations were determined by the National Hormone and Peptide Program (Dr A F Parlow, Harbor-UCLA, CA, USA) using a radioimmunoassay.
Imaging
All images were captured using a Leica DMLB microscope fitted with an Insight digital camera (Leica Microsystems, Germany) and SPOT image capture software v4.0.2 (Diagnostic Instruments, Sterling Heights, MI, USA), and all figures were designed and assembled in CorelDRAW 12.0 (Corel Corp., Eden Prairie, MN, USA).
Statistics
All reported values are mean values and standard error of the mean (S.E.M.). For the cell count data, each of the three dependent variables was analysed using a single multifactorial analysis of variance (ANOVA) with repeated measures, using TREATMENT as the between subject factor and REGION (e.g. each CA group at a specific bregma level) as the within subject factor for repeated measures analysis. The left and right hemispheres of the brain were not distinguished so the cell count data represents the total of the respective left and right sides. Regions in which the cell counts of the treatment groups had mean values of zero were excluded from the multi-factorial ANOVA with repeated measures. Missing values were replaced by the method of Peterson (1985) prior to the multifactorial ANOVA with repeated measures, but the original data were used for post-hoc analysis and for representation of the data in the tables and figures. When TREATMENT effects or TREATMENT*REGION interactions were observed, multiple pair-wise comparisons were made using Fischer’s Protected Least Significant Difference tests. Corticosterone and prolactin data were analysed using Student’s t-test for independent samples. Correlations between (1) plasma corticosterone or prolactin and cell count data and (2) cell count data across different regions were determined using Pearson correlation. In all cases, significance was accepted at P < 0.05.
Results
CA cell groups and regions studied are depicted in Fig. 1, and a detailed list of regions that were counted and their abbreviations is presented in Table 1. Out of all the different regions studied (Table 1), only the cell count data in which post-hoc analysis revealed a significant TREATMENT effect is presented schematically (Fig. 1) and in detail (Table 2). Seventy-two different cell counts (from CA cell groups at specific bregma levels) had no double immunostaining in either treatment group, and these regions were not used in the multifactorial ANOVA with repeated measures. Multifactorial ANOVA with repeated measures revealed a TREATMENT*REGION interaction for the number of c-Fos-positive CA neurons (F(50,500) = 2.815, < P 0.001), and revealed a TREATMENT*REGION interaction for the number of c-Fos-positive non-CA nuclei (F122,1220 = 6.864, < P 0.001). There was neither a TREATMENT effect nor a TREATMENT*REGION interaction in the total numbers of CA neurons between treatment groups (F1,10 = 2.925, P = 0.118; F122,1220 = 0.598, P = 0.948).
Medullary noradrenergic (A1/A2) and adrenergic (C1/C2/C3) neurons
LPS-injected mice had increased numbers of c-Fos-positive CA neurons along the rostral–caudal axis of the medulla compared with vehicle-injected controls (Table 2, Fig. 2). LPS administration resulted in increased numbers of c-Fos-positive CA neurons within the area postrema (AP), A1, A1/C1, A2/C2 and C1 regions compared with vehicle-injected mice. LPS administration also resulted in increased numbers of c-Fos-positive non-CA nuclei within the AP, A1/C1, A2 and A2/C2 regions. There were no differences in the numbers of c-Fos-positive CA neurons or c-Fos-positive non-CA nuclei in the C3 region between treatment groups. In summary, i.p. administration of LPS increased the number of c-Fos-positive CA neurons within the AP and several medullary CA groups.
Pontine noradrenergic neurons (A5/A6/A7)
Analysis of pontine regions containing noradrenergic neurons revealed an increase in the numbers of c-Fos-positive CA neurons within subdivisions of the A6 cell group (Table 2, Fig. 3). LPS-injected mice had increased numbers of c-Fos-positive CA neurons within the rostral levels of the locus coeruleus (LC), the medial parabrachial nucleus (MPB), central parts of the lateral parabrachial nucleus (LPBC) and ventral subcoeruleus (SubCV) regions when compared with vehicle-injected mice. There were no differences in the numbers of c-Fos-positive CA neurons or c-Fos-positive non-CA nuclei within the A5 and A7 regions between treatment groups, and no differences in the numbers of c-Fos-positive non-CA nuclei within the A6 subdivisions between treatment groups. In summary, i.p. administration of LPS increased the number of c-Fos-positive CA neurons within subdivisions of the A6 cell group including the LC, MPB, LPBC and SubCV.
Midbrain catecholaminergic neurons (A8, A9, A10, A11)
Analysis of midbrain regions containing putative dopaminergic neurons revealed an increase in the numbers of c-Fos-positive CA neurons solely within the rostral portion of the VTA, but not other A10 cell groups (Table 2, Fig. 4). The main dopaminergic cell groups of the midbrain are those within the substantia nigra (SN, A9) and VTA (A10); additional putative dopaminergic cells are present in the pedunculopontine tegmental nucleus (PPTg, A8), retrorubral field (RRF, A8), rostral linear nucleus (RLi, A10) and periaqueductal grey (PAG, A11). There was virtually no double immunostaining in any of the divisions of A8, A9 or A11. Mice given injections of LPS had decreased numbers of c-Fos-positive non-CA nuclei within the regions of the ventral subdivision of A8, the substantia nigra lateral (SNL; A9), the VTA (A10) and the PAG (A11) cell groups compared with vehicle-injected mice. In summary, i.p. administration of LPS increased the number of c-Fos-positive CA neurons within the rostral portion of the VTA and decreased the numbers of c-Fos-positive non-CA nuclei within subdivisions of the A8, A9, A10 and A11 regions.
Hypothalamic catecholaminergic neurons (A11, A12, A13, A14, A15)
Analysis of hypothalamic and thalamic regions containing putative dopaminergic neurons revealed differential responses in the numbers of c-Fos-positive CA neurons that were restricted to subdivisions of the Arc (Table 2, Fig. 5). LPS-injected mice had increased numbers of c-Fos-positive CA neurons in the medial posterior Arc (ArcMP), dorsal Arc (ArcD) and rostral Arc compared with vehicle-injected mice. In contrast, LPS administration resulted in decreased numbers of c-Fos-positive CA neurons within the lateral posterior Arc (ArcLP). There were no differences in the numbers of c-Fos-positive CA neurons within the A11, A13 or A14 cell groups between treatment groups. Due to lack of tissue containing the A15 cell group, the region was not included in the analysis. Mice given injections of LPS had decreased numbers of c-Fos-positive non-CA nuclei within the ZI at one bregma level, and increased numbers of c-Fos-positive non-CA nuclei within all subdivisions of the Pa studied compared with vehicle-injected controls. In summary, i.p. administration of LPS had differential effects on the numbers of c-Fos-positive CA neurons within subdivisions of the Arc but not other hypothalamic CA cell groups.
Corticosterone, prolactin and correlation analysis
LPS-injected mice had increased plasma corticosterone and decreased plasma prolactin concentrations measured 2 h after injection compared with vehicle-injected controls, and these measurements were correlated with cell count data within various brain regions (Fig. 6). Correlation analysis revealed correlations between plasma corticosterone concentrations and (1) the number of c-Fos-positive CA neurons within the AP, A1/C1, A2/C2, MPB and VTA, and (2) the number of c-Fos-positive non-CA nuclei within the Pa and the region of the A2 cell group. Correlation analysis also revealed correlations between plasma prolactin concentrations and (1) the numbers of c-Fos-positive CA neurons within the AP, A1/C1, A2/C2, ArcMP and ZI, and (2) the number of c-Fos-positive non-CA neurons within the ArcMP.
Discussion
This study provides a detailed analysis of topographically organised CA systems in the mouse brain that are candidates for components of the neural systems mediating neuroendocrine responses to peripheral immune activation. We have been able to provide indirect evidence that CA neurons within the parabrachial and subcoeruleus regions play a role in relaying signals of peripheral immune activation within the mouse CNS. Although previous studies have described increases in dopaminergic neurotransmission in the CNS in response to peripheral immune activation, this is the first study to illustrate responses to peripheral immune activation within midbrain dopaminergic cell groups. We have also investigated the functional properties of dopaminergic neurons within the Arc, and for the first time have shown differential c-Fos responses to peripheral immune activation in subdivisions of the Arc which also correlate with plasma prolactin concentrations. Interestingly, we have identified multiple regions within the brain where peripheral immune activation results in decreased c-Fos expression (ventral A8, substantia nigra lateral, VTA, PAG) compared with controls. Thus, the present data suggest that there are distinct, topographically organised subsets of medullary, pontine, midbrain and hypothalamic CA systems that modulate the neuroendocrine responses to peripheral immune system activation.
In order to identify c-Fos-immunoreactivity within topographically organised subpopulations of CA neurons with a high level of neuroanatomical detail, double immunostaining of the protein product of the immediate-early gene c-fos was used in conjunction with TH immunostaining of CA neurons. While a change in the c-Fos-immmunoreactivity does not necessarily indicate a change in electrophysiological or electrochemical properties of a cell (e.g. firing rate, depolarization, calcium influx) (Morgan & Curran 1986, Luckman et al. 1994), it does provide a useful method to identify functional cellular responses at single cell resolution among large populations of cells. It is also important to be aware of temporal factors when comparing the present data with previous physiological and behavioural studies because the expression of c-Fos is transient and varies across the circadian cycle (Grassi et al. 1994, Kovacs 1998).
Although omission of the primary antibodies from the immunostaining protocol resulted in a lack of positive staining, it does not confirm that the antibodies are in fact specific for the protein markers of interest. Control peptides were not used but the antibodies used in the present study have been used previously for the identification of changes in c-Fos-immunoreactivity (Zhang et al. 2003) and the identification of CA neurons in mouse tissue (Horger et al. 1998). In this study, we describe TH immunostaining in several regions not previously described in mouse brain (MPB, lateral parabrachial nucleus (LPB), A8, RLi, PAG); however all of these regions have been found to contain CA neurons in at least one of the multiple vertebrate species studied to date (Smeets & Gonzalez 2000).
Medullary and pontine CA systems
The LPS-induced effects we observed in the medulla are consistent with previous findings in rats where LPS administration resulted in increased numbers of c-Fos-positive CA neurons within specific CA groups (A1, A1/C1, A2/C2, C2) along the rostral–caudal extent of the medulla (Ericsson et al. 1994, Dayas et al. 2001). Our observation that LPS failed to alter the numbers of c-Fos-positive CA neurons within the mouse A2 region (−7.48 to −7.32 mm bregma) appears to conflict with a recent study in rats demonstrating that interleukin-1 (IL-1) treatment results in increased numbers of c-Fos-positive noradrenergic neurons that are more rostral than in rats exposed to other stressors such as restraint or footshock (Dayas et al. 2001). However, the bregma levels in mouse brain containing noradrenergic neurons (−7.20 to −6.96 mm bregma) equivalent to the IL-1-responsive regions in rat brain described by Dayas and colleagues are more rostral than the bregma levels we defined as the A2 cell group. We defined this region of the mouse nTS (−7.20 to −6.96 mm bregma) as part of the A2/C2 because it contains a mixture of noradrenergic and adrenergic neurons that we could not distinguish using an antibody raised against TH. Taking into account these differences in nomenclature used to describe mouse and rat brain, the present results demonstrating that medullary noradrenergic and/or adrenergic CA neurons are responsive to peripheral immune activation are consistent with previous reports (Ericsson et al. 1994, Dayas et al. 2001).
The existence of CA neurons within the AP has previously been described (Dahlstrom & Fuxe 1964), and to our knowledge this is the first study describing LPS-induced increases in c-Fos-positive CA neurons within the AP. The AP has long been implicated in rapid homeostatic responses such as those associated with osmo-regulatory and nutrient challenges, and is an important centre for the integration of systemic and visceral sensory information as it is a circumventricular organ and also receives sensory information from peripheral vagal ganglia (Oldfield & McKinley 2004). The projections of the AP have been shown to be primarily restricted to the brainstem, with dense projections to CA neurons of the nTS (Cunningham et al. 1994) and moderate projections to the external lateral parabrachial nucleus, the dorsal motor nucleus of the vagus (DMN), nucleus ambiguus and VLM (Shapiro & Miselis 1985). The AP has been shown to be critical for brainstem c-Fos responses, hypothalamic paraventricular c-Fos responses, and HPA axis responses to immune stimuli (Lee et al. 1998). The projections and functions of the CA neurons within the mouse AP are not known, although CA projections from the AP to the rostral VLM have been found in the rabbit (Blessing et al. 1987) so it is feasible that the mouse brain contains similar connections.
The present study is the first to identify CA neurons of the A6 cell group (MPB, LPBC, SubCV) other than the LC that are responsive to peripheral LPS injection. The CA neurons within the LC have repeatedly been shown to be responsive to immune stimuli (Zalcman et al. 1991, Hare et al. 1995, Kaneko et al. 2001), and the present results confirm these previous studies. The external portion of the LPB (LPBE) is known to be especially responsive to immune activation (Elmquist et al. 1996), although no CA cells are located within this region. Because only regions containing CA cells were focused on in this study, single c-Fos counts were not performed within all subdivisions of the parabrachial nucleus. The LPBC projects to the Pa in the rat (Bester et al. 1997), although the neuroanatomical projections of PB subdivisions in the mouse have not been studied and it is not known if the neurochemical phenotypes of this presumed connection includes CA neurons. Reciprocal connections between the MPB and subdivisions of the nTS (caudal region of commissural subdivision, rostral portions of nTS) have been shown in the rat (Herbert et al. 1990, Krukoff et al. 1993), and this may be an important circuit for relaying visceral signals related to immune stimuli within the mouse CNS as well. Cells of the SubCV are known to project to the spinal cord in the rat (Clark & Proudfit 1991), although the functional significance of this connection is not known. This study provides evidence that there are further topographical subdivisions of the mouse A6 cell group (MPB, LPB, SubCV) that may relay signals of peripheral immune activation within the CNS.
Midbrain CA systems
The A10 dopaminergic neurons within the VTA are often associated with motivation, incentive, and reward (Le Moal & Simon 1991), and it is both interesting and novel that LPS administration resulted in increased c-Fos-positive CA neurons at the level of the A10 dopaminergic cell body. The VTA is the major source of the mesocorticolimbic dopaminergic system that projects to forebrain structures including the nucleus accumbens, prefrontal cortex and the amygdala (Swanson 1982). Peripheral immune activation results in increased extracellular concentrations of dopamine within the nucleus accumbens and prefrontal cortex (Lacosta et al. 1994, Hayley et al. 2001), and our results support the hypothesis that the VTA may be the source of limbic dopamine changes in response to peripheral immune activation. Although the role of the VTA during immune responses is not known, it is feasible that the VTA is involved in behavioural reinforcement during states of sickness or disease.
Hypothalamic CA systems
To our knowledge, this is the first study reporting differential c-Fos responses within topographically organised subsets of Arc dopaminergic (DA) neurons of the mouse in response to immune activation. The Arc is known to express IL-1 receptors, and peripheral immune activation results in c-fos mRNA induction within the Arc (Brady et al. 1994, Ericsson et al. 1995). The majority of IL-1 receptors within the Arc are within the medial division that contains the majority of the median eminence (ME)-projecting tuberoinfundibular dopaminergic (TIDA) neurons thought to be the primary source of dopamine acting to regulate prolactin secretion (Rethelyi 1985, Reymond & Porter 1985). The lateral divisions of the Arc are not thought to be as important in prolactin regulation, as they are thought to project primarily to the lateral divisions of the ME (Lofstrom et al. 1976). In the rat, these cells within the lateral divisions of the Arc contain putative CA neurons that do not express the enzyme aromatic amino acid decarboxylase that is necessary for dopamine synthesis (Meister et al. 1988), and the functions of these cells have yet to be elucidated. One study reported similar results in the Arc as in the present study after administering a dopamine receptor antagonist instead of LPS; however this also resulted in increased plasma prolactin concentrations (Hentschel et al. 2000). Prolactin is thought to be the main negative-feedback regulator of TIDA neurons because TIDA neurons lack autoreceptor feedback (Demarest & Moore 1979). However, treatment with prolactin alone has provided controversial results, resulting in increased numbers of c-Fos-positive CA neurons within either both subdivisions of the Arc (Hentschel et al. 2000) or only the medial subdivision (Cave et al. 2001), suggesting prolactin is not the sole regulator of TIDA neurons. The present results add further evidence that the medial and lateral subdivisions of the Arc are indeed functionally distinct, and are differentially responsive to peripheral immune activation.
The present results are in accordance with a recent study showing decreased plasma prolactin concentrations following immune activation (De Laurentiis et al. 2002); however there are multiple other studies reporting either increased plasma prolactin concentrations (Rettori et al. 1994, Phelps et al. 2001, Gonzalez et al. 2004) or no change in prolactin (Abreu et al. 1994) following immune activation. The major difference between these conflicting results is the route of LPS administration or, in the case of studies which found no changes in plasma prolactin i.c.v., IL-1β was administered instead of LPS. In studies using LPS given intraperitoneally (De Laurentiis et al. 2002), including the present study, LPS injection resulted in decreased serum prolactin concentrations. In contrast, in studies using high doses of LPS given i.v. (Rettori et al. 1994, Phelps et al. 2001) or LPS given i.c.v. (Gonzalez et al. 2004), LPS injection resulted in increased serum prolactin concentrations. One explanation is that when LPS is given intravenously or intracerebroventricularly, LPS actions at the level of the Arc result in increased release of prolactin. The Arc is highly vascularised and capable of directly sensing circulating signals of immune activation (Brady et al. 1994, Ericsson et al. 1995, Herkenham et al. 1998). When LPS is given in compartments such as the peritoneal cavity, activation of the Arc may be primarily due to ascending signals arising from brainstem nuclei such as medullary CA systems. The medullary neuronal systems innervate the Arc (Ricardo & Koh 1978) and adrenergic fibres actually terminate on TIDA neurons (Hrabovszky & Liposits 1994). Because LPS administration results in c-Fos responses within medullary CA systems, this could be a possible mechanism of relaying signals of peripheral immune activation from the periphery to the Arc.
Correlation analyses
Correlation analyses support the involvement of particular CA systems in both the LPS-induced increases in plasma corticosterone concentrations and the LPS-induced decreases in plasma prolactin concentrations. The correlation between the numbers of c-Fos-positive CA neurons within medullary CA systems and plasma corticosterone concentrations largely confirms previous studies, whereas the correlation between the numbers of c-Fos-positive CA neurons in the MPB and VTA are unreported as far as the authors are aware. However, the MPB and VTA have not been shown to project to the Pa or to be directly involved in the increased HPA axis activity in response to immune activation (Elmquist & Saper 1996), so the correlations should not be over interpreted. The negative correlation between the numbers of c-Fos-positive CA neurons in the ArcMP and plasma prolactin concentrations, although not strongly correlated, further supports the involvement of TIDA neurons in prolactin release in response to immune activation. The strong correlation between the numbers of c-Fos-positive CA neurons within the ZI and plasma prolactin concentrations was unexpected, although the ZI is largely associated with the reproductive axis and is known to be involved in the regulation of other neuro-endocrine hormones (MacKenzie et al. 1988). The correlation between the numbers of c-Fos-positive CA neurons within the medullary CA cell groups and plasma prolactin concentrations further supports the idea that LPS injection, when given intraperitoneally, may relay signals to the Arc through medullary CA cell groups.
Conclusions
The present study provides evidence that the CA systems of the CNS are composed of functionally distinct, topographically organised subgroups that are likely to be involved in a variety of neuroendocrine, autonomic and behavioural responses to immune activation. Our results provide further evidence for the involvement of medullary (A1, A2, C1, C2) and pontine (LC) CA systems, and identify new CA subpopulations within the medulla (AP), pons (MPB, LPB, SubCV) and midbrain (VTA) that may be involved in neuroendocrine responses to peripheral immune activation. This study is the first to demonstrate that peripheral immune activation results in differential c-Fos responses within topographically distinct subpopulations of TIDA neurons which are known to have important roles in neuroendocrine hormone regulation, including the LPS-induced decrease in the concentration of plasma prolactin. This study has also identified various brainstem (A1/C1, A2/C2) and hypothalamic (ArcMP, ZI) CA cell groups where cell counts were correlated with plasma prolactin concentrations. These subpopulations of CA neurons deserve further attention in efforts to define neural systems contributing to dysregulation of CNS and endocrine systems in acute and chronic illness and immune-related disease.
Catecholaminergic groups and regions studied, with abbreviations
| Brain region | Bregma level (mm) | |
|---|---|---|
| Catecholaminergic cell groups, brain regions and abbreviations of the mouse were defined based on the work in the rat by Everitt et al.(1992), in multiple species by Smeets and Gonzalez (2000) and in the mouse by Paxinos and Franklin (2001). CVO, circumventricular organ. | ||
| CA group | ||
| CVO | Area postrema (AP) | −7.48 |
| A1 | Ventrolateral medulla (VLM) | −7.48, −7.32 |
| A2 | Nucleus of the solitary tract (nTS) | −7.48, −7.32 |
| A1/C1, C1 | Ventrolateral medulla (VLM) | −7.20, −6.96, −6.84 |
| A2/C2, C2 | Nucleus of the solitary tract (nTS) | −7.20, −6.96, −6.84 |
| C3 | Dorsal midline of rostral medulla oblongata (RMO) | −6.96, −6.84 |
| A5 | Ventrolateral periolivary area | −5.40 |
| A6 | Locus coeruleus (LC) | −5.80, −5.68, −5.52, −5.34 |
| Dorsal subcoeruleus (SubCD) | −5.20, −5.02, −4.96 | |
| Ventral subcoeruleus (SubCV) | −5.20, −5.02, −4.96 | |
| Medial parabrachial nucleus (MPB) | −5.20 | |
| Central part of the lateral parabrachial nucleus (LPBC) | −5.20, −5.02 | |
| A7 | Lateral pons | −4.96 |
| A8 | Pedunculopontine tegmental nucleus (PPTg; caudal part of A8) | −4.24 |
| Retrorubral field (RRF; dorsal part of A8) | −4.16, −4.04, −3.88 | |
| Pedunculopontine tegmental nucleus (PPTg; ventral part of A8) | −4.16, −4.04 | |
| A9 | Substantia nigra, compact (SNC) | −3.80, −3.64, −3.52, −3.16, −3.08 |
| Substantia nigra, lateral (SNL) | −3.80, −3.64, −3.52, −3.16 | |
| Substantia nigra, reticular (SNR) | −3.88, −3.80, −3.64, −3.52 | |
| A10 | Ventral tegmental area (VTA) | −3.88, −3.80, −3.64, −3.52 |
| Caudal linear nucleus (CLi) | −4.24, −4.16, −4.04 | |
| Rostral linear nucleus (RLi) | −3.88, −3.80, −3.64, −3.52 | |
| A11 | Dorsal raphe nucleus, caudal part (DRC) | −4.96 |
| Dorsal raphe nucleus (DR) | −4.24, −4.16, −4.04 | |
| Periaquaductal gray (PAG) | −4.24, −4.16, −4.04, −3.88, −3.80, −3.64, −3.52, −3.16, −3.08, −2.70, −2.54 | |
| Periventricular field (pv) | −2.46, −2.30, −2.18, −2.06 | |
| A12 | Medial posterior arcuate nucleus (ArcMP) | −2.70, −2.54, −2.46, −2.30, −2.18 |
| Dorsal arcuate nucleus (ArcD) | −2.06 | |
| Lateral posterior arcuate nucleus (ArcLP) | −2.54, −2.46, −2.30, −2.18 | |
| Lateral arcuate nucleus (ArcL) | −2.06 | |
| Arcuate nucleus (Arc) | −1.58, −1.46, −1.34, −1.22 | |
| A13 | Zona incerta (ZI) | −1.58, −1.46, −1.34, −1.22 |
| Dorsomedial nucleus of the hypothalamus (DM) | −2.18, −2.06, −1.58, −1.46, −1.34 | |
| Anterior hypothalamic area (AHP) | −1.22 | |
| A14 | Periventricular hypothalamic nucleus (Pe) | −1.58, −1.46, −1.34, −1.22, −1.06, |
| −0.94, −0.46, −0.10, 0, +0.14 | ||
| Paraventricular hypothalamic nucleus, posterior part (PaPo) | −1.06 | |
| Paraventricular hypothalamic nucleus, magnocellular part (PaM) | −0.94 | |
| Paraventricular hypothalamic nucleus, medial parvocellular part (PaMP) | −1.06, −0.94 | |
| Suprachiasmatic nucleus (SCh) | −0.46 | |
| Bed nucleus of the stria terminalis, medial division, posteromedial part (BSTMPL) | −0.46 | |
| Anterodorsal preoptic nucleus (ADP) and bed nucleus of the stria terminalis, medial division, ventral part (BSTMV) | −0.10 | |
| Medial preoptic nucleus, medial and central part (MPOM, MPOC) | 0.00, +0.14 | |
Cell counts of c-Fos-positive CA (TH-positive) neurons, total CA (TH-positive) neurons, and c-Fos-positive non-CA (TH-negative) nuclei within medullary, pontine, midbrain and hypothalamic catecholaminergic cell groups in which LPS injection resulted in a significant change compared with vehicle injection. Cell counts are means ±s.e.m.
Sagittal (A) and coronal (B) diagrams depicting catecholaminergic cell groups of mouse brain used in analysis and affected by lipopolysaccharide treatment. The effects of LPS are depicted by the degree of shading: no shading, no change in c-Fos; light grey, a change in c-Fos-positive non-CA nuclei; dark grey, a change in c-Fos-positive CA neurons; black, a change in both c-Fos-positive CA and non-CA nuclei. The diagram is adapted from a mouse brain stereotaxic atlas (Paxinos & Franklin 2001). 4 V, fourth ventricle; LV, lateral ventricle. For additional abbreviations see Table 1.
Citation: Journal of Endocrinology 184, 2; 10.1677/joe.1.05839
Medullary CA neurons are responsive to peripheral LPS injection (vehicle injection: A–C, G–I; LPS injection: D–F, J–L). LPS administration resulted in increased numbers of c-Fos-positive CA neurons within the AP (D, E), A1 (D, F), C1 (J, L) and C2 (J, K) cell groups compared with vehicle-injected controls (A–C, G–I). B,C, E,F, H,I and K,L are high magnifications of A, D, G and J respectively. Open arrows represent c-Fos-negative CA neurons while closed arrows indicate c-Fos-positive CA neurons. Arrowheads indicate c-Fos-positive non-CA nuclei. Scale bar = 200 μm (A, D, G, J) or 50 μm (B,C, E,F, H,I, K,L). For abbreviations see Table 1.
Citation: Journal of Endocrinology 184, 2; 10.1677/joe.1.05839
Multiple subpopulations of the A6 cell group in addition to the LC are responsive to peripheral LPS injection (vehicle injection: A,B, E,F; LPS injection: C,D, G,H). LPS administration resulted in increased numbers of c-Fos-positive CA neurons within the LC (C, D) and MPB (G, H) but not the LPB (G, H) or SubCD (G) at −5.20 mm bregma compared with vehicle-injected controls (A, B, E, F). B, D, F and H are higher magnifications of A, C, E and G respectively. Open arrows indicate c-Fos-negative CA neurons while closed arrows indicate c-Fos-positive CA neurons. Scale bar = 200 μm (A, C, E, G), 50 μm (B, D) or 25 μm (F, H). PB, parabrachial nucleus; scp, superior cerebellar peduncle. For additional abbreviations see Table 1.
Citation: Journal of Endocrinology 184, 2; 10.1677/joe.1.05839
Dopaminergic midbrain neurons within the ventral tegmental area (VTA) are responsive to peripheral LPS injection. LPS administration resulted in increased numbers of c-Fos-positive CA neurons within the VTA (C, D) but not the SN, RLi or PAG when compared with vehicle-injected controls (A, B). B and D are higher magnifications of A and C respectively. Open arrows indicate c-Fos-negative CA neurons while closed arrows indicate c-Fos-positive CA neurons. Scale bar = 200 μm (A, C) or 25 μm (B, D). For abbreviations see Table 1.
Citation: Journal of Endocrinology 184, 2; 10.1677/joe.1.05839
Dopaminergic neurons within subdivisions of the arcuate nucleus are differentially responsive to peripheral LPS injection (saline injection: A-F, M-R; LPS injection: G-L, S-X). LPS administration resulted in increased numbers of c-Fos-positive CA neurons in the ArcMP at −2.18 mm bregma (K, W), in the ArcD at −2.06 mm bregma (L), and in the Arc at −1.58 and −1.46 mm bregma (S, T), but decreased numbers of c-Fos-positive CA neurons in the ArcLP at −2.18 mm bregma (K, X) compared with vehicle-injected controls (A-F, M-R). Q and R are higher magnifications of E while W and X are higher magnifications of K. Open arrows represent c-Fos-negative CA neurons and closed arrows represent c-Fos-positive CA neurons. Scale bar = 100 μm (A-P, S-V) or 25 μm (Q, R, W, X). For abbreviations see Table 1.
Citation: Journal of Endocrinology 184, 2; 10.1677/joe.1.05839
Peripheral LPS injection results in neuroendocrine hormone changes that are correlated with cell count data from various brain regions. LPS administration resulted in increased plasma corticosterone concentrations compared with vehicle-injected controls (A), and corticosterone concentrations correlated with the numbers of c-Fos-positive CA neurons within the MPB (B) and VTA (C). LPS administration resulted in decreased plasma prolactin concentrations compared with vehicle-injected controls (D), and prolactin concentrations correlated with the numbers of c-Fos-positive CA neurons within the A1/C1 (E), ArcMP (F) and ZI (G) and with the numbers of c-Fos-positive non-CA neurons within the ArcMP (H). The slope of each line (correlation coefficient, r) and P value of the correlation analysis are shown for each scatter plot. *P < 0.05, **P < 0.001 versus vehicle-injected controls. For abbreviations see Table 1.
Citation: Journal of Endocrinology 184, 2; 10.1677/joe.1.05839
This work was supported by the Neuroendocrinology Charitable Trust, UK (grant no. PMS/VW-00/01-606 to S L L and C A L). Dr Christopher A Lowry is a Wellcome Trust Research Fellow (RCDF 068558/Z/02/Z). The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
References
Abreu P, Llorente E, Hernandez MM & Gonzalez MC 1994 Interleukin-1 beta stimulates tyrosine hydroxylase activity in the median eminence. Neuroreport 5 1356–1358.
Bester H, Besson JM & Bernard JF 1997 Organization of efferent projections from the parabrachial area to the hypothalamus: a Phaseolus vulgaris-leucoagglutinin study in the rat. Journal of Comparative Neurology 383 245–281.
Blessing WW, Hedger SC, Joh TH & Willoughby JO 1987 Neurons in the area postrema are the only catecholamine-synthesizing cells in the medulla or pons with projections to the rostral ventrolateral medulla (C1-area) in the rabbit. Brain Research 419 336–340.
Brady LS, Lynn AB, Herkenham M & Gottesfeld Z 1994 Systemic interleukin-1 induces early and late patterns of c-fos mRNA expression in brain. Journal of Neuroscience 14 4951–4964.
Buckingham JC, Loxley HD, Christian HC & Philip JG 1996 Activation of the HPA axis by immune insults: roles and interactions of cytokines, eicosanoids, glucocorticoids. Pharmacology, Biochemistry and Behavior 54 285–298.
Cave BJ, Wakerley JB, Luckman SM & Tortonese DJ 2001 Hypothalamic targets for prolactin: assessment of c-Fos induction in tyrosine hydroxylase- and proopiomelanocortin-containing neurones in the rat arcuate nucleus following acute central prolactin administration. Neuroendocrinology 74 386–395.
Chuluyan HE, Saphier D, Rohn WM & Dunn AJ 1992 Noradrenergic innervation of the hypothalamus participates in adrenocortical responses to interleukin-1. Neuroendocrinology 56 106–111.
Clark FM & Proudfit HK 1991 The projection of locus coeruleus neurons to the spinal cord in the rat determined by anterograde tracing combined with immunocytochemistry. Brain Research 538 231–245.
Cunningham ET Jr, Bohn MC & Sawchenko PE 1990 Organization of adrenergic inputs to the paraventricular and supraoptic nuclei of the hypothalamus in the rat. Journal of Comparative Neurology 292 651–667.
Cunningham ET Jr, Miselis RR & Sawchenko PE 1994 The relationship of efferent projections from the area postrema to vagal motor and brain stem catecholamine-containing cell groups: an axonal transport and immunohistochemical study in the rat. Neuroscience 58 635–648.
Curtis JR & Patrick DL 2003 The assessment of health status among patients with COPD. European Respiratory Journal Supplement 41 36s–45s.
Dahlstrom A & Fuxe K 1964 Localization of monoamines in the lower brain stem. Experientia 20 398–399.
Dayas CV, Buller KM, Crane JW, Xu Y & Day TA 2001 Stressor categorization: acute physical and psychological stressors elicit distinctive recruitment patterns in the amygdala and in medullary noradrenergic cell groups. European Journal of Neuroscience 14 1143–1152.
De Laurentiis A, Pisera D, Caruso C, Candolfi M, Mohn C, Rettori V & Seilicovich A 2002 Lipopolysaccharide- and tumor necrosis factor-alpha-induced changes in prolactin secretion and dopaminergic activity in the hypothalamic–pituitary axis. Neuroimmunomodulation 10 30–39.
Demarest KT & Moore KE 1979 Comparison of dopamine synthesis regulation in the terminals of nigrostriatal, mesolimbic, tuberoinfundibular and tuberohypophyseal neurons. Journal of Neural Transmission 46 263–277.
Elmquist JK & Saper CB 1996 Activation of neurons projecting to the paraventricular hypothalamic nucleus by intravenous lipopolysaccharide. Journal of Comparative Neurology 374 315–331.
Elmquist JK, Scammell TE, Jacobson CD & Saper CB 1996 Distribution of Fos-like immunoreactivity in the rat brain following intravenous lipopolysaccharide administration. Journal of Comparative Neurology 371 85–103.
Ericsson A, Kovacs KJ & Sawchenko PE 1994 A functional anatomical analysis of central pathways subserving the effects of interleukin-1 on stress-related neuroendocrine neurons. Journal of Neuroscience 14 897–913.
Ericsson A, Liu C, Hart RP & Sawchenko PE 1995 Type 1 interleukin-1 receptor in the rat brain: distribution, regulation, and relationship to sites of IL-1-induced cellular activation. Journal of Comparative Neurology 361 681–698.
Everitt BJ, Meister B & Hokfelt T 1992 The organization of monoaminergic neurons in the hypothalamus in relation to neuroendocrine integration. In Neuroendocrinology, 1st edn, pp 87–128. Ed. CB Nemeroff. Boca Raton: CRC Press.
Freeman ME, Kanyicska B, Lerant A & Nagy G 2000 Prolactin: structure, function, and regulation of secretion. Physiological Reviews 80 1523–1631.
Gonzalez MC, Abreu P, Barroso-Chinea P, Cruz-Muros I & Gonzalez-Hernandez T 2004 Effect of intracerebroventricular injection of lipopolysaccharide on the tuberoinfundibular dopaminergic system of the rat. Neuroscience 127 251–259.
Grassi ZG, Menegazzi M, Carcereri DP, Vescia S, Ranucci G & Bentivoglio M 1994 Different programs of gene expression are associated with different phases of the 24 h and sleep-wake cycles. Chronobiologia 21 93–97.
Hare AS, Clarke G & Tolchard S 1995 Bacterial lipopolysaccharide-induced changes in FOS protein expression in the rat brain: correlation with thermoregulatory changes and plasma corticosterone. Journal of Neuroendocrinology 7 791–799.
Hayley S, Lacosta S, Merali Z, van Rooijen N & Anisman H 2001 Central monoamine and plasma corticosterone changes induced by a bacterial endotoxin: sensitization and cross-sensitization effects. European Journal of Neuroscience 13 1155–1165.
Hentschel K, Moore KE & Lookingland KJ 2000 Effects of prolactin on expression of Fos-related antigens in tyrosine hydroxylase-immunoreactive neurons in subdivisions of the arcuate nucleus. Brain Research 857 110–118.
Herbert H, Moga MM & Saper CB 1990 Connections of the parabrachial nucleus with the nucleus of the solitary tract and the medullary reticular formation in the rat. Journal of Comparative Neurology 293 540–580.
Herkenham M, Lee HY & Baker RA 1998 Temporal and spatial patterns of c-fos mRNA induced by intravenous interleukin-1: a cascade of non-neuronal cellular activation at the blood–brain barrier. Journal of Comparative Neurology 400 175–196.
Horger BA, Nishimura MC, Armanini MP, Wang LC, Poulsen KT, Rosenblad C, Kirik D, Moffat B, Simmons L, Johnson E Jr, Milbrandt J, Rosenthal A, Bjorklund A, Vandlen RA, Hynes MA & Phillips HS 1998 Neurturin exerts potent actions on survival and function of midbrain dopaminergic neurons. Journal of Neuroscience 18 4929–4937.
Hrabovszky E & Liposits Z 1994 Adrenergic innervation of dopamine neurons in the hypothalamic arcuate nucleus of the rat. Neuroscience Letters 182 143–146.
Itoi K, Suda T, Tozawa F, Dobashi I, Ohmori N, Sakai Y, Abe K & Demura H 1994 Microinjection of norepinephrine into the paraventricular nucleus of the hypothalamus stimulates corticotropin-releasing factor gene expression in conscious rats. Endocrinology 135 2177–2182.
Kaneko YS, Ikemoto K, Mori K, Nakashima A, Nagatsu I & Ota A 2001 Expression of GTP cyclohydrolase I in murine locus coeruleus is enhanced by peripheral administration of lipopolysaccharide. Brain Research 890 203–210.
Kovacs KJ 1998 c-Fos as a transcription factor: a stressful (re)view from a functional map. Neurochemistry International 33 287–297.
Krukoff TL, Harris KH & Jhamandas JH 1993 Efferent projections from the parabrachial nucleus demonstrated with the anterograde tracer Phaseolus vulgaris leucoagglutinin. Brain Research Bulletin 30 163–172.
Lacosta S, Merali Z, Zalcman S & Anisman H 1994 Time-dependent in vivo mesolimbic dopamine variations following antigenic challenge. Brain Research 664 225–230.
Le Moal M & Simon H 1991 Mesocorticolimbic dopaminergic network: functional and regulatory roles. Physiological Reviews 71 155–234.
Lee HY, Whiteside MB & Herkenham M 1998 Area postrema removal abolishes stimulatory effects of intravenous interleukin-1 beta on hypothalamic–pituitary–adrenal axis activity and c-fos mRNA in the hypothalamic paraventricular nucleus. Brain Research Bulletin 46 495–503.
Lofstrom A, Jonsson G & Fuxe K 1976 Microfluorimetric quantitation of catecholamine fluorescence in rat median eminence. I. Aspects on the distribution of dopamine and noradrenaline nerve terminals. Journal of Histochemistry and Cytochemistry 24 415–429.
Luckman SM, Dyball RE & Leng G 1994 Induction of c-fos expression in hypothalamic magnocellular neurons requires synaptic activation and not simply increased spike activity. Journal of Neuroscience 14 4825–4830.
MacKenzie FJ, James MD & Wilson CA 1988 Changes in dopamine activity in the zona incerta (ZI) over the rat oestrous cycle and the effect of lesions of the ZI on cyclicity: further evidence that the incerto–hypothalamic tract has a stimulatory role in the control of LH release. Brain Research 444 75–83.
Meister B, Hokfelt T, Steinbusch HW, Skagerberg G, Lindvall O, Geffard M, Joh TH, Cuello AC & Goldstein M 1988 Do tyrosine hydroxylase-immunoreactive neurons in the ventrolateral arcuate nucleus produce dopamine or only L-dopa? Journal of Chemical Neuroanatomy 1 59–64.
MohanKumar SM, MohanKumar PS & Quadri SK 1999 Lipopolysaccharide-induced changes in monoamines in specific areas of the brain: blockade by interleukin-1 receptor antagonist. Brain Research 824 232–237.
Morgan JI & Curran T 1986 Role of ion flux in the control of c-fos expression. Nature 322 552–555.
Oldfield BJ & McKinley MJ 2004 Circumventricular organs. In The Rat Nervous System, 3rd edn, pp 389–406. Ed G Paxinos. London: Elsevier Academic Press.
Paxinos G & Franklin KBJ 2001 The Mouse Brain in Stereotaxic Coordinates, 2nd edn. London: Academic Press.
Peterson RG 1985 Design and Analysis of Experiments. New York: Marcel Dekker, Inc.
Phelps CP, Dong JM, Chen LT & Menzies RA 2001 Plasma interleukin-1 beta, prolactin, ACTH and corticosterone responses to endotoxin after damage of the anterior hypothalamic area. Neuroimmunomodulation 9 340–351.
Rethelyi M 1985 Dendritic arborization and axon trajectory of neurons in the hypothalamic arcuate nucleus of the rat–updated. Neuroscience 16 323–331.
Rettori V, Dees WL, Hiney JK, Lyson K & McCann SM 1994 An interleukin-1-alpha-like neuronal system in the preoptic–hypothalamic region and its induction by bacterial lipopolysaccharide in concentrations which alter pituitary hormone release. Neuroimmunomodulation 1 251–258.
Reymond MJ & Porter JC 1985 Involvement of hypothalamic dopamine in the regulation of prolactin secretion. Hormone Research 22 142–152.
Ricardo JA & Koh ET 1978 Anatomical evidence of direct projections from the nucleus of the solitary tract to the hypothalamus, amygdala, and other forebrain structures in the rat. Brain Research 153 1–26.
Rivest S 2001 How circulating cytokines trigger the neural circuits that control the hypothalamic–pituitary–adrenal axis. Psychoneuroendocrinology 26 761–788.
Sawchenko PE & Swanson LW 1981 Central noradrenergic pathways for the integration of hypothalamic neuroendocrine and autonomic responses. Science 214 685–687.
Shapiro RE & Miselis RR 1985 The central neural connections of the area postrema of the rat. Journal of Comparative Neurology 234 344–364.
Smeets WJ & Gonzalez A 2000 Catecholamine systems in the brain of vertebrates: new perspectives through a comparative approach. Brain Research. Brain Research Reviews 33 308–379.
Swanson LW 1982 The projections of the ventral tegmental area and adjacent regions: a combined fluorescent retrograde tracer and immunofluorescence study in the rat. Brain Research Bulletin 9 321–353.
Swiergiel AH, Dunn AJ & Stone EA 1996 The role of cerebral noradrenergic systems in the Fos response to interleukin-1. Brain Research Bulletin 41 61–64.
Wagner CK, Eaton MJ, Moore KE & Lookingland KJ 1995 Efferent projections from the region of the medial zona incerta containing A13 dopaminergic neurons: a PHA-L anterograde tract-tracing study in the rat. Brain Research 677 229–237.
Zalcman S, Shanks N & Anisman H 1991 Time-dependent variations of central norepinephrine and dopamine following antigen administration. Brain Research 557 69–76.
Zhang YH, Lu J, Elmquist JK & Saper CB 2003 Specific roles of cyclooxygenase-1 and cyclooxygenase-2 in lipopolysaccharide-induced fever and Fos expression in rat brain. Journal of Comparative Neurology 463 3–12.
