Glutamatergic input in the PVN is important in renal nerve response to elevation - 《American Journal of Physiology》美国生理学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 32 下一页

返回列表

查看: 594254\|回复: 313	go Glutamatergic input in the PVN is important in renal nerve response to elevation [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:46 |只看该作者 |倒序浏览 |打印

作者：Emilio Badoer, Chi-Wai Ng,  Robert De Matteo作者单位：School of Medical Sciences, Royal Melbourne Institute of TechnologyUniversity, Bundoora 308 Melbourne, Victoria, Australia 7 x* x2 J( n( {# V1 Y* o; w



                     0 P% ^; c& }$ m3 [' Z

         + A( s8 |% ^  W& q


                  4 l1 E% F5 |, X6 B$ T. O9 o

      3 z1 E, t3 u& u) \' ^& S4 r$ D! g
      & l4 _, d0 n+ w* B# ]' E) m
      【摘要】; M1 o- f* u; v* G/ v$ u
   Elevations in plasma osmolality elicit reflex humoral and neural responses.The hypothalamic paraventricular nucleus (PVN) is important in humoralresponses. We have investigated whether the PVN contributed to the renal nervereduction that is normally elicited by increased plasma osmolality in theconscious rabbit. Renal sympathetic nerve activity (RSNA) was monitored after an intravenous infusion of hypertonic saline (1.7 M NaCl, 2 ml/min for 7 min).The responses were examined in animals microinjected with muscimol (10 nmol)into, and outside, the PVN to acutely inhibit neuronal function or withkynurenate (25 nmol) to block glutamate receptors. Compared with vehicle, themaximum reduction in RSNA elicited by hypertonic saline was significantly less with muscimol or kynurenate pretreatment into the PVN. A similar study withkynurenate was also performed in sinoaortically denervated rabbits, andsimilar effects were observed. The effect was specific to the PVN becausemicroinjections of the drugs outside the PVN had no effect on the response.The findings suggest that excitatory inputs into the PVN may be important inthe neural responses elicited by elevations in plasma osmolality. 6 z+ l- W  }! @4 S6 _$ J
      【关键词】 glutamate paraventricular nucleus plasma osmolality renal nerve activity
   AN INCREASE IN PERIPHERAL osmolality elicits reflex changes insympathetic nerve activity, including a marked reduction in renal sympatheticnerve activity (RSNA) and changes in vasopressin and plasma renin levels( 40 ). The renal excretion ofsodium and the increase in urinary flow rate that accompany an intravenous infusion of hypertonic saline are heavily dependent on intact renal nerves( 30 ). Thus the reflexreduction in RSNA invoked by an elevation in plasma sodium is an importanthomeostatic mechanism designed to reduce the sodium load.! y6 [7 P! @  w( q$ a; i# F* G

Surprisingly, we know very little about the central pathways mediating thereflex reduction in RSNA after an intravenous infusion of hypertonic saline.However, the detection of the protein Fos, a marker of neuronal activation,after this stimulus suggests that the hypothalamic paraventricular nucleus(PVN) is likely to be an important integrative site in mediating the responses( 16, 32, 36, 41 ).
  l0 a5 i9 e, K% P3 m+ s% R: m2 r
The PVN contains distinct subdivisions of magnocellular and parvocellularneurons. Vasopressin-containing magnocellular neurons project to the posteriorpituitary and release the hormone into the bloodstream. As expected, thesemagnocellular neurons in the PVN were activated by the hypertonic saline stimulus. Additionally, many neurons in the parvocellular PVN were alsoactivated by the stimulus ( 16, 32, 36, 41 ).

Parvocellular PVN neurons project to different brain regions that are knownto be important in the regulation of the sympathetic nervous system( 2, 34, 39 ). These include theintermediolateral cell column of the thoracolumbar spinal cord, where thesympathetic preganglionic motoneurons are located( 26 ). Additionally, parvocellular PVN neurons have been shown to have direct connections to therostral ventrolateral medulla, where the tonic generation of sympathetic nerveactivity is believed to originate( 15, 33, 35 ). These pathways are likelyto mediate the changes in sympathetic nerve activity in which the PVN isinvolved.

On the basis of these anatomic connections and on the previous work thatsuggested the parvocellular neurons were excited after peripheral hypertonicsaline ( 18, 32, 41 ), we hypothesize thatexcitation of the PVN is involved in the sympathetic nerve activity responsesevoked by peripheral administration of hypertonic saline. Furthermore, wehypothesize that excitation of parvocellular neurons in the PVN is importantin mediating those responses. Surprisingly, the role of the PVN in mediatingthe renal nerve responses elicited by hypertonic saline has not been examined previously in conscious animals.

Thus the aim of the study was to determine the effect of neuronal inhibition of the PVN and the effect of blocking excitatory glutamatergicinput in the PVN on the RSNA response elicited by an intravenous infusion ofhypertonic saline. In the study, we microinjected into the PVN 1 )muscimol, the GABA agonist, to acutely inhibit neuronal function and 2 ) the glutamate antagonist kynurenate and determined the effect onthe reflex RSNA. In separate experiments, we investigated whether the drugsacted specifically within the PVN by also examining the effects of administering the drugs into regions adjacent to, but outside, the PVN. Tofurther study the role of the excitatory inputs into the PVN, we alsoperformed experiments in rabbits that underwent sinoaortic denervation (SAD),in which we investigated the effects of kynurenate on the reflex RSNA elicitedby hypertonic saline infusion.

The experiments were performed in conscious rabbits, alleviating theinterference of anesthesia that is known to dampen reflex function.

MATERIALS AND METHODS4 t% z4 Z0 }) E2 J0 g" `0 L
! d' s9 _' O2 K. d- z3 X
New Zealand White rabbits of either sex (2.5-3.0 kg) were used inthis study and were obtained from Monash Animal Services (Monash University,Clayton, Victoria, Canada) or Nanawie Stud Farm (Geelong, Victoria, Canada).All experimental protocols were approved by the Royal Melbourne Institute ofTechnology University Animal Ethics Committee and conform to the Guiding Principles for Research Involving Animals and Human Beings ( 1 ) and guidelines set up bythe National Health and Medical Research Council of Australia.7 G/ w6 X! D! I" B
# F+ c: E, L+ r$ `+ }# A0 x
Surgical Procedures
& j9 @! ?  w' X. z- X* i
Before the experiment, the animals underwent surgical procedures that wereperformed under general anesthesia using ketamine hydrochloride (40 mg/kg im)and xylazine hydrochloride (5.0 mg/kg im), with additional ketamine (20 mg/kgim) administered every 30 min. Diazepam (5 mg) was injected intramuscularly as a premedication. Analgesia (buprenorphrine HCl, 60 µg im) was administeredafter each surgical procedure to alleviate any postoperative pain, and anantibiotic (chloramphenicol, 100 mg sc) was routinely administered after eachoperation. At least 2 wk separated each surgical procedure.
( U5 A# J/ y+ P! |
Implantation of PVN guide cannulas. Under general anesthesia, thehead of the rabbit was placed in a David Kopf stereotaxic apparatus. The headclamps were modified so that a small pin protruded from each clamp to allowthe clamps to fix onto the zygomatic arch. A longitudinal incision was made onthe head to expose both the bregmoid and lambdoid sutures. The head was positioned so that bregma and lambda were on the same horizontal plane. Thehead was vertically aligned by dividing the skull into quadrants at the levelof bregma using the bregmoid and midsagittal sutures as the dividing lines. Areference point in each quadrant of the skull was identified, equidistant from the midsagittal suture, and the dorsoventral coordinate was determined. Thehead was considered level when the dorsoventral coordinate of the referencepoints on the left side of the skull was within 0.2 mm of those on the right.Subsequently, the anterior-posterior position of bregma was determined, and areference point 3 mm rostral of it was marked. This latter point subsequentlyacted as our anterior-posterior reference point because bregma was removed bydrilling the burr hole to enable placement of our guide cannula." e& ~. k1 X1 S
" V" R5 K5 o6 A$ o1 X* F3 E
A small burr hole ( 5 mm in diameter) was made in the skull centered at1.7 mm caudal to bregma. Three very small drill holes, in separate quadrantsand 1 cm from the larger burr hole, were also made to allow smalljeweler's screws to be screwed into the bone to act as anchors. The guidecannulas were positioned bilaterally, at 1.7 mm caudal to bregma (i.e., 4.7 mmcaudal to our reference point) and 0.8 mm lateral to the midline, identified as the sagittal separation of the brain hemispheres. Each guide cannula waslowered 5.9 mm from the brain surface so that the tip lay 5 mm dorsal to theanticipated position of the PVN. Dental cement was used to secure the guidecannula. Protection from accidental damage was provided by a small plasticcylinder that was embedded into the dental cement and encircled the guide cannulas. Stainless steel stoppers were placed into each guide cannula.
7 E6 i8 c; i4 B
Renal nerve electrode. The implantation of the renal nerve electrode was performed as described previously( 5 ). Briefly, the left kidneywas exposed via a lumbar incision. The sympathetic nerve going to the kidneywas cleared from the surrounding tissue and placed onto bipolar spiralelectrodes, which were sutured to the adventitial wall of the renal artery.Wacker Sil-Gel (Wacker Chemie, Munich, Germany) was used to cover the electrode and nerve to insulate them from surrounding tissue. The free ends ofthe electrode wires were buried subcutaneously on the back, and the wound wasclosed. The rabbits were used 3-4 days later.

SAD. In a separate group of rabbits ( n = 12 in total), weperformed SAD to remove the arterial baroreceptor afferent input. This procedure was performed immediately after implantation of the guide cannulas.The procedure for SAD has been described at length previously( 9, 11, 17 ). Denervation wasconsidered satisfactory when an intravenous bolus infusion of phenylephrine, which raised mean arterial pressure (MAP) by 20-30 mmHg, evoked changesin heart rate (HR) of
3 W1 D: ]0 J8 j# V  u' y) i
Minor surgical procedures. On the day of the experiment, under local anesthesia (0.5% lignocaine), an ear artery and vein were catheterized,and the free ends of the electrode wires were exposed. After the completion ofthe minor surgical procedures, at least 1 h was allowed to elapse before thestart of the experiment.! ]8 J* E9 P' {( Y# t
' g8 Z5 {4 s$ s1 q+ z
Experimental Protocols

On the experimental day, after a quiet rest period of at least 1 h,muscimol (a GABA agonist), kynurenate (a glutamate-receptor antagonist), orappropriate vehicles were microinjected (200 nl) intracerebrally into theconscious rabbit. Five to fifteen minutes later, hypertonic saline was infusedintravenously (1.7 M NaCl, 2 ml/min for 7 min). MAP, HR, and RSNA weremonitored before the intracerebral injections and for 40 min after the startof the infusion. A 1-ml blood sample was taken before the infusion and at 15and 45 min postinfusion.

Muscimol (10 nmol, 200 nl/side) was injected bilaterally into the PVN ofsix rabbits 15 min before the start of the hypertonic saline infusion on 1experimental day. On another day, vehicle (Ringer solution) replaced muscimol.The experimental days were separated by 48 h, and the administration ofvehicle or muscimol was randomized. In a separate series, performed in eightrabbits, the same protocol was followed, except the injections were centeredoutside the PVN (defined as 0.5 mm or more from the PVN boundary).
& S4 ^* v" ?2 s  Y; t" f  X5 c6 T
Similar experimental procedures were performed with kynurenate (25 nmol,200 nl/side; n = 7 normal, n = 5 SAD; pH adjusted to 7.4with 0.1 M HCl and 1.0 M NaOH) microinjected into the PVN of the consciousrabbits. In these experiments, kynurenate was administered 5 min before thestart of the hypertonic saline infusion. On a separate day in these sameanimals, vehicle (Ringer solution containing 0.1 M HCl and 1.0 M NaOH, pH = 7.4) replaced kynurenate. In separate rabbits, microinjections of kynurenateand vehicle were made into areas adjacent to the PVN ( n = 7 normal, n = 7 SAD). The dose of kynurenate was chosen on the basis ofprevious work in the rabbit( 22 ).
# c3 W6 v' s0 ]. x
The intracerebral injections were performed using a stainless steelinjection needle, which extended 5 mm past the end of the fixed guide cannula.The injection needle was connected by thin tubing to a 100-µl Hamiltonsyringe and a micromanipulator, which enabled the accurate injection of thevolume required.

Monitoring Cardiovascular Variables

Blood pressure was monitored using an indwelling arterial catheter connected to a pressure transducer. The signal was recorded using a MacLabdata-acquisition system (AD Instruments). MAP and HR were determinedelectronically using the blood pressure signal.+ {& P0 z0 s* ^

Raw RSNA was amplified using a low-noise differential amplifier (ENG models187B and 133, Baker Institute, Melbourne, Australia), filtered (band-pass,100-5,000 Hz), rectified, and integrated at 0.5-s intervals. Thethreshold was set visually to cut out background nerve activity during quietperiods between bursts. We have found that this method provides a similarestimate of noise level as after a maximum pressor response. The average integrated RSNA over 1- to 2-min periods was calculated and expressed as apercentage of the resting period before each stimulus( 5 ).% b% i# i* F7 }0 X% T0 C' t
- m4 {; e5 ]2 q' \( D
Brain Histology
; X0 @3 a  l/ s- {
After the completion of all experiments, 200 nl of rhodamine or fluoresceinbeads were injected intracerebrally as an aid in determining the site ofinjection. Rabbits were deeply anesthetized with pentobarbitone sodium (60mg/kg iv), injected with 1,000 U heparin (iv), then perfused transcardiallywith 1 liter of 0.1 M phosphate-buffered saline (pH = 7.2) followed by 1 liter of 4% paraformaldehyde in phosphate buffer. The brain was removed and storedin fixative solution containing 20% sucrose. The hypothalamus was cut intosections (40 µm thick), and every fourth serial section was taken forhistological examination. The sections were mounted onto subbed slides andallowed to dry before being counterstained with cresyl violet andcoverslipping with De-Pex mounting medium (BDH Laboratory Supplies).
2 r- x7 P$ G4 a& c# x+ O, ^
The sections were examined using light microscopy, and the site of the dyeinjection was recorded in relation to the PVN. An injection site wascategorized as "in the PVN" if it was mm from the boundaryof the PVN. The positions of the injections sites relative to the PVN weredrawn onto maps.
- I% E0 q, T8 k2 X2 I. v. N
Plasma Electrolytes
0 X3 Y: k/ F( v* X
Arterial blood samples were taken to determine the plasma osmolality andthe concentration of sodium. Samples were spun at 3,000 rpm for 10 min, theplasma was collected, and the concentration of sodium was determined byindirect ion-selective electrodes using a DADE Dimension random accessanalyzer. Osmolality was determined by freezing-point depression using anADVANCED Osmometer.
/ E1 y7 M, U8 c6 Z
Statistical Analysis

The basal resting MAP and HR levels were compared using a paired orunpaired Student's t -test as appropriate. Data were expressed as thechanges from resting levels, and comparisons between treatments were madeusing a two-way ANOVA with repeated measures. Comparisons between time pointswere made using Student's t -test and applying Bonferroni'smodification to compensate for multiple comparisons.

In the case of RSNA, it is well known that a comparison of absolute levelsof the sympathetic nerve activity between animals and between days isinappropriate because of technical reasons. Thus data were expressed as apercentage of the resting level before hypertonic saline, and the changes werecompared between treatments as described above.

RESULTS

Effect of Muscimol on Responses Elicited by Hypertonic Saline

Inside the PVN. In the conscious rabbits pretreated with vehicle ( n = 6), intravenous hypertonic saline infusion was accompanied by anacute increase in MAP of 13.1 ± 1.5 mmHg, which was maximal within 5min and returned toward resting levels over the observation period( Fig. 1 ). This was not observed on the muscimol pretreatment day. Comparison of the individual time pointsbetween the treatment days indicated that when muscimol was microinjected intothe PVN, there were significant differences at the 5- and 20-min time points( P with vehicle for each time point)( Fig. 1 ). There was nostatistically significant difference in the HR responses observed afterhypertonic saline between the muscimol and vehicle treatment days( Fig. 1 ).
3 l" a) T* ~% c/ N4 L- p+ T
Fig. 1. Change in ( ) mean arterial pressure (MAP; A ), heart rate(HR; B ), and renal sympathetic nerve activity (RSNA; C )after intravenous hypertonic saline infusion (2 ml/min for 7 min, filledhorizontal bars) in conscious rabbits ( n = 6). Before infusion, theanimals were bilaterally microinjected into the hypothalamic paraventricularnucleus (PVN) with muscimol (10 nmol/side, ) or vehicle (Ringer, )on separate days. b/min, Beats/min. * P
# [4 P+ |( G  g- C# b8 Y) [' |, B
RSNA was reduced by 73% at 5 min and 60% at 10 min after the start of theintravenous hypertonic saline infusion into the conscious rabbits that hadbeen microinjected with vehicle into the PVN( Fig. 1 ). This effect wasmaximal within 5 min of the start of the infusion and slowly returned towardthe resting level during the observation period( Fig. 1 ). In contrast, withmuscimol pretreatment, the maximum reduction in RSNA was only 43% at 5 min and41% at 10 min. These were significantly different from the changes seen on thevehicle treatment day ( P Fig. 1 ).

Outside the PVN. When vehicle was microinjected into regions adjacent to, but outside, the PVN ( n = 8), the intravenous hypertonic saline infusion was accompanied by an increase in MAP of 13.6 ± 2.7mmHg. The increase was maximal at 5 min after the start of the infusion( Fig. 2 ). A similar acuteincrease in MAP also occurred when muscimol was microinjected before thehypertonic saline (MAP increased by 11.1 ± 2.8 mmHg) ( Fig. 2 ). There was nosignificant difference between the time points. In addition, there was nosignificant difference between the 2 treatment days in the HR response afterhypertonic saline infusion ( Fig.2 ).2 E+ `5 ^8 Y. f% ^9 K
7 e! a+ \% ]3 [/ y
Fig. 2. Change in MAP ( A ), HR ( B ), and RSNA ( C ) afterintravenous hypertonic saline infusion (2 ml/min for 7 min, filled horizontalbars) in conscious rabbits ( n = 8). Before infusion, the animals werebilaterally microinjected into areas outside the hypothalamic PVN withmuscimol (10 nmol/side, ) or vehicle (Ringer, ) on separatedays.

Figure 2 also shows therewas a marked reduction in RSNA in response to hypertonic saline in the rabbitsin which intracerebral microinjections were made into regions adjacent to thePVN. The RSNA fell by 65% within 5 min of the start of the hypertonic saline infusion on the vehicle treatment day( Fig. 2 ). When muscimol wasmicroinjected, the maximal reduction in RSNA in response to the hypertonicsaline infusion was 62% at 5 min, which was similar to that seen on thevehicle treatment day ( Fig.2 ).

Effect of Kynurenate on Responses Elicited by Hypertonic Saline inNormal Intact Rabbits
0 u5 c/ p4 |5 l9 ^" x3 k7 G- I
Inside the PVN. In animals pretreated with vehicle microinjected into the PVN, the hypertonic saline infusion was not accompanied by any markedchange in MAP ( Fig. 3 ). HR wasincreased by 50 beats/min at 5 and 10 min after the start of thehypertonic saline infusion ( Fig.3 ). In these rabbits, RSNA was dramatically reduced by 64% atboth 5 and 10 min after the start of the hypertonic saline infusion( Fig. 3 ). The magnitude and the time course of the change in RSNA were similar to that observed in the animalspretreated with the muscimol vehicle.
# `7 y' m% l+ J$ G+ X5 z. V
Fig. 3. Change in MAP ( A ), HR ( B ), and RSNA ( C ) afterintravenous hypertonic saline infusion (2 ml/min for 7 min, filled horizontalbars) in conscious rabbits ( n = 7). Before infusion, the animals werebilaterally microinjected into the hypothalamic PVN with the glutamateantagonist kynurenate (25 nmol/side, ) or vehicle ( ) on separatedays. * P
9 ~$ f5 S1 z" V. p5 z
MAP and HR responses that accompanied the hypertonic saline infusion inrabbits that were pretreated with kynurenate in the PVN were similar to thoseobserved on the vehicle pretreatment day( Fig. 3 ). However, thereduction in RSNA elicited by the hypertonic saline infusion was significantlyattenuated after kynurenate pretreatment. RSNA fell by only 19% at 5 min after the start of the infusion and by only 30% at 10 min after the start of thehypertonic saline infusion ( P time point comparedwith vehicle) ( Fig. 3 ).

Outside the PVN. When vehicle was microinjected outside the PVN,hypertonic saline was accompanied by an increase in MAP that was maximal at 5min after the start of the infusion (11.7 ± 3.0 mmHg)( Fig. 4 ). HR was also elevatedby over 50 beats/min at 5 and 10 min after the start of the infusion ( Fig. 4 ). RSNA was reduced by 53% within 5-10 min after the start of the infusion( Fig. 4 ).

Fig. 4. Change in MAP ( A ), HR ( B ), and RSNA ( C ) afterintravenous hypertonic saline infusion (2 ml/min for 7 min, filled horizontalbars) in conscious rabbits ( n = 7). Before infusion, the animals werebilaterally microinjected into areas outside the hypothalamic PVN with theglutamate antagonist kynurenate (25 nmol/side, ) or vehicle ( ) onseparate days. * P P

On the day in which kynurenate was microinjected outside the PVN, thehypertonic saline infusion was not accompanied by an increase in MAP at 5 min,and this was significantly different from the vehicle pretreatment day( P Fig.4 ). An increase in HR accompanied the hypertonic saline infusion, and there was no significant difference in the magnitude or the time course ofthe HR responses observed between the kynurenate and vehicle pretreatment days( Fig. 4 ).
* S* D/ _$ o; u
The reduction in RSNA that was elicited by the hypertonic saline infusionon the kynurenate pretreatment day was of similar magnitude to that observedon the vehicle pretreatment day ( Fig.4 ). However, the time course of the response appeared longer withkynurenate pretreatment, such that the magnitude of the reduction at 40 minafter the start of the hypertonic saline infusion was 60%. This wassignificantly greater than the reduction that was observed at that time on thevehicle pretreatment day (30% reduction; P Fig. 4 ).
5 Y7 X2 Q- X# t
Effect of Kynurenate on Responses Elicited by Hypertonic Saline inSAD Rabbits: Z! u4 b$ Z! T: J
# D0 K3 x( M* ~: J; k  Y
Inside the PVN. MAP and HR responses elicited by intravenous hypertonic saline infusion were not significantly different between vehicleand kynurenate pretreatment days in the conscious SAD rabbits( Fig. 5 ). In contrast, the RSNAresponse differed markedly between the two pretreatments( Fig. 5 ). With vehicle pretreatment, RSNA fell by 50-60% after the hypertonic saline infusion,whereas with the kynurenate pretreatment, there was no marked reduction inRSNA. This difference was significantly different from the response observedwith vehicle pretreatment. ( Fig.5 ) ( P$ _4 R1 u2 K5 L9 I, J# h$ [5 d

Fig. 5. Change in MAP ( A ), HR ( B ), and RSNA ( C ) afterintravenous hypertonic saline infusion (2 ml/min for 7 min, filled horizontalbars) in conscious sinoaortcally denervated (SAD) rabbits ( n = 5).Before infusion, the animals were bilaterally microinjected into thehypothalamic PVN with the glutamate antagonist kynurenate (25 nmol/side, ) or vehicle ( ) on separate days. * P9 A" p) O, ^6 o- ?
+ a- o1 J3 D- i5 s2 z! t
Outside the PVN. When kynurenate was microinjected outside the PVN, the MAP response elicited by the hypertonic saline infusion was notdifferent from that observed on the vehicle pretreatment day( Fig. 6 ). This was similar tothe HR response, although HR appeared to remain higher longer with kynurenatepretreatment (see Fig. 6 ).Hypertonic saline elicited a marked reduction in RSNA on both the kynurenateand vehicle pretreatment days ( Fig.6 ). Thus, as was observed in nondenervated animals, kynurenate microinjected into the PVN attenuated the reflex reduction in RSNA elicited bythe hypertonic saline infusion, and this effect was specific to the PVN.( p) m$ p" }7 w

Fig. 6. Change in MAP ( A ), HR ( B ), and RSNA ( C ) afterintravenous hypertonic saline infusion (2 ml/min for 7 min, filled horizontalbars) in conscious SAD rabbits ( n = 7). Before infusion, the animalswere bilaterally microinjected into areas outside the hypothalamic PVN withthe glutamate antagonist kynurenate (25 nmol/side, ) or vehicle ( )on separate days. * P

Cardiovascular Levels Before Hypertonic Saline Infusion! C/ U0 |; h% U5 ?

Effect of muscimol. INSIDE THE PVN. MAP levels beforehypertonic saline were similar on the vehicle and muscimol pretreatment days( Table 1 ). HR before theintravenous infusion of hypertonic saline was significantly lower on themuscimol pretreatment day compared with the vehicle pretreatment day( P ( Table1 ). This difference in HR level was due to a significant reductionin HR elicited by the administration of muscimol (mean difference pre- vs.post-muscimol treatment = -54 ± 16 beats/min; P n = 6).% L$ V; G% R2 P) {- Z( t0 g
% T/ @% U- J# B% P
Table 1. Mean arterial pressure and heart rate before intravenous infusion ofhypertonic saline in rabbits pretreated with muscimol (10 nmol), kynurenate(25 nmol), or their respective vehicles microinjected inside or outside thehypothalamic paraventricular nucleus' c2 o) P* s  O1 j
8 G7 ?7 q6 d5 j. W$ F% q0 }
Although it is well accepted that it is difficult for technical reasons tocompare absolute levels of sympathetic nerve activity between animals, RSNAbefore hypertonic saline was always higher with muscimol treatment. This wasdue to the significant increase in RSNA observed after muscimol (meandifference pre- vs. post-muscimol treatment = 199 ± 58%; P n = 6). The change in RSNA reached a plateau within10-15 min after muscimol was administered. Vehicle did not significantlyaffect MAP, HR, or RSNA.

OUTSIDE THE PVN. In eight rabbits in which the intracerebral microinjections were made outside the PVN, there was no significant differencein resting blood pressure before the hypertonic saline infusion betweenmuscimol and vehicle pretreatments ( Table1 ). However, HR was significantly lower on the day that muscimolwas administered ( Table 1 ).This difference was due to the effect of muscimol. Microinjection of the drugoutside the PVN elicited a significant reduction in HR (mean difference pre-vs. post-muscimol treatment = -33 ± 8 beats/min; P  B1 ~2 D3 X" C3 i. F2 z+ X4 a
* F/ x8 @! f: N: @& T- F$ T0 n4 u
Microinjection of muscimol outside the confines of the PVN resulted in amean increase in RSNA of 77 ± 23% ( P It isnoteworthy that the magnitude of the increase in RSNA was significantlysmaller than that observed after muscimol microinjection into the PVN( P significantly affect MAP, HR, orRSNA.3 Q2 S) k+ Q, o  `# g& P
; g. D8 _# b; |! v2 n
Effect of Kynurenate0 z8 |  e) L3 \6 W

Inside the PVN. Before hypertonic saline, there were nostatistically significant differences in the levels of MAP and HR between thekynurenate and vehicle treatment days( Table 1 ). RSNA beforehypertonic saline infusion was higher after kynurenate pretreatment comparedwith the vehicle treatment, because RSNA was significantly elevated bykynurenate microinjected into the PVN (mean difference pre- vs.post-kynurenate treatment, 24 ± 9%; P affect the cardiovascular variables.
6 O5 X7 o( Z# Z+ r' T5 w
In SAD animals, MAP and HR before hypertonic saline infusion weresignificantly lower on the kynurenate pretreatment day than on the vehiclepretreatment day ( Table 1 ). Asin intact rabbits, RSNA was elevated by kynurenate (mean difference pre- vs.post-kynurenate treatment = 22 ± 8%).) t% I0 g/ |/ @  D
! F3 n* J- E$ F, N5 P3 z. ~- s
Outside the PVN. There were no significant differences in the levels of MAP and HR in both intact and SAD animals before hypertonic salinebetween the vehicle and kynurenate treatment days( Table 1 ). In intact rabbits,RSNA was elevated by kynurenate administered outside the PVN, but this effectdid not gain statistical significance (mean difference pre- vs.post-kynurenate treatment = 18 ± 10%). In addition, this increase wasnot significantly different from that observed after kynurenate microinjectedinto the PVN. In SAD animals, RSNA was not significantly affected by kynurenate (mean difference pre- vs. post-kynurenate treatment = 13 ±11%). Vehicle did not affect the cardiovascular variables( Table 1 ).

Effect of Hypertonic Saline on Plasma Sodium and Osmolality
! c1 F6 _4 w- e% T+ v
There was no significant difference in the size of the increase inosmolality elicited by hypertonic saline observed in rabbits pretreated withvehicle (average increase over 45 min = 22 ± 1 mosmol/kgH 2 O)or muscimol microinjection into the PVN (average increase over 45 min = 23± 3 mosmol/kgH 2 O) ( Fig.7 ). Plasma sodium was elevated slightly more on the muscimol treatment day ( P plasmalevels of sodium remaining significantly elevated longer compared with theRinger solution day ( P t = 45 min after thestart of the infusion). Resting levels were not significantly differentbetween the 2 treatment days.

Fig. 7. Effect of hypertonic saline infusion (1.7 M NaCl, 2 ml/min for 7 min) onplasma osmolality and sodium concentration. Plasma samples were taken before( t = 0) and 15 and 45 min after the start of the infusion. Opensymbols represent data from animals microinjected with vehicle, and solidsymbols represent data from the animals microinjected with muscimol.Microinjections were made into ( A; n = 6) or outside the PVN( B; n = 8). * P4 Q7 U' i0 f: h, ]0 b' ^9 W

When muscimol was injected into brain regions adjacent to the PVN, therewas a significantly greater increase in osmolality after hypertonic salinecompared with intracerebral vehicle pretreatment (average increase over 45 min= 25 ± 2 vs. 20 ± 1 mosmol/kgH 2 O, muscimol vs.vehicle; P 0.05 between days)( Fig. 7 ). This waspredominantly due to the greater increase in osmolality at 15 min after thestart of the infusion ( P increase in plasma sodium after muscimol pretreatment, which wassignificant at the final time points examined after the start of the infusion( Fig. 7 ). Basal levels ofosmolality and sodium were not significantly different between the 2 treatment days.

Unfortunately, technical problems prevented us from determining the plasmaosmolality and sodium levels in the majority of the rabbits used in thekynurenate study. We were able to gather results from four animals: two hadmicroinjections into the PVN, and two had microinjections outside the PVN.Similar levels of osmolality and sodium were reached after the hypertonic sodium infusion in those rabbits as in the animals that received muscimol/vehicle.3 S9 \4 E/ [$ g4 a* |7 I; A
7 {* @) |. |2 A* H/ a
Intracerebral Microinjection Sites+ I% K5 _0 v- X2 |+ B! M

For the muscimol studies, microinjections were distributed from the mid- tocaudal levels of the PVN, with the majority located at the midlevel of the PVN( Fig. 8 ). Microinjections that were found to be outside the PVN were located either dorsal or caudal to it,predominantly in the reuniens nucleus. One site was found to be in thedorsomedial hypothalamic nucleus ( Fig.8 D ).5 K; q* x$ z) [2 ~1 m+ `* [

Fig. 8. Schematic representation of coronal sections of the rabbit hypothalamusshowing the outline of the PVN., Sites in which muscimol wasmicroinjected within the PVN (only 1 site/animal is shown);, sitesoutside the PVN (only 1 site/animal is shown). Details incorporated in the insets in B and C are also shown in highermagnification ( right ). Sections were drawn from 1 rabbit, and theinjection sites are positioned in the section that most closely correspondedto that actually observed. OT, optic tract; III, 3rd ventricle; Fx, fornix;MMT, mamillothalamic tract.: a7 q% s3 _9 ]) t6 Y! P1 [" r1 b

For the kynurenate studies, microinjections into the PVN were spreadthroughout the rostral-caudal extent ( Fig.9 ) but were predominantly located at the midlevel, at similarlevels to the muscimol injections. Microinjections of kynurenate that wereoutside the PVN were located dorsal and caudal to it, but the majority wererostral., z- r# ~$ b% X0 F+ z

Fig. 9. Schematic representation of coronal sections of the rabbit hypothalamusshowing the outline of the PVN., Sites in which kynurenate wasmicroinjected within the PVN (for clarity, sites are in intact animals, andonly 1 site/animal is shown);, sites outside the PVN (for clarity,sites are in intact animals, and only 1 site/animal is shown). Detailsincorporated in the insets in B - D are alsoshown in higher magnification ( right ). Sections were drawn from 1rabbit, and the injection sites are positioned in the section that mostclosely corresponded to that actually observed.

DISCUSSION
7 P+ v4 H/ g' _9 Q1 X7 \7 M3 E8 d
The present study highlights for the first time an important role for thePVN in the renal nerve response elicited by elevations in plasma osmolality inthe conscious animal. We found that administration of kynurenate into the PVNto block glutamatergic inputs resulted in an attenuation of the renalsympathoinhibition elicited by the hypertonic saline infusion in both normalintact and SAD rabbits. Inhibition of neuronal activity in the PVN withmuscimol had a similar effect (see below). Changes in HR accompanying theelevation in plasma osmolality were not markedly affected by blockade ofglutamatergic function nor the inhibition of neuronal activity within the PVN.The study shows that the PVN plays an important role in the neuralcontribution to the response elicited by hypertonic saline, and our novel findings indicate that excitatory amino acids within the PVN are involved.; N% x- w9 J- O+ N% }

In the present study, we investigated the role of endogenous glutamatergicinput in the PVN in the renal sympathoinhibition elicited by an infusion ofhypertonic saline. We found that kynurenate microinjected into the PVNmarkedly attenuated the reduction in RSNA. This effect was restricted tomicroinjections into the PVN because it was not observed when kynurenate was microinjected into areas adjacent to, but outside, the PVN. Similarly, inanimals in which the arterial baroreceptor afferents had been removed (i.e.,SAD), kynurenate microinjected into the PVN attenuated the reduction in theRSNA normally observed after hypertonic saline infusion. Thus these findingssuggest that the PVN plays an important role in the renal nerve inhibition elicited by intravenous hypertonic saline. This concurs with studies using Fosas a marker of activated neurons, which suggest that PVN neurons are activatedby intravenous hypertonic saline ( 8, 32, 36, 41 ). The present findingssuggest for the first time that excitatory glutamatergic inputs within the PVNmediate the contribution this nucleus makes to the renal nerve response. Inlight of this, it is interesting to note that subunits of glutamate receptorsare differentially distributed in the PVN. The NMDA subunits NMDAR1 and NMDAR2and -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptorsubunits, especially GluR1, appear to be particularly concentrated in regionsof the PVN where neurons that project to the brain stem and spinal cord arepredominantly located ( 3, 4, 21, 39 ).

It is not possible from our present work to determine the origin of theglutamatergic input into the PVN. Glutmatatergic terminals within the PVN havebeen found to arise from neurons within the PVN, within adjacent hypothalamicnuclei, and from nuclei farther afield( 14 ). Neurophysiologicalstudies provide evidence for a physiological role of local excitatoryglutamatergic inputs in the hypothalamus( 10 ). Thus the glutamatergicinput contributing to the RSNA response elicited by hypertonic saline mightarise from neurons outside the PVN or from interneurons within the PVN.Further investigations will be required to examine this question.$ Q" E  v+ G, K

The microinjection of muscimol into the PVN attenuated the reduction inRSNA elicited by the intravenous hypertonic saline infusion. Muscimol inhibitsneuronal function; thus the results suggest that neurons within the PVN areimportant in mediating the renal nerve inhibition elicited by hypertonicsaline. This effect of muscimol was specific to the PVN becausemicroinjections of muscimol outside the nucleus did not result in anattenuation of the response. However, there are two issues that may interfere with the interpretation of the muscimol data. First, there were marked changesin the resting levels of RSNA induced by muscimol. The increase in RSNAelicited by the administration of muscimol into the PVN was almost a doublingof the resting RSNA. We have reported this interesting observation in a recentpublication ( 6 ). Such a changein resting RSNA may make it difficult to interpret the effect of muscimol onthe responses elicited by the infusion of hypertonic saline. However, themarked attenuation of the hypertonic saline-induced RSNA response that wasobserved after kynurenate microinjection into the PVN, in both intact and SADrabbits, strongly supports the view that the PVN is important in mediating thereduction in RSNA elicited by the infusion of hypertonic saline.

The second issue is related to the observation that in the muscimol seriesof experiments there was a brief increase in MAP that accompanied thehypertonic saline infusion in the rabbits that were microinjected into the PVNwith vehicle. This response was not observed in the animals in which muscimolwas microinjected into the PVN. Thus the attenuation in the renalsympathoinhibitory response elicited by hypertonic saline that was seen inthese animals could be interpreted as resulting from the attenuation of theacute pressor response and the resultant lack of baroreceptor-mediated renalnerve inhibition. However, several important observations in the present studyargue against this explanation. First, in the series of experiments in whichkynurenate was injected into regions outside the PVN, there was no pressorresponse accompanying the hypertonic saline infusion, but RSNA fell as incontrol animals (see Fig. 4 ).Second, hypertonic saline was not accompanied by an acute pressor responsewhen vehicle was administered into the PVN, but there was the normal fall inRSNA in that group (see Fig. 3,vehicle curve). Thus it appears that the pressor response is not a consistentobservation after hypertonic saline, but the reduction in RSNA occursirrespective of a change in blood pressure. This has been observed previously ( 8, 30 ). Third, and importantly,we found that in SAD rabbits, in which arterial baroreceptor afferents hadbeen removed, there was still a marked reduction in RSNA elicited byhypertonic saline infusion (see Figs. 5 and 6 ). Finally, the ability ofkynurenate microinjected into the PVN to attenuate the renal sympathoinhibition elicited by intravenous hypertonic saline infusion in SADrabbits suggests that the contribution of the PVN to the renal nerve responseelicited by hypertonic saline was not dependent on an intact arterialbaroreceptor reflex.

The pathways involved in the transmission of information about plasmaosmolality to the PVN and relaying the appropriate responses from the PVN tothe peripheral sympathetic nerves are not known. However, there is aconsiderable body of evidence supporting the view that peripheral osmolalitychanges are detected by sensors in the anterior wall of the third ventricle,in nuclei that lack a blood-brain barrier and that form part of the lamina terminalis ( 28, 29 ). These nuclei areactivated by elevations in plasma osmolality and are known to project directlyto the PVN ( 29, 32 ). The PVN projects tonuclei known to be important in regulating sympathetic nerve activity,including the sympathetic preganglionic motoneurons in the spinal cord and thepressor region of the rostral ventrolateral medulla( 3, 4, 34, 35, 38 ). Whether these pathwaysare mediating the renal nerve responses elicited by elevations in osmolalityawaits further study. However, recent evidence shows that sympatheticinnervation of the kidney includes polysynaptic pathways from the lamina terminalis ( 37 ), suggestingthat these osmosensitive nuclei can influence RSNA.* K9 A9 K- {6 r9 j+ I
& X# P# U" v# s7 u; u
Some studies also suggest that hepatic osmoreceptors may be activated byelevations in plasma osmolality and contribute to the renal nerve response( 23, 31 ). These may be important inpostprandial osmolality changes. It should also be noted that theaforementioned studies suggest that the reduction in renal nerve activity ismediated by a complex combination of afferents traveling in the vagus,carotid, and hepatic nerves, in which removal of all afferents was required toblock the response ( 23, 30, 31 ).* \! s; Y6 k/ e9 ~* t' a4 o' j
, f; S7 e% Q5 ~) X. g* W
Thus there may be several afferent pathways relaying information from theperipheral osmoreceptors to the central nervous system. Our present dataprovide evidence for a major role of the PVN in the central pathways utilizedin the neural responses elicited by elevations in plasma osmolality and thatglutamatergic inputs within the PVN are involved.

Accompanying the intravenous infusion of hypertonic saline was an increasein HR. This effect was not markedly altered by the administration of muscimolor by the microinjection of kynurenate into the PVN. This contrasts with theeffects of these drugs on the RSNA response. The results suggest that the roleof the PVN in the responses elicited by an infusion of hypertonic saline isnot generalized. Elevations in plasma osmolality can elicit nonuniform changesin sympathetic nerve activity( 40 ), and it would be of greatinterest in future studies to determine the role of the PVN in the responsesof sympathetic nerve activity to organs other than the kidneys.
7 N  F7 L0 Z  a$ g7 f
HR changes elicited by hypertonic saline appear to vary according tospecies and protocols used. In the rat, HR decreases, and this appears to beentirely mediated by arterial baroreceptors ( 7 ). In the rabbit, 1.5 M NaClinfusion has been reported previously to have no effect on HR, but higherconcentrations (3.3 M) reportedly lowered HR, and this effect was not mediatedby arterial baroreceptors( 30 ). In the present study, wefound that HR increased after hypertonic saline infusion in intact and SADconscious rabbits., v1 y8 G7 }8 [6 I6 G& g7 S  L7 S
% _0 C2 S3 g: a2 k% H( X
The PVN is known to be important in the humoral responses to stress andbody fluid disturbances. The PVN can also influence sympathetic nerve activity( 2, 13, 24, 25 ), but there is littleknowledge of the functional role of the PVN in the regulation of sympatheticnerve activity. The present study is the first to show that the PVN plays animportant role in the reduction in RSNA elicited by an elevation in plasmaosmolality in conscious animals. Additionally, in anesthetized SAD rats,endogenous angiotensin II within the PVN has been reported recently to contribute to the increase in RSNA elicited by intracarotid hypertonic salinein that preparation ( 12 ). Thusevidence suggests that the PVN is important in the sympathetic nerve activityresponses elicited by disturbances in blood volume ( 20 ). Taken together, thefindings suggest that the PVN may have a broad integrative role in theresponses elicited by perturbations in blood volume and osmolality bymediating the neural components as well as the humoral components of the responses involved in maintaining body fluid homeostasis.9 G: k  Q" b8 B
" S3 N: f& y0 {3 g1 E- s" B
The present study was performed in conscious animals. This is a majoradvantage in the examination of reflex responses that can be adverselyinfluenced in the presence of anesthesia. Additionally, by performing theseexperiments in conscious animals we were able to observe the effect of drugson rabbit behavior. We found that muscimol produced a hypnotic type of effect,in which the animals remained conscious but in a state of relaxation fromwhich they could easily be aroused if disturbed (e.g., by light prodding).This behavioral effect could not account for the effects of muscimol,microinjected into the PVN, on the responses elicited by hypertonic saline,because there was no influence on the RSNA response when muscimol was administered outside the PVN, but there was a similar behavioral response.

Finally, we have used hypertonic saline to raise plasma osmolality. Thisresults in an increase in plasma sodium and chloride ions and in osmoticpressure. It was not possible in the present study to determine thecontribution each makes to the effects observed.

Conclusions
: F# ?9 m; h7 P' w
We have found that the glutamate antagonist kynurenate microinjected intothe PVN of the conscious rabbit antagonizes the normal renal sympathetic nerveresponse evoked by an intravenous infusion of hypertonic saline in both intactand SAD animals. The findings suggest that endogenous glutamatergic inputs inthe PVN are important mediators of the renal sympathoinhibition initiated bythe elevation of plasma osmolality.
2 y3 a  r% x2 a/ r  R* `* K
The present findings, taken together with evidence in the literature ( 19, 27 ), suggest that the PVN isimportant in the neural responses that are elicited by volume expansion andaltered plasma osmolality. Thus the PVN appears to be an important integrative site that is involved in the neural as well as the hormonal responsesinitiated by perturbations in body fluid homeostasis.

DISCLOSURES( g/ d. S4 G; O! A- f

This work was supported by the National Health and Medical Research Councilof Australia, the National Heart Foundation of Australia, and the RoyalMelbourne Institute of Technology.8 e' v, V! O# W, ^# W$ ?
      【参考文献】
The American PhysiologicalSociety. Guiding Principles for Research InvolvingAnimals and Human Beings. Am J Physiol Regul Integr CompPhysiol 283:R281-R283, 2002.

* ], H- [# z, [( z/ D; Y

Badoer E. Cardiovascular role of parvocellular neurones in the paraventricular nucleusof the hypothalamus. News Physiol Sci 11: 43-47,1996.

7 Z5 h% M1 K% h4 J  n7 e

Badoer E,McKinley MJ, Oldfield BJ, and McAllen RM. A comparison of non-hypotensiveand hypotensive hemorrhage on Fos expression in spinally-projecting neurons ofthe paraventricular nucleus and rostral ventrolateral medulla. Brain Res 610:216-223, 1993.

Badoer E andMerolli J. Neurons in the hypothalamic paraventricular nucleus thatproject to the rostral ventrolateral medulla are activated by haemorrhage. Brain Res 791:317-320, 1998.+ T; z# u7 N( u% \8 o

8 w; B: A4 M. Q$ N, o" `
" b9 ^5 W7 e6 E9 _0 V5 T
Badoer E,Moguilevski V, and McGrath BP. Cardiac afferents play the dominant role inrenal nerve inhibition elicited by volume expansion in the rabbit. Am J Physiol Regul Integr Comp Physiol 274: R383-R388,1998.
/ n3 H+ _3 f7 x) u& ~

Badoer E, NgCW, and De Matteo R. Tonic sympathoinhibition arising from thehypothalamic PVN in the conscious rabbit. Brain Res 947: 17-24,2002.
$ Y5 t; U7 ]0 F3 H" [. v  i! O
) M* I) T1 y1 J% I; K5 |' s
- ~; F& P! C" C" k  o2 a! b
Bealer SL. Central control of cardiac baroreflex responses during peripheralhyperosmolality. Am J Physiol Regul Integr CompPhysiol 278:R1157-R1163, 2000.3 D* f8 n$ L. B9 X# H0 T" l

9 t, O1 k: I0 h( J7 O
Blaine EH,Denton DA, McKinley MJ, and Weller S. A central osmosensitive receptor forrenal sodium excretion. J Physiol 244: 497-509,1975." A! B% {8 r3 s! R( G. e. m  F
0 K" ]" H0 @  N  q- k1 S
& Q! R; a# D4 y2 [, Z/ i

Blombery PA andKorner PI. Relative contributions of aortic and carotid sinusbaroreceptors to the baroreceptor-heart rate reflex of the conscious rabbit. J Auton Nerv Syst 1:161-171, 1979.0 S( w4 J; `0 l5 N2 x
6 ]* e! j' W# Z/ Z
8 u! t* }( u- ^# l& |

Boudaba C,Schrader LA, and Tasker JG. Physiological evidence for local excitatorysynaptic circuits in rat hypothalamus. J Neurophysiol 77: 3396-3400,1997.
5 K  h( N: {2 P: a

( M# X) v+ }( H
Chalmers JP,Korner PI, and White SW. The relative roles of the aortic and carotidsinus nerves in the rabbit in the control of respiration and circulationduring arterial hypoxia and hypercapnia. J Physiol 188: 435-450,1967.

: X9 G5 d" ^! g% Q& h3 F
1 |, a% b7 o0 R
Chen QH andToney GM. AT 1 -receptor blockade in the hypothalamic PVN reducescentral hyperosmolality-induced renal sympathoexcitation. Am JPhysiol Regul Integr Comp Physiol 281:R1844-R1853, 2001.

/ M2 |+ G, Z" o! s
Coote JH, YangZ, Pyner S, and Deering J. Control of sympathetic outflows by thehypothalamic paraventricular nucleus. Clin Exp PharmPhysiol 25:461-463, 1998.0 u3 _4 L/ n# C( C% N( I
, a( U' L$ e/ N

0 D: y: n" R3 a% T) o- V
Csaki A, KocsisK, Halasz B, and Kiss J. Localization of glutamatergic/aspartatergicneurons projecting to the hypothalamic paraventricular nucleus studied byretrograde transport of [ 3 H] D -aspartate autoradiography. Neuroscience 101:637-655, 2000.2 x0 s6 M4 o& M- O
, [9 z7 W! g# n& r
1 }6 H( e4 F' P- e3 A: I
* g; A) \1 e. |  J
Dampney RAL. Functional organization of central pathwaysregulating the cardiovascular system. Physiol Rev 74: 323-364,1994.: o) e( w5 u; @6 u# c; W! \. M# n

1 H$ m) k! r: P. P7 M* U
5 }! K3 `6 \' |. L7 V
Ding YQ, Lu BZ,Guan ZL, Wang DS, Xu JQ, and Li JH. Neurokinin B receptor (NK3)-containingneurons in the paraventricular and supraoptic nuclei of the rat hypothalamussynthesize vasopressin and express Fos following intravenous injection ofhypertonic saline. Neuroscience 91: 1077-1085,1999.

; U; A( f5 r% q3 R/ B. M2 x

Dorward PK,Riedel W, Burke SL, Gipps J, and Korner PI. The renal sympatheticbaroreflex in the rabbit. Arterial and cardiac baroreceptor influences,resetting, and effect of anesthesia. Circ Res 57: 618-633,1985.

* D1 f. o" K# M! b* @1 q

Han L andRowland NE. Dissociation of Fos-like immunoreactivity in lamina terminalisand magnocellular hypothalamic nuclei induced by hypernatremia. Brain Res 708:45-49, 1996.
* p  M' R5 P8 H* s9 X. F
& D1 e  C7 W- ^+ o. \4 }6 F
) J7 N& C: e$ j$ {( U; ]
Haselton JR,Goering J, and Patel KP. Parvocellular neurons of the paraventricularnucleus are involved in the reduction in renal nerve discharge during isotonicvolume expansion. J Autonom Nerv Syst 50: 1-11,1994.
1 P# @0 ~- ]( {1 j$ r
# c+ ~# r2 m1 B) ]2 K9 |( \" T

Haselton JR andVari RC. Neuronal cell bodies in paraventricular nucleus affect renalhemodynamics and excretion via the renal nerves. Am J Physiol RegulIntegr Comp Physiol 275:R1334-R1342, 1998.2 B. X: d: t! ]9 ^: r# X

Herman JP,Eyigor O, Ziegler DR, and Jennes L. Expression of ionotropic glutamatereceptor subunit mRNAs in the hypothalamic paraventricular nucleus of the rat. J Comp Neurol 422:352-362, 2000.
. `, {$ k# D$ B1 V

% p9 O& ^% y( _- O4 x) Z7 w
Hirooka Y,Polson JW, and Dampney RAL. Pressor response from rostral dorsomedialmedulla is mediated by excitatory amino acid receptors in rostral VLM. Am J Physiol Regul Integr Comp Physiol 267: R309-R315,1994.

1 I- q  @9 P7 d  b3 j3 U
  N. F% K) s0 R
Ishiki K,Morita H, and Hosomi H. Reflex control of renal nerve activity originatingin the hepato-portal region. J Autonom Nerv Syst 36: 139-148,1991.. J0 e! J; R3 W5 |5 h! D7 b

% ]+ N# g  M7 [- g5 I1 R/ K
2 O8 i& y0 a; ^
Kannan H,Hayashida Y, and Yamashita H. Increase in sympathetic outflow byparaventricular nucleus stimulation in awake rats. Am J PhysiolRegul Integr Comp Physiol 256:R1325-R1330, 1989.$ H. l" y& k6 b5 y2 E

( w& |- q$ b5 v9 p- G! z

Kannan H,Niijima A, and Yamashita H. Inhibition of renal sympathetic nerve activityby electrical stimulation of the hypothalmic paraventricular nucleus inanesthetized rats. J Autonom Nerv Syst 21: 83-86,1987.

( o% D6 J! Q% V
2 z+ k& F7 m; s7 Y
Kuypers HGJM and Maisky VA. Retrograde axonal transport to horseradish peroxidase fromthe spinal cord to brainstem cell groups in the cat. NeurosciLett 1:9-14, 1975.* {& j. }5 f- F0 ^* _+ s% ]7 @

Lovick TA andCoote JH. Effects of volume loading on paraventriculospinal neurones inthe rat. J Autonom Nerv Syst 25: 135-140,1988.
* i( C3 h' H% L' m% W( @2 i2 X
5 B4 ^8 O- Z5 R  b8 B) N& d% e8 u
; ^2 P, P+ w8 Q; Z5 \1 ]7 E
McKinley MJ,Denton DA, Coghan JP, Harvey RB, McDougall JG, Rundgren M, and ScogginsBA. Cerebral osmoregulation of renal sodium excretion-a responseanalogous to thirst and vasopressin release. Can J PhysiolPharmacol 65:1724-1729, 1986.( H4 _. _% t- r/ A2 e/ \) J, S* g

$ K8 a5 r' C3 K* f

McKinley MJ andOldfield BJ. Circumventricular organs. In: The Human NervousSystem. New York: Academic, 1990, p.415-438.6 _3 d- K. U" Z; _

! E% p* |. a& b! s! ]# M& M! J  n

Morita H,Nishida Y, and Hosomi H. Neural control of urinary sodium excretion duringhypertonic NaCl load in conscious rabbits: Role of renal and hepatic nervesand baroreceptors. J Autonom Nerv Syst 34: 157-170,1991.+ b: x# N5 _# }: b; i+ A& M
! s* O' z  g( J0 S: ?6 E, q( N5 o
% [3 r2 _- a" |8 l0 t8 k4 W5 h

Nishida Y,Sugimoto I, Morita H, Murakami H, Hosomi H, and Bishop V. Suppression ofrenal sympathetic nerve activity during portal vein infusion of hypertonicsaline. Am J Physiol Regul Integr Comp Physiol 278: R97-R103,2000., T" U  R4 W- `

7 f! }: N8 [" q" h3 Q
Oldfield BJ,Bicknell RJ, McAllen RM, Weisinger RS, and McKinley MJ. Intravenoushypertonic saline induces Fos immunoreactivity in neurons throughout thelamina terminalis. Brain Res 561: 151-156,1991.$ I9 x% q# G4 ~

Pyner S, ClearyJ, Buchan PM, and Coote JH. Tracing functionally identified neurones in amultisynaptic pathway in the hamster and rat using herpes simplex virusexpressing green fluorescent protein. Exp Physiol 86: 695-702,2001.
; n! c/ x* m( C9 I  y8 i; n6 G
. ^; i6 q/ l; Q6 s- G+ z0 y; _

Ranson RN,Motawei K, Pyner S, and Coote JH. The paraventricular nucleus of thehypothalamus sends efferents to the spinal cord of the rat that closely apposesympathetic preganglionic neurones projecting to the stellate ganglion. Exp Brain Res 120:164-172, 1998.
8 M) d6 S  |0 P$ _7 l  i, l

Shafton AD,Ryan AR, and Badoer E. Neurons in the hypothalamic paraventricular nucleussend collaterals to the spinal cord and to the rostal ventrolateral medulla inthe rat. Brain Res 801:239-243, 1998.8 D, Y) m: V  I3 f. C) `" f
  E  A- g! o8 B4 L: A
5 h9 Z- n  q7 N% W7 a% y- y

Sharp FR, SagarSM, Hicks K, Lowenstein D, and Hisanaga K. c-fos mRNA, Fos, andFos-related antigen induction by hypertonic saline and stress. JNeurosci 11:2321-2331, 1991.

0 O7 p" K- \+ @, }, f. L
2 ~% q) ]- N+ W6 y/ \4 n" [. K
Sly DJ, ColvillL, McKinley MJ, and Oldfield BJ. Identification of neural projections fromthe forebrain to the kidney, using the virus pseudorabies. JAutonom Nerv Syst 77:73-82, 1999.& n( R6 n# m3 Z6 t) {; Q% r
, A$ L  d% T" P
- `' J: W4 Q+ r0 ]# N7 F6 E
% f/ h0 H# V' h2 [
Strack AM,Sawyer LM, Hughes JH, Platt KB, and Loewy AD. A general pattern of CNSinnervation of the sympathetic outflow demonstrated by transneuronalpseudorabies viral infections. Brain Res 455: 187-191,1989.* U$ @) A/ q4 g  c* T$ I% K
; U2 i" }/ G$ a" w0 z& L6 \

Swanson LW andKuypers HGJM. The paraventricular nucleus of the hypothalamus:cytoarchitectonic subdivisions and organisation of projections to thepituitary, dorsal vagal complex, and spinal cord as demonstrated by retrogradefluorescence double-labeling methods. J Comp Neurol 194: 555-570,1980.
5 M  }5 _- _* b  R# H" f" \
! H. f/ E$ |9 z

Weiss ML,Claasen DE, Hirai T, and Kenney MJ. Nonuniform sympathetic nerve responsesto intravenous hypertonic saline infusion. J Autonom NervSyst 57:109-115, 1996.

Xu Z, Glenda C,Day L, Yao J, and Ross MG. Osmotic threshold and sensitivity forvasopressin release and fos expression by hypertonic NaCl in ovine fetus. Am J Physiol Endocrinol Metab 279: E1207-E1215,2000.

收藏0 分享0 顶0 踩0