7_differential induction of c-fos immunoreactivity

Upload: tacitus58

Post on 07-Apr-2018

219 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/6/2019 7_Differential Induction of C-Fos Immunoreactivity

    1/7

    BrainResearch ullerin, ol . 32,pp. 581-587, 1993 0361-9230/93 6.00 + .OOPrinted in the USA. All rights reserved. Copyright0 1993Pergamon Press Ltd.

    Differential Induction of c-Fos Immunoreactivityin H~othalamus and Brain Stem Nuclei

    following Central and PeripheralAdmi~stration of Endotoxin

    WEIHUA WAN,* LOREN JANZ,* CATHERINE Y. VRIEND,* CRAIG M. SORENSEN,?ARNOLD H. GREENBERG* AND DWIGHT M. NANCE*

    *~epu~~ents of Patho~ o~, Phy~io~o~ and P~y~hology~ Cell Biology,Universityofkfm itoba, W innipeg, Mani toba, R3E 0 W 3 CanadafO ncogene Sciences, Inc., Uniondale, NY 11553Received 16 March 1993; Accepted 7 May 1993

    WAN, W., JANZ, Y. M.in hypothalamusand brain stem nuclei ollow ing central and peripheral administ rationof endotoxin.BRAIN RES BULL 32(6) 58 l-587, 1993.-Lipopolysaccharide (LPS), an endotoxin associated with gram-negative bacteria, isa potent activator of the immune system. We have tested the effects of ICV infusions of LPS (10 ng) or Ringers solution on theinduction of the proto-oncogene protein c-Fos in the brain as well as plasma levels of corticosterone and splenic concentrationsof norepinep~ne (NE) and VIP. At 3 h post-ICV infusion of LPS, numerous labeled neurons were observed in the pamvent~~ularnucleus (PVN) of the h~~~arnus and the nucleus tractus solitarius (A2) region of the brain stem. Also, corticosterone andsplenic NE and VIP levels were all elevated post-ICV LPS. Analysis of the time course for the induction of c-Fos protein in thebrain following IP injections of LPS indicated that, relative to control injections, increased numbers of c-Fos-positive cells weredetected in the PVN 0.5 h following IP injections (100 pg), peaked at 2-3 h postinjection, and then returned to control levels atlater intervals. Additional dose-response data for IP LPS indicated a small increase in the number of labeled cells at a dose of4.0 ~8, and the number and staining intensity increased up to a dose of 100 pg. Corticosterone levels followed a similar patternand were elevated at the 4.0 pg IP dose of LPS and increased to peak levels at 40 rg and higher. In contrast to ICV injections,splenic NE levels were unaltered by IP injections of LPS. An additional difference noted between 10 and IP injections of LPSwas the presence of c-For-labeled neurons in the supraoptic nucleus, arcuate nucleus, and the ventrolateral (A I) region of thebrain stem at IP doses of 40 pg and higher. These results indicate selective and differential effects of central and peripheral LPSon the induction of #OS protein in hypothalamic and brain stem nuclei. The concurrent changes in corticosterone and splenicneurotransmitter levels produced by ICV LPS suggest that the endocrine and autonomic nervous systems are primary targets forthese activational effects of endotoxin.Li~~ly~~ha~deBrain stem Oncogene proteins Pamvent~~ul~ nucleus Supraoptic nucleus Arcuate nucleus

    LIPOPOLYSACCHARIDE (LPS) is a unique glucosamine-based phospholipid that makes up the monolayer of the outermembrane of most gram-negative bacteria (34). LPS produceschanges in behavior (22), endocrine function (45), as well asimmune function (43). Many of these effects of LPS are sec-ondary to the production of cytokineq such as tumor necrosisfactor-alpha (TNF-a) and Interleukin- l/3 (IL- 1 i), by macro-phages (531). Elevations in serum TNF and IL-1 have beenfound in animal models of endotoxin-induced shock (f&27,55),

    and these macrophage-derived mediators have been shown toexhibit potent immunologic and inflammatory properties (42).The central rote of these mediators in LPS-induced shock hasalso been described in various animal models (6,48). Recently,Sundar et al. (43) demonstrated that central injections of LPSproduced a profound suppression of splenic immune functionthat was comparable to the effects of centrally injected IL-I.Furthermore, the immunosuppressive effects of central LPS wereantagonized by (w-MSH, an endogenous IL- 1 antagonist. These

    Requests for reprints should be addressed to D. M. Nance, Department of Pathology, University of Manitoba, 770 Bannatyne, Winnipeg,Manitoba, R3E 0W3 Canada.

    581

  • 8/6/2019 7_Differential Induction of C-Fos Immunoreactivity

    2/7

    WAN ET AL.

    results suggest that the immunosuppressive effects of central LPSinfusions are mediated by the induction of IL-l release. Addi-tional evidence provided by Sundar et al. (43) indicates that theimmunosuppressive effects of central IL- 1 are mediated throughthe activation of both the hypothalamic-pituitary-adrenal axisand the sympathetic nervous system by corticotropin-releasingfactor (CRF).

    In support of these two immune regulatory pathways, theeffects of corticosterone, sympathetic neurotransmitters, andCRF on immune function have been well documented. Immunechanges in the spleen produced by LPS or IL-l infusions areparalleled by a concurrent elevation of ACTH and corticosteronelevels (4,8,12,45,49). Cutting the splenic nerve, which depletessplenic norepinephrine (NE) levels, blocks the immunosup-pressive effects of central IL-l on macrophage function (8) andprovides further evidence that the autonomic nervous systemmay mediate the immunosuppressive effects of IL- 1. In addition,the neurotransmitter NE is an important modulator of immunefunction (3,21,23). For example, increases in hypothalamic andsplenic NE turnover are induced by both LPS and IL-l(12,24,33,50), and increased NE metabolism in the forebrain,brain stem, and thoracic spinal cord are observed following IL-1 infusions ( 19). Recently we showed that cutting the splenicnerve abrogated foot shock-induced immunosuppression ofsplenic cell function in the rat (51). Furthermore, Irvin et al.( 18) have demonstrated that ICV infusions of CRF suppresssedsplenic immune function, and the immunomodulatory effect ofCRF on the spleen was mediated via the sympathetic nervoussystem. Together, these results suggest at least two possible path-ways for LPS-induced immunosuppression. First, LPS-inducedproduction of IL-l (9) may, in turn, stimulate CRF release andthereby increase plasma ACTH and corticosterone (12). Sec-ondly, stimulation of central CRF neurons by LPS/IL-1 mayactivate the autonomic nervous system and produce changes inimmune function by modifying both central and peripheral NEactivity. However, the exact mechanisms and possible centralsites of action for LPS-induced immune changes have yet to bedetermined.

    The expression of immediate early response genes, such asc-pas, has become a powerful tool for structural and functionalanalysis of the nervous system (29,30,41) and may provide in-formation on the mechanism of LPS-induced immunosuppres-sion. The induction of c-Fos has been implicated in differentia-tion and cell proliferation (35,52), and it is known that c-Fosprotein is differentially expressed in regions of the brain andspinal cord depending on the physiological challenge (17,30,54).However, the exact role of c-Fos and related oncogenes is stillunknown, and whether or not c-Fos is involved in the immunechanges induced by LPS remains to be determined. To under-stand better the behavioral, physiological, and immune changesproduced by LPS, we have tested the effects of both central andperipheral injections of LPS on the induction of the proto-on-cogene protein c-Fos in the brain of rats as well as measuredchanges in corticosterone levels in the plasma and neurotrans-mitter content of the spleen.

    METHODExperiment (ICV)Infusions

    mg/kg) and stainless steel cannulausing the stereotaxic coordinates:

    A-P = 0.8 mm, L = 1.3 mm, D-V = -3.0 mm, and securedwith dental cement and screws. Four to 7 days can-nulation, Biol. Campbell,

    Three followingsodium nitrite followed

    Brainswere subsequently cryoprotected in 30% buffered sucrose.Immunocytochemical Serial 40 pm sections

    trans-ferred plates containingbuffer saline (PBS) washed for 30 min. Sections

    rocker with a rabbit antibody to pro-tein (Oncogene Sci., 1:200 in a PBSsolution containing 1% normal serum and 1% Triton

    washed in PBS and then incubatedagain washed three times in PBS and

    incubated rabbit PAP (1:300; Cappel).rinsed in PBS and transferred plates containing

    sayed following method described previously by GreenbergThree

    followingtrunk and spleens were collected. Blood

    plasma was collectedstored at -20C. frozen im-

    mediately assay.Norepinephrine assay. Frozen spleenyield a tissue concentration of 20 mg/ml. Homogenates

    were centrifuged taken for alumina extraction using the ESAplasma catecholamine methodology

    using an ESA 5700solvent delivery with a 0DS2 5-pm column.Analysis

    tissue weight.VIP assay. Pieces of spleen (approximately 50 mg) wereplaced tubes containingdiced and a 40X volume added. werethen placed water bath for 5 min to deactivate

    placed on a shakertubes. Samples were then centrifuged 30,000

    rpm. One milliliter subse-quently reconstituted assay buffer, and 100 ~1 was

    for RIA. The RIA procedureLabs, Belmont,

    Adult (250-300housed individuallycages with food and ad lib and maintained Adult (170-550River, Dorval, groups ofSurgical three per cage with food and ad lib in a reversed

    deeply anesthetized light room with a 12L: 12D lighting schedule.

  • 8/6/2019 7_Differential Induction of C-Fos Immunoreactivity

    3/7

    EFFECTS OF LPS 583

    FIG. 1,PVN of the hypothalamus of a rat showing numerous C-For-positive neurons 3 h following an ICV infusion of 10 ng ofLPS (left) and the PVN of an animal given an ICV injection of Ringers solution 3 h before sacrifice (right). Serial 40 rrn frozensections were used for the immunocy&hemistry. _

    Time course and dose response. In the LPS time-course ex-periment, rats were injected IP with 100 pg of LPS (055:B5,Sigma, St. Louis, MO) in 0.2 ml of saline or 0.2 ml of saline forthe control animals. Animals were overdosed with Nembutaland perfused with fixative as outlined above either 0.5, 1.0,2.0,3.0,6.0, or 24.0 h later. Brains were subsequently sectioned andprocessed for c-Fos immunocytochemistry as outlined previ-ously. In the LPS dose-response experiment, rats were injectedIP with LPS at a dose of 0.4,4.0,40.0, 100.0, or 200.0 pg in 0.2ml of saline. The control animals were injected IP with 0.2 mlof saline. At 3 h postinjection, rats were overdosed and perfusedwith fixative. Brains were subsequently processed for c-Fos im-munocytochemist~. In additional groups of rats, the dose-re-sponse study was repeated and the rats killed by decapitation.Trunk blood was collected for subsequent assay of plasma cor-ticosterone levels and the spleens were removed and frozen ondry ice for later determination of splenic NE levels as outlinedabove.iMa analysis. Brain sections taken at comparable levels ofthe paraventricular nucleus of the hy~thalamus (PVN) of eachanimal were used to index the number of c-Fos-labeled cells foreach group. The most heavily labeled single section through thePVN ofeach animal was selected for cell counts. All cells labeledin a single section were counted, regardless of the intensity ofstaining, in a standardized manner through a microscope usinga grid reticule. All data were analyzed by an analysis of variance.

    RESULTSThe Eflects of ICV Infusions of LPS

    Effects on the induction c-Fos protein in the brain. Figure 1illustrates the large number of labeled neurons focused in thePVN of an adult male rat 3 h following ICV injection of 10 ngof LPS (left side), relative to an animal injected with Ringerssolution shown on the right side. Figure 4 (right side) shows themean number of c-Fos positive-labeled cells in the PVN 3 hfollowing ICV infusion of 10 ng of LPS vs. the number of c-Fospositive-labeled cells in the PVN following an ICV infusion ofRingers. The number of c-Fos-labeled cells following ICV in-jections of 10 ng of LPS was increased by >800% comparedwith the Ringers controls, F(1, 17) = 31.65, p < 0.0001. Inaddition, we found that relative to control animals injected withRingers solution, cells located in the nucleus tractus solitarius

    (A2 cell group) showed positive c-Fos labeling following ICVinfusions of 10 ng of LPS (data not shown).Eflects on Corticosterone Levels and Splenic NE and VIPLevel.7

    We found that ICV injection of 10 ng of LPS produced adramatic increase in corticosterone levels, F(1, 16) = 119.25,p < 0.000 1, and also produced a significant increase in splenicNE, fll, 15) = 14.63, p < 0.002, and VIP, F(1,17) = 5.03,p < 0.037, levels 3 h following injection, relative to animalsinfused with Ringers solution (Figs. 2 and 3).Eflects of Systemic Injections of LPS

    Dose-response effects on c-Fos protein in the brain. Therewas a dose-related increase in the number of c-Fos positive cellslocalized in the PVN of the hypothalamus following IP injectionof LPS (Fig. 4). Relative to saline or the 0.4 pg dose, a modest

    8001

    Ringers LPS7

    FIG. 2. ICV injection of 10 ng of LPS produces a dramatic increase inco~~~terone levels 3 h following infusion, relative to Ringers injectedcontrols. Values of corticosterone levels are expressed as the mean(+SEM), and the number of animals per group are indicated within thebar graphs. *Denotes significantly different from Ringers control.

  • 8/6/2019 7_Differential Induction of C-Fos Immunoreactivity

    4/7

    584 WAN ET AL.

    1100

    s??Ic- 900%.o5 700

    ::

    500

    *I

    Time Course for t he Induction of c-Fos Labeled Neurons inthe PVNBased upon a 100 rg dose of LPS and relative to the control

    injections, increased numbers of c-Fos positive cells were detectedin the PVN 0.5 h following IP injection of LPS, F( 1, 22) = 4.62,p < 0.04, peaked at 2-3 h postinjection, then decreased at 6 h,F( 1, 17) = 3.04, p < 0.09 vs. saline, and returned to belowcontrol levels by 24 h postinjection (Fig. 5).cl_0 fl0- Dose-Response Effects on Corticosterone and Splenic NELevels7

    *I9-

    Ringers LPSThere was a dose-related increase in corticosterone levels fol-

    lowing IP injection of LPS. As illustrated in Fig. 6, the dose-response relationship for corticosterone levels followed the samepattern as that observed for c-Fos protein labeling in the PVN(see Fig. 4). Corticosterone levels were significantly elevated atthe 4.0 pg dose of LPS relative to saline, F( 1, 25) = 6.7 1,p < 0.02, and the 0.4 pg dose, F(l, 25) = 5.60, p < 0.03, andincreased to peak levels at doses of 40 pg, F( 1, 25) = 20.64, p< 0.0003, and higher. However, in marked contrast to ICV in-jections of LPS, splenic NE levels were unaltered by IP injectionsof LPS, relative to saline injected controls, F(5, 24) = 0.73, p