| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Physiology and Functional Genomics and McKnight Brain Institute, University of Florida, Gainesville, Florida, USA; and
Department of Physiology, University of Texas Health Science Center at San Antonio, San Antonio, Texas, USA
1Correspondence: Department of Physiology and Functional Genomics, College of Medicine, Box 100274, 1600 SW Archer Rd., University of Florida, Gainesville, FL 32610-0274, USA. E-mail: csumners{at}phys.med.ufl.edu
ABSTRACT
Macrophage migration inhibitory factor (MIF) acts intracellularly to counteract the angiotensin (ANG) II type 1 receptor (AT1-R)-mediated chronotropic effect of ANG II in hypothalamic neurons, an effect mediated by the thiol-protein oxidoreductase (TPOR) activity of the MIF molecule. Here we determined the in vivo actions of MIF in regulating the physiological actions of ANG II that are mediated via the paraventricular nucleus (PVN), an area that serves as a relay point in the central nervous system (CNS)-mediated effects of ANG II on cardiovascular functions and water intake. Intracerebroventricular (icv) injection of ANG II into normotensive rats selectively increased MIF protein levels in the PVN and produced significant pressor and drinking responses that were inhibited by PVN administration of the AT1-R antagonist losartan. Overexpression of MIF in PVN neurons via Ad-Syn-MIF gene transfer attenuated the pressor and drinking responses produced by icv-injected ANG II. Consistently, intracellular application of MIF or MIF-(50–65) (which harbors the TPOR activity of MIF) into PVN sympathetic regulatory neurons, blunted the electrophysiological actions of ANG II at these cells. These observations establish for the first time that MIF within the PVN, acting via TPOR, is an intracellular regulator of the central cardiovascular and dipsogenic effects of ANG II.—Li, H., Gao, Y., Freire, C. D., Raizada, M. K., Toney, G. M., Sumners, C. Macrophage migration inhibitory factor in the PVN attenuates the central pressor and dipsogenic actions of angiotensin II
Key Words: ANG II • blood pressure • paraventricular nucleus • thiol-protein oxidoreductase
ANGIOTENSIN II(ANG II) acts within the brain to increase blood pressure, and the physiological mechanisms involved in this effect include stimulation of sympathetic vasomotor pathways, release of vasopressin (AVP), and dampening of baroreflexes (1
, 2)
. These central nervous system (CNS) effects of ANG II involve activation of neuronal ANG II type 1 receptors (AT1-R) at specific circumventricular organs (CVO) including the subfornical organ (SFO) and the organum vasculosum of the lamina terminalis (OVLT), with subsequent activation of hypothalamic and brainstem sites such as the paraventricular nucleus (PVN), rostral ventrolateral medulla (RVLM), and nucleus tractus solitarius (NTS; ref.3
4
5
6
). Certain of these sites, including the SFO and PVN, are also involved in mediating the central dipsogenic actions of ANG II (1
, 2)
. Much data exist on the intracellular signals that mediate neuronal actions of ANG II, including activation of a calcium-dependent protein kinase C (PKC) and calcium-calmodulin-dependent kinase II, generation of reactive oxygen species (ROS), and modulation of membrane K+, Ca2+, and nonselective cation currents (7
8
9
10)
. In contrast, there is very little information on how these rapid neuronal actions of ANG II are counter regulated or turned off.
It is generally accepted that the interaction of ligands with G protein-coupled receptors (GPCR) initiates a series of intracellular signaling events, including activation of ß-arrestins, that ultimately leads to desensitization and down-regulation of the receptor (11)
. However, there is little cellular or physiological evidence to indicate that such mechanisms exist for the AT1-R. For example, the acute peripheral or CNS actions of ANG II to increase blood pressure are greater in hypertensive animals (which express an overactive renin-angiotensin system) than in normotensive rats (12
13
14)
. At the cellular level, it appears that arrestin-mediated signaling for the AT1-R may be distinct from that observed with other GPCRs and may not be linked to receptor desensitization (15
16
17)
. Finally, there is little evidence that exposure of the neuronal AT1-R to ANG II results in its desensitization/down-regulation. Collectively, these observations suggest that alternate mechanisms may exist to counteract AT1-R-mediated actions in neurons.
One such mechanism may be macrophage migration inhibitory factor (MIF). MIF is a 12.5 kDa protein that is expressed in immune tissues and has established roles in immune responses (18)
. In addition, in the nervous system MIF is expressed in neurons within the hypothalamus, cortex, hippocampus, and pons (19)
, and evidence indicates that MIF may regulate the sensitivity of neurons to glucocorticoids (18)
. Despite its hypothalamic localization, there are no reports as yet that MIF has a role in blood pressure control or water balance. There is growing evidence that MIF can exert intracellular enzymatic actions via its intrinsic tautomerase or thiol-protein oxidoreductase (TPOR) activities (20
, 21)
. Our studies have demonstrated that MIF is produced intracellularly in normotensive rat hypothalamic/brain stem neurons in culture and in rat hypothalamus in vivo in response to ANG II (AT1-R-mediated) stimulation (22
, 23)
. Further, MIF acts intracellularly to depress the acute AT1-R-mediated chronotropic actions of ANG II on hypothalamic neurons in culture, via a TPOR-mediated mechanism (23)
. Based on this, we hypothesized that MIF serves as a counter regulator of the neuronal actions of ANG II via AT1-R, and we decided to determine whether MIF can regulate the acute CNS actions of ANG II on neuronal discharge and blood pressure. Here we demonstrate that ANG II increases the expression of MIF within rat PVN, a critical sympathetic relay point in the CNS-mediated effects of ANG II on cardiovascular functions (3)
. We further demonstrate that transient overexpression (adenoviral-mediated) of MIF within the PVN attenuates the pressor and dipsogenic effects of centrally injected ANG II. Lastly, we show that intracellular application of MIF blunts electrophysiological responses to ANG II in PVN sympathetic-regulatory neurons via a TPOR-mediated mechanism. These data support our hypothesis and provide the first indication that MIF within the PVN can operate as a novel intracellular negative regulator of the cardiovascular and body fluid regulatory actions of ANG II.
MATERIALS AND METHODS
Animals
For the experiments described here, we used a total of 80 adult male Sprague-Dawley (SD) or Wistar Kyoto (WKY) rats, purchased from Charles River Farms (Wilmington, MA, USA). All experimental procedures were approved by the University of Florida or the University of Texas Health Science Center at San Antonio Institutional Animal Care and Use Committees.
Materials
Rabbit anti-rat MIF antibody (Ab) was purchased from Torrey Pines Biolabs, Inc (Houston, TX, USA). Mouse recombinant MIF (rMIF) was prepared as the native protein from an Escherichia coli expression system and purified free of endotoxin by C8 chromatography as described previously (24)
. Mouse MIF differs from rat MIF by a single amino acid substitution (mMIF: asn; ref 54; rMIF Ser-54) that has to date not been found to influence the bioactivity or immunoreactivity of the protein’s in different murine bioassays (24
25
26)
. The peptide fragments of rat MIF, MIF-(50–65), and C57S/C60S-MIF-(50–65) (26)
were synthesized by the Tufts University Core Facility (Boston, MA, USA). These peptides were unmodified at the N and C termini. MIF-(50–65) displays TPOR activity, while C57S/C60S-MIF-(50–65) lacks TPOR activity (25)
. Structures were as follows: MIF-(50–65): [H]-F-S-G-T-S-D-P-C-A-L-C-S-L-H-S-I-[OH]; and C57S/C60S-MIF-(50–65): [H]-F-S-G-T-S-D-P-S-A-L-S-S-L-H-S-I-[OH]. ANG II was purchased from Sigma-Aldrich Chemicals (St. Louis, MO, USA). Monoclonal anti-NeuN Ab was purchased from Chemicon International (Temecula, CA, USA). Alexa Fluor 594 goat anti-mouse IgG was purchased from Molecular Probes (Eugene, OR, USA).
Recombinant adenoviral constructs
Preparation of the adenoviral constructs Ad5-Syn-MIF and Ad5-Syn-enhanced GFP, which produce selective expression of MIF and GFP in neurons in vitro and in rat brain, was performed exactly as detailed previously (23)
.
CNS cannulae/microinjections
Adult male rats (250–300 g) were anesthetized with a mixture of O2 (1 l/min) and 4% isoflurane and placed in a Kopf stereotaxic frame. Anesthesia was maintained using an O2/isoflurane (2%) mixture delivered through a specialized nose cone for the duration of the surgery. An analgesic agent (buprenorphine; 0.05 mg/kg sc) was administered before waking.
Intracerebroventricular cannulae
Injections of 0.9% saline or ANG II were made via a stainless steel guide cannula (22 Gauge; Plastics One) placed into the right lateral cerebroventricle as detailed previously (27)
. Stereotaxic coordinates for the lateral cerebroventricle were as follows: –1.3 AP (bregma), 1.50 ML, and –4.50 DV (skull surface), according to Paxinos and Watson (28)
. Injections of sterile 0.9% saline (2 µl) or ANG II (10 pmol/2 µl) were made using a 5 µl Hamilton microsyringe and a stainless steel injector.
PVN cannulae
For PVN injections of losartan, a stainless steel guide cannula (26 Gauge; Plastics One, Roanoke, VA, USA) was implanted using similar procedures as for the intracerebroventricular (icv) cannulae. Stereotactic coordinates for the PVN were: –1.6 to 1.8 AP (bregma), + 2.0 ML, angled at 14°, and –8.0 DV (skull surface).
PVN microinjections
Bilateral microinjections of either of the recombinant adenoviral constructs Ad5-Syn-enhanced GFP or Ad5-Syn-MIF were performed as described recently, using the following stereotaxic coordinates: –1.6 to 1.8 AP (bregma), ±0.3 to 0.5 ML, and –7.8 DV (skull surface; ref 29
). Localization of the injection site and neuron-specific transduction was confirmed by visualization of green fluorescent protein (GFP) and immunostaining with antibodies against the neuron-specific protein NeuN, as described previously (29)
.
Ventral reuniens thalamic nucleus microinjections
Bilateral microinjections of either of the recombinant adenoviral constructs were performed into the ventral reuniens thalamic nucleus (RTN), 
0.5mm dorsal to the PVN, using the following stereotactic coordinates: –1.6 to 1.8 AP (bregma), ±0.3 to 0.5 ML, and –7.3 DV (skull surface; ref 28
).
Analysis of MIF protein in hypothalamic nuclei
Three hours after icv injection of either 0.9% saline or ANG II into WKY or SD rats, animals were euthanized and brains were removed and sectioned. The PVN, anterior hypothalamus (AH), or lateral hypothalamic area (LH) were removed by micropunch, and MIF protein expression was assessed via Western immunoblot as detailed previously (22)
. Similar procedures were used to assess the levels of MIF protein within the PVN 7 and 14 days after bilateral injections of either Ad5-Syn-enhanced GFP or Ad5-Syn-MIF at this site.
Analysis of MIF transgene in the PVN
Seven days after bilateral injections of either Ad5-Syn-EGFP or Ad5-Syn-MIF into SD rat PVN, animals were euthanized and and brains were removed and sectioned. The PVN was removed by micropunch, and MIF transgene expression was assessed by reverse transcriptase-polymerase chain reaction (RT-PCR) as follows. Total RNA was isolated from micropunches using an RNeasy kit (Qiagen, Valencia, CA). Samples were treated with DNase I, and 25 ng RNA were used to perform RT-PCR. The forward and reverse primers were designed from MIF and the vector separately to amplify a specific fragment with the transduced MIF cDNA. The primers were as follows: forward primer (vector): 5'-AGTCGTGTCGTGCCTGAGA-3'; reverse primer (MIF): 5'-AAGAACAGCGGTGCAGGTAA-3'
RT-PCR reactions were performed using a OneStep RT-PCR Kit (Qiagen) under the following conditions: 1 cycle of 50°C for 30 min, 95°C for 15 min; 35 cycles of 94°C for 40 s, 50°C for 1 min, 72°C for 1 min; and 1 cycle of 72°C for 8 min. PCR products were electrophoresed on a 2% agarose gel and identified by restriction endonuclease analysis.
Cardiovascular and water intake measurements
Measurements of mean arterial pressure (MAP), heart rate (HR), and locomotor activity were made via telemetry transducers (Data Sciences Int., St. Paul, MN) implanted into the abdominal aorta under isofluorane anesthesia as above. The procedures used were detailed by us previously (30)
. Raw data were analyzed using Dataquest IV software (Data Sciences Int.). Water intake was assessed over a 30 min recording period via gravimetric analysis (27)
.
Patch clamp recordings from PVN sympathetic regulatory neurons
Retrograde labeling
This was achieved as detailed previously (10)
. In brief, male SD rats (100–120 g) were anesthetized with pentobarbital sodium (65 mg/kg ip and placed in a stereotaxic head frame. A glass micropipette (tip: 35 µm) was filled with rhodamine-labeled fluorescent microspheres (Lumafluor Corp., Naples, FL) and lowered into the pressor region of the RVLM using the following coordinates (in mm): –10.8 AP (bregma), 1.6 ML, and –9.5–9.7 DV (skull surface). Microspheres were then injected in a 50 nl vol. delivered slowly (2 min) by pressure ejection, and the pipette was left in place for 5 min after each injection to allow for diffusion of tracer away from the injector tip. A period of 5–7 days was allowed for recovery and retrograde transport of the tracer to the PVN.
Hypothalamic slice preparation
Hypothalamic slices were prepared as described previously (10
,31)
. In brief, rats that received microinjections of rhodamine-labeled latex microspheres as above were decapitated under halothane anesthesia. The brains were quickly removed and placed in ice-cold artificial cerebrospinal fluid (ACSF) in which NaCl was replaced with equal osmolar sucrose. The buffer was equilibrated with a 95% CO2/5% O2 gas mixture. Coronal sections (300 µm thick) were cut through the hypothalamus using a vibratome (Series 1000, Technical Products, Inc., St. Louis, MO). Sections were then placed in a holding chamber containing gassed ACSF at room temperature until use (1–2 h).
Electrophysiological recordings
Whole cell patch-clamp recordings were made from PVN neurons that had been retrogradely labeled from the RVLM. Recordings were performed in current- and voltage-clamp modes. Details of these procedures, including the composition of intracellular and extracellular solutions, were published previously (10
,31)
.
Discharge responses to bath application of ANG II were recorded in current clamp mode at resting Vm. After a stable whole cell recording was obtained for a period of 5 min, ANG II dissolved in normal extracellular solution was applied to the bath in a concentration of 1 µM for 3 min. Recordings were maintained for 
30 min to observe the full time course of discharge responses. Responses to ANG II obtained with pipettes containing normal intracellular solution were compared with those recorded during intracellular dialysis of rMIF, which was added to the patch pipette at a final concentration of 0.8 nM.
Responses to locally applied ANG II were recorded 5 min after achieving stable whole-cell configuration. ANG II was delivered by pressure ejection (Dagan Corp., Minneapolis, MN) using three-barreled glass micropipettes (3B120F-4, World Precision Instruments, Sarasota, FL). For these experiments, only two barrels with nearly identical tip diameters (12 µm optical density ea.) were used. Each pipette was positioned in the slice at the depth of the recorded neuron. The tip was separated from the cell surface by a distance of 25 µm. In a previous study of PVN-RVLM neurons, local application of ANG II for 5 s was demonstrated to produce consistent responses that are dose-dependent and mediated by AT1 receptors (10)
. The maximally effective dose was determined to be 2 pmol when delivered in an average volume of 0.4 nl. By using this protocol, responses to pressure ejected ANG II were recorded in voltage-clamp mode in the presence of tetrodotoxin (TTX) at a holding potential of –60 mV and in current-clamp mode (without TTX) at resting Vm. The baseline effect was determined as the average of at least 2 nearly identical responses to "puffed" ANG II spaced 1 min apart. Responses to ANG II obtained with pipettes containing normal intracellular solution were compared with those recorded during intracellular dialysis of either MIF-(50–65) or C57S/C60S-MIF-(50–65), which were added to the patch pipette at a final concentration of 0.8 nM.
Data analysis
Data are mean ± SE. Statistical significance was evaluated with the use of a one-or two way ANOVA as appropriate, followed by a Bonferroni posthoc test to compare individual means. Differences were considered significant at P < 0.05, and individual P values are noted in the figure legends.
RESULTS
Effects of MIF in the PVN on the pressor and dipsogenic actions of CNS-applied ANG II
Administration of ANG II (icv) increases MIF levels in the rat hypothalamus (22)
. Our first objective was to determine if this increase was localized to specific hypothalamic nuclei. Figure 1
shows that icv injection of ANG II (10 pmol/2 µl 0.9% saline, 3 h) elicited a significant (P<0.001) 3-fold increase in MIF protein exclusively in the PVN, without affecting MIF levels in the AH or LH. This increase in the PVN was comparable in two different rat strains (SD and WKY; Fig. 1
). In a preliminary experiment icv injection of the AT1-R antagonist losartan (Los; 10 nmol/ 2 µl) blocked the increase in MIF expression in the PVN produced by ANG II (data not shown).
|
Considerable evidence indicates that icv-injected ANG II increases blood pressure in part by stimulating forebrain sites that subsequently promote local actions of ANG II at AT1-R expressing PVN neurons. The latter consist principally of neurons that project to the RVLM and/or spinal cord to increase sympathetic outflow (32
33
34
35
36)
. Similarly, ANG II activation of AT1-R in the PVN contributes to the drinking effect produced by SFO-injected ANG II (37)
. Thus, we next sought to establish the role of AT1-R within the PVN in the pressor and dipsogenic actions of icv-injected ANG II. Data in Fig. 2
A, Bdemonstrate that icv injection of ANG II produces a significant and dose-dependent increase in MAP and water intake, reaching a maximal effect at 10 pmol ANG II (P<0.001). Injection of Los (2.1 nmol/µl) directly into the PVN significantly (P<0.001) reduced the pressor and drinking effects produced by 10 pmol ANG II injected icv (Fig. 2C, D
). Based on the data from Figs. 1
and 2
, our subsequent studies with MIF were focused on the PVN.
|
A gene transfer approach was used to overexpress MIF within the PVN and determine its effects on the pressor and drinking responses induced by icv-injected ANG II. Unilateral microinjection of an adenoviral vector containing enhanced GFP (EGFP), driven by the neuron-specific promoter synapsin (Ad5-Syn-EGFP; 1 µl of 1x107 ifu), into the PVN resulted in highly efficient and localized gene transfer as indicated by robust green fluorescence (Fig. 3
A, B). Immunostaining with NeuN antibodies revealed colocalization of green fluorescence within neurons (Fig. 3C, D, E
), thereby confirming that the Ad-Syn vector predominantly transduces neurons. Bilateral microinjection of Ad5-Syn-MIF (1 µl of 1x107 ifu) into the PVN produced an increase in MIF transgene expression and MIF protein levels within 3–4 days, reaching maximal levels by 7 days (Fig. 4
A, B). Over the same time course, Ad5-Syn-EGFP did not alter basal MIF expression. MIF expression declined to basal levels by 14 days after transduction with Ad5-Syn-MIF (Fig. 4C
).
|
|
To test the effects of MIF overexpression in the PVN on cardiovascular and dipsogenic responses to ANG II, rats were implanted with radiotelemetry transducers in the abdominal aorta. One week later, rats received injections of either Ad5-Syn-MIF or Ad5-Syn-EGFP bilaterally into the PVN as above and a single icv cannula. A control group of rats received an icv cannula but no injection of either viral construct into the PVN. After 7 days, at which time MIF overexpression is maximal, telemetry recordings revealed that MIF overexpression in the PVN did not alter basal MAP compared with the Ad5-Syn-EGFP-treated or control (no virus) rats (Table 1
). Basal heart rate in the MIF overexpressing rats was not different from that of the control group, but was significantly greater than that observed in the Ad5-Syn-EGFP-treated rats (Table 1)
. Injection of ANG II (10 pmol/2 µl icv) into the Ad5-Syn-EGFP-treated rats 7 days after viral transduction elicited a significant pressor action compared with icv injections of 2 µl 0.9% saline. This ANG II-induced pressor action was similar in magnitude to that observed in control rats (Table 1)
. However, the pressor action of icv-injected ANG II was diminished by >50% (P<0.001) in the Ad5-Syn-MIF-treated rats, failing to reach the same maximal response as observed in the Ad5-Syn-GFP-treated or control rats (Table 1
; Fig. 5
A, B). By 14 days postviral injections, when MIF levels in the PVN return to baseline (Fig. 3C
), the ANG II-induced pressor action was not significantly different between the Ad5-Syn-MIF and Ad5-Syn-EGFP-treated rats (Fig. 5C, D
).
|
|
Overexpression of MIF within the PVN as above also significantly reduced (P<0.01) the water intake induced by ANG II (10 pmol/2 µl, icv) compared with that of control or Ad5-Syn-EGFP-treated rats (Fig. 6
A). By 14 days after Ad5-Syn-MIF transduction into the PVN, the ANG II-induced increase in water intake was not different from that observed in the Ad5-Syn-EGFP-treated rats (not shown). Recordings of locomotor activity revealed no significant differences between the three treatment groups (control, Ad5-Syn-EGFP or Ad5-Syn-MIF), either in the absence or presence of ANG II (Fig. 6B
).
|
To test the anatomical specificity of the inhibitory action of MIF, Ad5-Syn-MIF or Ad5-Syn-EGFP were injected 0.5 mm dorsal to the PVN (the RTN). Seven days later the basal MAP, and the pressor and drinking responses to icv-injected ANG II (10 pmol/2 µl) were similar in both groups of rats (Fig. 7
A, B).
|
MIF depresses electrophysiological responses to ANG II in PVN neurons
The previously observed inhibitory actions of MIF on ANG II-induced neuronal firing in hypothalamic neurons in culture were produced by increasing intracellular levels of MIF or MIF-(50–65) (23)
. To determine the effect of MIF on the response of identified sympathetic regulatory neurons to ANG II, whole-cell patch-clamp recordings were made from PVN neurons retrogradely labeled from the ipsilateral RVLM. In the first set of experiments, control responses to ANG II, recorded with electrodes filled with standard intracellular solution, were compared with those recorded using electrodes filled with rMIF (0.8 nM). Representative current clamp traces (Fig. 8
A) illustrate the effect of intracellular rMIF on the discharge response to bath applied ANG II (1 µM). The peak discharge frequency achieved under control conditions (top trace) was greater than the ANG II response recorded after 15 min of continuous intracellular dialysis with rMIF (bottom trace). The bar charts shown in Fig. 8B
demonstrate that the baseline discharge of control neurons or neurons that had been dialyzed intracellularly with rMIF was not significantly different. However, the peak discharge response produced by bath-applied ANG II in the presence of intracellular rMIF was significantly blunted (40%; P<0.01) compared with control neurons. The average latency to the peak ANG II response in control and rMIF-treated neurons was similar and averaged 13.5 ± 2.1 min. Consistent with previous reports (10
,38
,39)
, resting membrane potential under control conditions averaged –56 ± 1.1 mV; n=5) and was not significantly altered by intracellular rMIF (–54±1.2 mV; n=5). Similarly, intracellular rMIF failed to affect the average amplitude of action potentials (control: 98±4.1 mV vs. rMIF: 96±4.4 mV).
|
Next, we tested whether the peptide MIF-(50–65), which harbors the TPOR activity of the MIF molecule, depresses the stimulatory action of ANG II on PVN neuron discharge. In whole-cell recordings, ANG II delivered locally by pressure ejection (2 pmol, 0.4 nl, 5 s) on to retrogradely labeled PVN-RVLM neurons produced a discharge response, recorded 5 min after achieving whole-cell configuration (Fig. 8C
, top). Fifteen minutes after achieving whole-cell mode, reapplication of ANG II in the presence of MIF-(50–65) (0.8 nM) in the patch pipette produced a smaller peak response (Fig. 8C
, bottom). Summary data reveal that the maximum discharge response to ANG II was significantly reduced after 15 min of intracellular dialysis with MIF-(50–65) (P<0.01). In contrast, the control response to ANG II was not altered after 15 min of intracellular dialysis with standard pipette solution lacking MIF-(50–65) (Fig. 8D
).
To further assess the role of the TPOR activity of MIF on the neuronal responses to ANG II, voltage-clamp recordings were performed during bath perfusion with ACSF containing tetrodotoxin (1.0 µM). Intracellular perfusion of either 0.8 nM MIF-(50–65) or 0.8 nM C57S/C60S-MIF-(50–65), which lacks TPOR activity, did not affect holding current [control: 7.9 ± 2.2 pA (n=8); MIF-(50–65): 8.9 ± 2.6 pA (n=4); C57S/C60S-MIF-(50–65): 13.3 ± 3.2 pA (n=4)]. However, pressure applied ANG II (2 pmol) induced an inward current that was significantly reduced (P < 0.05) after 15 min of dialysis with intracellular MIF-(50–65) (Fig. 8E
, top trace, and F). In contrast, the control inward current response to ANG II in a separate group of neurons was unaltered after 15 min of intracellular exposure to C57S/C60S-MIF-(50–65) (Fig. 8E
, bottom trace, and F).
DISCUSSION
The major findings from this study are 1) ANG II increases MIF levels in the rat PVN; 2) Adenoviral-mediated overexpression of MIF within rat PVN blunts the pressor and drinking actions induced by centrally administered ANG II; and 3) MIF, acting intracellularly via a TPOR-mediated mechanism, can depress the electrophysiological effects of ANG II at PVN-RVLM sympathetic regulatory neurons. The rationale for this study arose from previous investigations in which we had demonstrated that ANG II increases MIF expression in hypothalamic neurons in culture and in normal rat hypothalamus in vivo (22
, 23)
. Furthermore, MIF acts intracellularly to depress the acute AT1-R-mediated chronotropic actions of ANG II on hypothalamic neurons in culture, via a TPOR-mediated mechanism (23)
. Thus, we determined whether MIF can modulate the CNS actions of ANG II on neuronal discharge, blood pressure and drinking.
Since the central actions of ANG II on blood pressure and drinking involve activation of AT1-R at multiple CVO, hypothalamic, and brainstem sites, we initially investigated whether ANG II could modulate MIF levels at any of these areas. Our focus was on the hypothalamus, because this area contains MIF localized in neurons (19)
, and we had determined previously that ANG II increases MIF expression in this region (22)
. The data presented here indicate that ANG II increases MIF expression in rat PVN. This nucleus contains AT1-R and is a critical relay point in the ANG II-control of sympathetic outflow, blood pressure, and water intake (3
, 32
33
34
35
36
37)
. Furthermore, our data (Fig. 2)
indicate that blockade of PVN AT1-R prevents the pressor and drinking responses to icv-injected ANG II. Thus, we concentrated on the PVN to investigate the possible regulatory actions of MIF on ANG II-induced cellular and physiological actions.
To examine the effects of MIF on ANG II-induced physiological effects, one approach would have been to test the effects of icv treatment with ANG II (which raises MIF levels in the PVN) on the acute actions of this peptide on blood pressure and water intake. However, aside from inducing MIF expression, icv-injected ANG II also induces expression of many other factors and genes (40)
, making it difficult to ascribe any physiological changes specifically to increased MIF expression. Therefore, to produce a selective increase in neuronal MIF within the PVN, it was overexpressed via Ad5-Syn-MIF. Using this approach, we determined that increased MIF expression within PVN neurons blunts the pressor and dipsogenic actions of CNS-injected ANG II, when compared with ANG II effects in control or Ad-Syn-EGFP-treated animals. Conversely, once MIF levels within the PVN have returned to normal (basal) values, the ANG II-induced increases in MAP and water intake are restored to normal. These results parallel our in vitro findings that increased levels of MIF suppress the neuronal chronotropic actions of ANG II (23)
and provide the first in vivo indication that MIF can serve as a counter regulator of the CNS-mediated actions of ANG II on cardiovascular regulation and fluid balance. Our future studies will use an adeno associated virus (AAV2)-Syn-MIF construct to produce longer term (months) neuronal overexpression of MIF in the PVN, and also MIF knockout mice. These approaches will allow us to further explore the idea that this protein has an inhibitory role with regard to ANG II actions in the brain.
Since overexpression of MIF in PVN neurons blunted the physiological actions of CNS-applied ANG II, we determined whether MIF could modulate the electrophysiological effects of ANG II at PVN-RVLM sympathetic regulatory neurons. The data indicate that intracellular application of rMIF attenuates the stimulatory actions of ANG II (AT1-R-mediated) on neuronal discharge and inward current via a TPOR-mediated mechanism, consistent with our cell culture experiments (23)
. Collectively, the data demonstrate that MIF depresses the stimulatory actions of ANG II at PVN neurons that are involved in regulating sympathetic outflow and ultimately blood pressure.
Although these findings provide strong evidence for a novel role of MIF in regulating the CNS actions of ANG II, several questions remain. The first concerns the time course over which MIF might act in vivo, and the mechanisms involved. Based on the time course of MIF expression after icv injection of ANG II (1–3 h.; refs 22
, 23
), it appears unlikely that MIF will attenuate an initial increase in neuronal activity produced by ANG II. However, it is possible that MIF provides a longer term regulation of ANG II actions. In terms of the physiological mechanisms involved, it is well known that the PVN is a critical relay point in the control of sympathetic outflow (3)
and receives ANG II inputs from CVOs such as the SFO and OVLT (32
, 35
, 37
38
39)
. Excitation of the SFO or OVLT (via ANG II or osmotic stimuli, for example) leads to activation of AT1-R in the PVN, which then appear to transmit the excitatory drive to the RVLM and the intermediolateral column of the spinal cord, subsequently increasing sympathetic outflow and blood pressure (3
, 35
, 41
, 42)
. Thus, one possibility is that MIF acting at PVN neurons interrupts the ANG II-induced activation of sympathetic efferents to the RVLM and IML and so reduces the ANG II-induced pressor effect. However, it is also known that activation of other AT1-R containing PVN neurons elicits release of vasopressin (AVP) from terminals at the posterior pituitary and that AVP elicits a pressor effect (1
, 2)
. Considering that the Ad5-Syn-MIF used here does not allow selective transduction of MIF into functionally distinct populations of PVN neurons, it is also possible that the inhibitory action of MIF involves inhibition of ANG II-induced AVP secretion.
Another question raised by the current study is whether the actions of MIF are limited to the PVN or whether ANG II actions at other AT1-R-containing nuclei are also tempered by MIF. Clearly, additional studies are needed to resolve this important question.
One further issue that bears discussion is the mechanism of action of MIF in regulating the neuronal actions of ANG II. Our data indicate that MIF is working via one of its intrinsic enzymatic activities (TPOR) to exert an inhibitory action over the effects of ANG II in vitro and in vivo. In our previous studies, we have demonstrated that viral mediated overexpression of MIF in SD rat neurons in culture prevents the ANG II-induced accumulation of ROS (23)
. Since the centrally mediated pressor and dipsogenic actions of ANG II depend on the availability of intracellular ROS (8)
, it is tempting to speculate that the inhibitory action of MIF in the brain in vivo may involve a TPOR-mediated ROS scavenging action. However, other modes of action, such as direct effects at membrane K+ and Ca2+ channels that control neuronal activity, at signaling intermediates (e.g., PKC, CaMKII, and NADPH oxidase) or at AT1-R cannot be discounted at this time, and future investigations will address these points. An extrapolation of this discussion is the possibility that MIF may serve to regulate the neural-excitatory actions of other transmitters, via a general cellular redox action.
To summarize, although it is clear that multiple factors may influence and regulate the CNS actions of ANG II, the present findings provide the first indication that MIF can provide a counter-regulatory influence over the cardiovascular and fluid intake actions of ANG II mediated through the PVN.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health Grants HL-076803 (C.S.) and HL-071645 (G.M.T.).
Received for publication February 8, 2006. Accepted for publication March 20, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  H. Li, Y. Gao, Y. Qi, M. J. Katovich, N. Jiang, L. N. Braseth, D. A. Scheuer, P. Shi, and C. Sumners Macrophage migration inhibitory factor in hypothalamic paraventricular nucleus neurons decreases blood pressure in spontaneously hypertensive rats FASEB J, September 1, 2008; 22(9): 3175 - 3185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Yao, C. Sumners, S. T. O'Rourke, and C. Sun Angiotensin II increases GABAB receptor expression in nucleus tractus solitarii of rats Am J Physiol Heart Circ Physiol, June 1, 2008; 294(6): H2712 - H2720. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Matsuura, R. A. Harrison, A. D. Westwell, H. Nakamura, A. E. Martynyuk, and C. Sumners Basal and angiotensin II-inhibited neuronal delayed-rectifier K+ current are regulated by thioredoxin Am J Physiol Cell Physiol, July 1, 2007; 293(1): C211 - C217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Sun, H. Li, Y. Gao, T. Matsuura, P. A. Upchurch, M. K. Raizada, and C. Sumners Lack of Macrophage Migration Inhibitory Factor Regulation Is Linked to the Increased Chronotropic Action of Angiotensin II in SHR Neurons Hypertension, March 1, 2007; 49(3): 528 - 534. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
