Participation of Fos protein at the nucleus tractus solitarius in inhibitory modulation of baroreceptor reflex response in the rat

BRAIN RESEARCH ELSEVIER

Brain Research 738 (1996) 39-47

Research report

Par icipation of Fos protein at the nucleus tractus solitaries in inhibitory modulation of baroreceptor reflex response in the rat Cheng-Dean Shih a, Samuel H.H. Chan b, Julie Y.H. Chan “ * ‘ Institute of Physiology, National Yang-Ming UniL)ersi@Taipei 11221, Taiwan ROC b Centerfor Neuroscience, Nutional Yang-Ming Unicersih. Taipei 11221, Taiwan, ROC ‘ Department of Medical Research, Veterans General Hospital-Taipei, Taipei 11217, Taiwan, ROC Accepted 18 June 1996

Abstract we investigated the physiologic role of FOS proteinat the nucleustractus solita~us(NTS) in the modulationof baroreceptor reflex (BRR) in adult, male Sprague-Dawleyrats that were anesthetized and maintained with pentobarbital sodium (40 mg/kg, i.p., with 10 mg/kg/h iv. infusionsupplements).Repeatedand scheduledactivationof the baroreceptorsby transienthypertensioninducedby iv. administrationof phenylephrine(2.5, 5.0 or 10.0 pg/kg) resulted in a significantincrease in Fos-like immunoreactivity (Fos-LI), primarily in the caudal part of the NTS. This increase in Fos-LI in the barosensitive NTS neurons was appreciably reduced by bilateral

microinjection into the caudal NTS of an antisense oligonucleotide (20 pmol, 20 n]) designed to target a region of the c-jios mRNA that flanks the initiation codon (5’-129 to 143-3’). The same treatment also discernibly enhanced the BRR response, but elicited no appreciable effect on systemic arterial pressure or heart rate. On the other hand, bilateral application to the NTS of the corresponding sense oligonucleotide (20 pmol, 20 nl) or an antisense cDNA (20 pmol, 20 nl) that targeted a different site of the c-@-mRNA (5’-135 to 149-3’) was ineffective. These results suggest that expression of the inducible c-@s gene in the NTS may represent an early step in the cascade of intracellular events that leads to long-term inhibitory modulation of baroreflex control of blood pressure. Keywords: Fos-like immunoreactivity; c-fix oligonucleotide; Nucleus tractus solitaries; Baroreceptor reflex; Rat

1. Introduction Accumulating evidence during the past few years suggests that neuronal activation initiates a cascade of intracellular events that results in the expression of the immediate–early genes, including the proto-oncogene c-jios [13]. In the central nervous system, the c-jh gene is transiently induced by numerous stimuli, including seizure, noxious stimulation, change in osmotic pressure, hypertension, hemorrhage and stress [13,25,33]. Fos protein, the product of c-fos gene expression, has accordingly been proposed to serve as an intracellular messenger that couples short-term responses at the neuronal membrane to long-term intracellular events upon external stimuli [24]. Thus, immunocytochemical detection of Fos-like proteins has been used successfully as a metabolic marker to identify neuronal groups and trace neuronal pathways that subserve specific physiologic functions [8,27]. In the area of central neural control of circulation, * Corresponding author. Fax: (886) (2) 8751562

studies exploiting immunocytochemical mapping revealed significant c-fos expression in specific brain regions that bear temporal relationship with cardiovascular challenges. As the principal terminal site of the primary baroreceptor afferents [4], the caudal nucleus tractus solitaries (NTS) has been a major focus in recent studies of this nature. For example, activation of the baroreceptors was demonstrated to induce the production of Fos protein at the NTS in both anesthetized [23,25,26] and conscious [9,20] animals. Likewise, electrical stimulation of the aortic depressor nerve [21], carotid sinus nerve [9] or carotid baroreceptors [7] resulted in an increase in the level of inducible c-fos at the NTS. The physiologic role of expression of c-fos gene at the NTS in baroreceptor reflex (BRR) control of cardiovascular functions, however, has not been fully characterized. Neuronal gene expression in the brain can be selectively blocked in vivo using antisense oligonucleotides complementary to strategically chosen sequences within the target mRNA [3,10,12,16,17]. A 15-mer antisense oligonucleotide to c-fos mRNA has been demonstrated to suppress

0006-8993/96/$15.00 Copyright 0 1996 Elsevier Science B.V. All rights reserved PII S0006-8993(96)0077 1-8

40

C.-D. Shih et al./Brain Research 738 (1996) 39-47

Fos-like immunoreactivity (Fos-LI) in the striatum in response to amphetamine [3,12,16]. The increase in Fos-LI at the rostral ventrolateral medulla, which resulted from reduction in neuronal activity at the caudal ventrolateral medulla, is similarly suppressed [31]. Functionally, a blockade of c-fos expression in the nucleus accumbens blunts cocaine-promoted locomotor stimulation [10]. The present study was carried out with two specific aims. First, to evaluate whether c-fos expression in the NTS evoked by baroreceptor activation can be blocked by antisense oligonucleotide treatment. Second, to investigate whether this blockade of c-fos expression can alter the BRR sensitivity. We found that bilateral microinjection into the caudal NTS of an antisense oligonucleotide designed to target a region of the c-fos mRNA that flanks the initiation codon (5’-129 to 143-3’) significantly suppressed the Fos-LI at the NTS to repeated baroreceptor activation. This suppression correlated positively with an enhancement of the BRR response that took place 180 min postinjection. These results suggest that expression of the inducible c-fos gene at the NTS may be involved in the long-term modulation of BRR sensitivity.

2. Materials and methods 2.1. General preparation Adult, male Sprague–Dawley rats (250–300 g) were used in the present study. Animals were anesthetized initially with pentobarbital sodium (40 mg/kg, i.p.). Routine surgical preparations included incubation of the trachea to facilitate ventilation using a rodent respirator (Harvard 683), and cannulation of the left femoral artery and vein for the measurement of systemic arterial pressure and administration of drugs. The right femoral vein was also cannulated for the maintenance of anesthesia by continuous infusion of pentobarbital sodium (10–15 mg/kg/h, iv.). Systemic arterial pressure was monitored via a pressure transducer (Statham P231D), and heart rate was determined using a cardiotachometer (Gould 20-4615-65) triggered by the arterial pressure pulses. Pulsatile and mean systemic arterial pressure, as well as heart rate, were recorded simultaneously on a polygraph (Gould ES 1000). The head of the rat was thereafter placed in a stereotaxic headholder (Kopf 1430), with the rest of the body positioned on a heating pad and elevated to a suitable position. All data were collected from animals with a maintained rectal temperature of 37°C and a steady mean systemic arterial pressure above 90 mmHg throughout the experiment. 2.2. Evaluation of baroreceptor rejlex response Similar to our previous study [28], the sensitivity of BRR response was evaluated by the slope method. We

measured the maximal reflex bradycardia in response to transient hypertension elicited by three doses of phenylephrine (2.5, 5.0 or 10.0 ~g/kg, iv.). The slope of the regression line that relates changes in these two circulatory parameters was used as the index for sensitivity of the BRR response. The order of different doses of phenylephrine, which were administered within 10 rein, was altered randomly to avoid sequential dependency on arterial pressure changes. 2.3. A4icroinjection of oligonucleotides into the caudal NTS Bilateral microinjection of sense or antisense c-fos oligonucleotide into the caudal NTS was carried out sequentially with a glass rnicropipette (50-70 pm tip), which was connected to a Hamilton microsyringe. The concentration of antisense oligonucleotide applied (20 pmol) was the same as that used to block the expression of c-fos gene at the striatum and rostral ventrolateral medulla [3,12,16,31]. The tip of the glass pipette was lowered 0.35 to 0.90 mm below the surface of the fourth ventricle, at –0.50 to +0.50 mm from, and 0.35 to 0.50 mm lateral to, the obex. This allowed us to deliver the oligonucleotide to sites in the NTS where the baroreceptor afferents terminate [4]. The volume of rnicroinjection was restricted to 20 nl to avoid the confounding effect of volume artifact. Three oligonucleotides [6,12] prepared by Quality Systems (Taipei, Taiwan) were used in this study. These included an antisense cDNA (AS1) designed to target a region of the rat c-fos mRNA that flanks the initiation codon (5’-129 to 143-3’), with a sequence of 5’-GAACAT-CAT-GGT-CGT-3’; and its corresponding sense (5’ACG-ACC-ATG-ATG-TTC-3’) oligonucleotide. A second antisense oligonucleotide (AS2) was designed to target the initiation site and a portion of the coding sequence (5’-135 to 149-3’), with a sequence of 5’-ACC-CGA-GAA-CATCAT-3’. The sense sequence showed no discernible complementarily to any part of the c-fos mRNA. None of the oligonucleotides showed any significant complementarily to any other gene sequence in the GenBank database [3,12,16]. All oligonucleotides were phosphorothioated in all positions, and were diluted in artificial cerebrospinal fluid (aCSF) at pH 7.4. 2.4. Immunocytochemical staining for Fos-like protein At the conclusion of each experiment (180 min after individual treatments), each animal was perfused intracardially consecutively with warm heparinized isotonic saline solution (500 ml), and ice-cold 3.8% paraformaldehyde in 0.1 M phosphate buffered saline (PBS) at pH 7.2 (500 ml). The brain was removed and postfixed by submersion in the latter solution overnight at 4°C, followed by cryoprotection with 30% sucrose in 0.1 M PBS at 4“C. Frozen serial transverse sections of the medulla oblongata were cut on a

C

Shih et al. /Brain Research 738 (1996) 39-47

cryostat (Leitz Cryostat 1720) at 20 ~m, and were collected in 0,1 M PBS. Free-floating sections were processed for Fos-LI based on our modification [19] of the peroxidase–antiperoxidase method described by Stemberger [30]. Briefly, sections were washed thoroughly in 0.1 M PBS, followed by 0.1 M PBS containing 370 normal rabbit serum (Sigma Inmmnochemicals), 0.590 gelatin and 0.0270 Triton X-1OO.They were then incubated for 48–72 h at 4°C in 0.1 M PBS containing a polyclonal sheep anti-Fos antiserum raised against Fos protein (Cambridge Research Biochemical OA-11-824; 1:6,000) and 1% normal rabbit serum. Sections were subsequently placed in 0.1 M PBS containing rabbit anti-sheep IgG (1: 16, Jackson Immuno Research) and IYo normal rabbit serum for 60–90 min at room temperature. Visualization of the immunoreactive product included incubation in sheep peroxidase anti-peroxidase complex (1: 160, Jackson Immuno Research) for 30 rein, followed by reaction with glucose oxidase and diaminobenzidine using a nickel/cobalt intensification procedure [22]. Sections were mounted on glass slides subbed with gelatin and air dried. To localize the distribution of Fos-LI with reference to anatomic structures, sections were counter-stained with IYo Neutral red. In control experiments, some sections were incubated without the primary antiserum or substituting Fos antiserum with normal sheep serum. No specific immunoreactivity was observed in these control sections. 2.5. Quantification of Fos immunoreactiuity in the NTS Tissue sections were examined under bright-field microscopy (Olympus BH-2) to localize Fos-LI in the NTS. The criterion for identification of Fos-LI neurons was a distinctly stained nucleus [25]. Nuclei within the confines of the NTS thus identified were counted bilaterally in each section with the aid of a camera lucida drawing attachment. A series of sections from each animal, taken at 200-pm intervals between 1 mm caudal and 0.4 mm rostral to the obex, were counted separately by two individuals, and the mean number of the Fos-positive cells in the NTS was tabulated. 2.6. Experimental protocols To delineate the physiologic role of Fos protein at the NTS in the BRR control machinery, the effect of bilateral microinjection into the caudal NTS of AS 1 oligonucleotide on BRR response was routinely followed for 180 min. The sensitivity of BRR was evaluated before, and at 25–35, 55–65, 85–95, 115–125 or 175–185 min postinjection. In a separate series of experiments, we also followed the effect of microinjection of AS1 oligonucleotide to the bilateral NTS on systemic arterial pressure and heart rate. To ascertain the specificity of the effects elicited by the antisense oligonucleotide, five series of control experi-

41

ments were carried out. First, bilateral microinjection of two oligonucleotides into the caudal NTS served as the treatment control. We employed the corresponding sense oligonucleotide and an antisense cDNA (AS2) that targeted a different site of the c-fos mRNA. Both oligonucleotides have been shown to be ineffective in blocking the expression of Fos-LI in the striatum following the administration of amphetamine in rats [12]. Second, repeated activation of the baroreceptors by the same dosing scheme with phenylephrine in animals that received only incubation of the trachea and cannulation of the femoral artery and veins served as the baroreceptor activation control. Third, iv. injection of saline, which replaced phenylephrine, in animals that received microinjection of aCSF into the bilateral NTS served as the volume and vehicle control. Fourth, animals that were surgically prepared and placed in the stereotaxic headholder, without subsequent experimental treatments, served as the sham control. Fifth, animals that were only maintained under pentobarbital anesthesia served as the anesthetic control. All groups of animals were immediately processed for immunocytochemical staining for Fos-LI at the end of the experiment. Additional animals that were anesthetized and immediately sacrificed provided the background levels of immunoreactive Fos-like protein in the NTS. 2.7. Data analysis All values are expressed as mean+ S.E.M. The number of Fos-positive neurons at each level of the NTS under various treatments was compared using one-way analysis of variance (ANOVA). This was followed by the Scheffe multiple range test for a posteriori multiple comparison of means. The temporal effect of treatments on various responses was assessed by two-way ANOVA with repeated measures. This was followed by the Scheffe multiple range test for a posteriori comparison of means at corresponding time intervals. In both analyses, P <0.05 was considered to be statistically significant.

3. Results 3.1. Effect of antisense oligonucleotide to c-fos mRNA on expression of Fos-LI in the NTS evoked by baroreceptor activation Repeated and scheduled activation of the baroreceptors for 180 min in control animals induced significant Fos-LI in the NTS (Fig. 1A and Fig. 2A). Most NTS neurons that expressed Fos-LI were distributed bilaterally and were concentrated primarily in the NTS caudal to the obex (Fig. 3). Within the NTS, Fos-positive cells were mainly found in the medial and commissural subnuclei, extending from the rostral extent of the pyramidal decussation to the level of area postrema. Within the medial and commissural

42

C.-D. Shih et al./Brain Research 738 (1996) 39-47

Fig. 1. Representative photomicrographs showing the distribution of Fos-like immunoreactivity (Fos-LI) at the caudal NTS in animats that were subject to repeated and scheduled transient hypertension for 180 min (A), and in animals that additionally received bilateral microinjection into the caudal NTS of a sense oligonucleotide (B) or an antisense cDNA (AS I) targeted against the initiation codon of the c-fos mRNA (C). Panel (D) was taken from the same brain stem section as in (C), and demonstrates the presence of Fos-LI at the caudal ventrolateral medulla. Calibration bar: 100 #m. AP, area postrema; NRL, nucleus reticularis lateralis; NTS, nucleus tractus solitaries; ts, tractus solitaries; X, dorsal motor nucleus of vagus.

subnuclei, a significant number of Fos-LI nuclei was observed in regions dorsomedial to the tractus solitaries (Fig. 1A). In animals that received microinjection into the bilateral NTS of a sense oligonucleotide (20 pmol), repeated and scheduled baroreceptor activation for 180 tin also induced Fos-LI expression in the NTS (Fig. IB and Fig. 2B). The distribution and number of Fos-positive neurons in all representative levels of the NTS were essentially the same

as animals that received baroreceptor activation alone (Fig. 3). Bilateral microinjection of an antisense oligonucleotide targeted against the initiation codon of c-fos mRNA (AS1, 20 pmol) into the caudal NTS significantly reduced (Fig. IC and Fig. 2C) Fos-LI at the NTS in response to 180 min of repeated and scheduled baroreceptor activation. This reduction extended across the rostral–caudal levels of the NTS examined (Fig. 3). It should be mentioned that Fos-LI

C.-D.

wet

A

aL/Bnrimaaeadl

0

738 fmw3%47

43 II

G

P .

-0.8

*

“0°

“’s NTS

.8

-0.4

.

v

:

““

X11

*

.

0

●

q“

0.4

“.

.

00

8

.

“p

Fig. 2. Distribution of Fos-likeimmunoreactivity in fourrepresentative rostral-caudats.wtionsof theNTSin animalsthatweresubjectto repeatedand intothe caudrdNTSof a sense scheduledtransienthypertension for 180nrin(A), and in mimals that additionally received bilateralmicroirrjection oligonucleotide (B),or an antisensecDNAtargetedagainsttheinitiationcodon@Sl) (c) or a differentregion@S2)@) of thec-fo~mRNA.Eachdot represents oneFos-positive nucleus.Numbersindicatethedistancetotheobex.AP,_pwi@&sM;NTS,nucleustractussotitarius;ts,tractussolitaries;X, dorsal motor nucleus of vagus;XII,hypoglossal nucleus.

was still demonstrated in the caudal (Fig. ID) and rostral ventrolateral medulla in the antistmse-treated animals. This ascertained that the appreciable decrease in Fos-LI at the NTS after treatment with this antisense oligon@eotide was not due to false-negative reaction. On the other hartd, microinjection into the bilateral NTS of a second antisense cDNA that targets a different site of the O$W mRNA (AS2) resulted in findings that were reminiscent of-those obtained from treatment with the sense oligmtuckmtide (Fig. 2D, Fig. 3). In animals that received repeated iv. injection of saline and microinjection of aCSF into the caudal NTS, there was

r= ~

[18P]

~

aCSF+saline sense+ [l BP]

m

AS1 +[IBP]

~

AS2+[I

only I&or demonstration of Fos-LI in the NTS, which r~~~, fi~ 3.2 * 0.3 to 9.8 + 0.9 nuclqi per section. More iin#xtantly, the Fos-positive neurons did not exhibit t.opo~phic distribution within the rostral-caudal extent of the NT’S,@t we examined (Fig. 3). Increased Fos-LI at the NTS was observed in sham-control animals that received only surgical p~paration and were placed in the stereotaxic h@holder (Fig. 4). The level of Fos-LI, however, was s@ificantly less than that evoked by baroreceptor activa$qn. In addition, there was also a lack of topographic &q@@on of Fos-positive neurons amonqst the rOstrSlcauckdextent of NTS examined. We also fo~nd that Fos-LI was below detection limit ( <1 nucleus per section) at all level of the NTS examined in animals sacrificed immedi-

BP]

*

m

bosol

m

anesthesia

~ ~

shorn ASl[non-NTS]+[TBP]

* .

-1.0

-0.s

-0.6

-0.4

-0.2

0.0

0.2

0.4

DISTANCE FROM OBEX (mm)

at eightrostral-caudal levels of the NTS in animalsthat were subjectto repeatedand scheduled transienthypertension for 180ruin([tJW, ~= 6), md in mi~s Mat additionally receivedbilateralrnicroirrjection into the calal NTS of a

Fig. 3. Distribution of Fos-positive neurons

sense oligonucleotide (n= 8), or an antisense cDNA targeted agaiust $& initiation codon (AS1) (rI = 10) or a different region (AS2) (n =-8) of tha c-jm nrRNA. Animals that received bilateral rnicroinjecti~n into the NTS of aCSF and scheduled iv. injection of saline (n= 3) served as the volume and vehicle control. values arc presented as mean+S.E.M. * P <0.05 vs. aCSF or sharn-contiol group (Fig. 4) and “ P < 0.0S vs. sense treatment group in the Scheffe analysis.

0

-1.0 -0,8 -0.6 -0.4 -0.2

0.0

0.2

0.4

DISTANCE FROM OBEX (mm)

Fig.4. Distribution of Fos-positive neurons at eight rostrat-caudal levels of the NTS in animals that were sacrificed immediately after pentobar&tal anesthesia (basrd, n = 3), received only continuous iv. infusion of pen@Wbr“tal sodium (20 mg/kg/h) for 180 mirr (anesthesia, n = 3), subject to surgicat preparations and placement in stereotaxic headholder without femher experimental procedures (sham, n = 3); or received micr@ irrjectiotr of AS1 oligonuckotide to sites outside the confines of the caudalNTS(non-NTS, n = 5). Values are presented as mean+ S.E.M. * P <0.05 vs. anesthesia group in the Scheffe analysis.

C.-D. Shih et al. /Brain Research 738 (1996) 39-47

44

ately after pentobarbital anesthesia (Fig. 4). Fos-positive neurons were also scarce and distributed sporadically amongst different levels of the NTS in animals maintained under pentobarbital anesthesia for 180 min (Fig. 4). Histologic verifications revealed that the above findings were obtained from animals in which the core of the microinjection sites was localized in the caudal NTS. Microinjection of the same amount of antisense oligonucleotide into more rostral part of the NTS (0.8 to 1.0 rostral to the obex) or areas adjacent to the NTS elicited no discernible effect on the Fos-LI in caudal NTS evoked by baroreceptor activation (Fig. 4).

I

1

[

1

T-/~

3.2. Effect of blocking the expression of inducible c-fos in the NTS on the BRR response In comparison to control animals, bilateral rnicroinjection of AS1 oligonucleotide into the caudal NTS enhanced the BRR response (Fig. 5). The trend of enhancement began 60 rein, became significant at 90 rein, and lasted at least until 180 min postinjection. Linear regression evaluation further revealed that, 180 min after AS1 oligonucleotide treatment, the enhancing effect on BRR response and the suppressive effect on Fos-LI in the eight levels of NTS we examined were positively correlated (y = 75.3 + 0.1 .x, r = 0.907, n = 10). on the other hand, bilateral microinjection into the caudal NTS of the corresponding sense or AS2 oligonucleotide, similar to animals that received scheduled transient hypertension, resulted in no discernible effect on BRR sensitivity (Fig. 5). In addition, rnicroinjection of AS1 oligonucleotide into more rostral part of (i.e., 0.8 to 1.0 mm rostra] to the obex), or areas adjacent but

m

[(BP]

~

sense+

m

AS1

~ ~

200 –30

O

30

60

90

120

180

POSTINJECTION TIME (rein) Fig. 6. Time-course changes in mean systemic arterial pressure (MSAP)

or heart rate (HR) in animals that received bilateral microinjection into the caudal NTS of a sense (n= 5) or AS I (n = 6) oligonucleotide. Values are presented as mean f S.E.M. No significant difference between groups (P> 0.05) in two-way ANOVA with repeated measures.

outside of, the NTS elicited no appreciable effect on the BRR response (Fig. 5). [l BP]

3.3. Effect of blocking the expression of inducible c-fos in the NTS on systemic arterial pressure and heart rate

+[TBP] As2+[rBP] AS I [.o.-NTs]+[I

BP]

In separate groups of animals, bilateral microinjection into the caudal NTS of either sense or AS 1 oligonucleotide produced no significant effect on baseline systemic arterial pressure or heart rate during the 3-h postinjection period (Fig. 6).

.

25-35

A .

55–65

8

POST INJECTION

i5–1 25

1-

4. Discussion

‘-

TIME (rein)

1

85

Fig. 5. Time-course changes in baroreceptor reflex (BRR) response to repeated and scheduled transient hypertension for 180 min ([‘TBp], n = fj), and in animals that additionally received bilateral microinjection into the caudat NTS of a sense oligonucleotide (n = 8), an arrtisense cDNA targeted against the initiation codon (AS1) (n = 10) or a different region (ASZ) (n = 8) of the c-jbs mRNA, or AS1 oligonucleotide to sites outside the confines of the caudat NTS (non-NTS, n = 5). Values are presented as mean+ S.E.M. * P <0.05 vs. [tBP] group in the Scheffe analysis.

Based on immunocytochemical staining coupled with physiologic evaluation, -the present study demonstrated that expression of the inducible c-fos gene in the NTS was related to a reduction in the sensitivity of BRR response. We demonstrated that bilateral microinjection into the caudal NTS of a 15-mer antisense oligonucleotide targeted against the initiation codon of C--OSmRNA (AS1) significantly reduced the Fos-LI evoked in the caudal subnuclei of this nucleus by repeated baroreceptor activation. Block-

C.-D. Shih et al. /Brain Research 738 (1996) 39-47

ade of the expression of c-jios gene in the NTS also enhanced the BRR response. At 180 min following treatment, the reduction in Fos-LI in the caudal NTS correlated positively with the augmentation in BRR sensitivity. The immediate-early gene c-jiis and its protein product Fos are integral components of the complex cellular mechanisms that link extracellular stimulation to long-term changes in cellular activity within neurons [24]. Numerous recent studies [8,27] demonstrated that immunocytochemical staining for Fos-containing neurons provides a powerful tool to delineate specific cell populations and neuronal pathways that subserve specific physiologic functions. Based on the same approach, the present study demonstrated an increase in neuronal expression of c-fos gene in the caudal NTS in response to repeated and scheduled baroreceptor stimulation. Further survey of the distribution of Fos-LI revealed that a large population of barosensitive neurons was located in the medial and commissural nuclei of the caudal NTS [7,9,20–22]. This distribution of Foscontaining neurons also overlapped with the terminal fields of baroreceptor afferent fibers in the NTS [4]. Together, our results strongly suggest that these Fos-positive NTS neurons may represent the relay neurons in the BRR pathway. A number of recent studies reported the use of antisense oligonucleotides to suppress gene expression in the central nervous system [3,10,12,16,17]. Furthermore, direct application of antisense c-fos oligonucleotide into specific regions of the brain produces discrete, reversible inactivation of c-fos gene in a highly selective manner [3,12,16,17]. We demonstrated in the present study that rnicroinjection into the caudal NTS of a 15-mer antisense oligonucleotide targeted against the initiation codon of c-fos mRNA (AS1) significantly reduced the number of Fos-positive NTS neurons in response to. baroreceptor activation. Since a closely related sense oligonucleotide and an antisense cDNA that targeted a different site of the c-fos mRNA (AS2) did not affect Fos-LI in the NTS to the same stimulus, it appears that the blunting effect of AS1 oligonucleotide was related ,to its complementarily with c-fos mRNA. The present results also revealed that significant suppression of Fos-LI in’ the NTS occurred only 180 rnin after AS1 oligonucleotide treatment. This finding was at variance with previous reports [3,31] of a longer (e.g. 6–8 h) onset latency. Nonetheless, our observations were in agreement with biochemical results that showed rapid uptake of antisense oligonucleotides by neurons in the vicinity of injection site within 15–30 min postinjection [34,35]. Another novel finding in the present study was that blockade of c-fos expression in barosensitive neurons of caudal NTS with AS1 oligonucleotide also enhanced the BRR response. In view of the suggested role of Fos protein as a transcription factor in stimulus-transcription coupling [24], we speculate that the expression of C+OSgene in the barosensitive NTS neurons may lead to the induction of other genes that encode neurotransmitters and/or neu-

...-.—..———— .. - ——..——-

45

ropeptides. These chemical signals may in turn exert a long-term inhibitory modulation on the sensitivity of BRR response. In support of this speculation, baroreceptor activation results in the expression of Fos-LI in neurons in the NTS [20,25,26] and ventrolateral medulla [25,26,32] that are also immunoreactive to tyrosine hydroxylase and/or phenylethanolamine-N-methyltransferase. A recent study [15] reported that seizure causes the expression of c-fos in NTS cells that contain prodynorphin mRNA, with a subsequent increase in prodynorphin mRNA levels in the same area. At the cellular level, induction of c-fos produces a long-term up-regulation of calcium channel currents in rat phaeochromocytoma cells to membrane depolarization [1]. Whether similar mechanisms may underlie the long-term modulation of BRR sensitivity following c-fos expression in the barosensitive NTS neurons, however, require fi@her elucidation. Apart from using sense and AS2 oligonucleotideas our treatment control, the results from three separate series of experiments ascertained the specificity of our observed effect of the antisense c-fos oligonucleotide. First, the lack of noticeable effect in animals that received bilateral rnicroinjection of aCSF into the caudal NTS, together with repeated and scheduled iv. injection of saline, precluded the confounding effects of solvent and transient increases in volume in brain tissues or systemic circulation. Phenylephrine does not cross the blood–brain barrier [11], thus minimizes the possibility of a direct effect on NTS neurons. Second, we did observe an increase in Fos-LI at the NTS in our sham-control animals, albeit at a level that was significantly less than that evoked by baroreceptor activation. In addition, there was a lack of topographic distribution of Fos-positive neurons amongst the rostral-caudal extent of NTS examined. Thus, it is deemed unlikely that stress associated with surgical preparations and placement in the stereotaxic headholder alone can account fully for the Fos-LI detected in the NTS of animals that received repeated and scheduled baroreceptor activation. Third, under the same anesthetic maintenance, microinjection of sense or antisense c-fos oligonucleotide into the caudal NTS differentially affected the number and topographic distribution of Fos-positive NTS neurons evoked by baroreceptor activation. Continuous iv. infusion of pentobarbital sodium for 180 rein, on the other hand, resulted in minor expression of c-fos. As such, although anesthetic agents induce or suppress the expression of c-fos in neurons depending on the region examined [18,33], the impact of anesthesia on our present results was considered nominal. The use of c-~os antisense oligonucleotide to block gene expression and to delineate potential role of c-fos to specific stimulation in the central nervous system has attracted substantial enthusiasm. However, several potential limitations to this approach [2,5,14] should be taken into consideration. First, our AS1 oligonucleotide specifically inhibited the expression of the target gene. Several

46

C.-D. Shih et al. /Brain Research 738 (1996) 39-47

studies [5,14,29] reported that an antisense oligonucleotide with a length of 11–15 nucleotides is able to bind selectively to a single RNA species in the cell. The antisense oligonucleotide we used in this study is designed to target a region of the c-fos mRNA that flanks the initiation codon (5’-129 to 143-3’). Hooper et al. [12] reported recently that it is able to eliminate the expression of c-jos gene to administration of amphetamine into the striatum. The same c-fos antisense oligonucleotide also reduces Fos protein without affecting another immediate-early gene product, c-Jun [17]. Second, we have chosen the appropriate control sequences of oligonucleotides. Sense oligonucleotides complementary to the antisense sequence [10,16,17,31], oligonucleotides of the same length as the antisense but targeted a different site of the mRNA [6,12], composed of a random mixture of all four nucleotides [12] or a mismatch of two or three base pairs [2], have all been used as control to verify the biologic activity of the c-fos antisense oligonucleotide. Whereas a suitable choice of control sequence for antisense oligonucleotide is still debatable, our present results demonstrate that a sense oligonucleotide complimentary to c-fos antisense sequence, and an antisense cDNA that targeted the initiation codon and portion of the coding sequence (5’-135 to 149-3’) did not affect the expression of c-fos in the NTS evoked by baroreceptor activation. The same treatment also resulted in minimal alteration in the BRR response. Third, there was minimal toxicity associated with the use of nuclease-resistant phosphorothioate oligonucleotide. Phosphorothioated antisense oligonucleotide was reported to be stable in brain tissue for at least 12 h after central application [3,10,12,16,17]. Several in vivo studies reported that a single injection of the phosphorothioate oligonucleotide into neural tissues does not appear to cause toxicity [2]. We also found that microinjection of the same amount of phosphorothioated antisense c-fos oligonucleotide into more rostral part of the NTS (0.8 to 1.0 rostral to the obex) or areas just adjacent to the NTS elicited no discernible effect on BBR response or Fos-LI in caudal NTS evoked by baroreceptor activation. In conclusion, the present study demonstrated that blockade of c-fos expression in the NTS with an antisense oligonucleotide targeted against a region that flanks the initiation codon of its mRNA diminished Fos-LI evoked in the same nucleus by baroreceptor activation. The same treatment also resulted in an enhancement of the BRR response. These observations support the notion that c-fos expression plays a critical role in coupling external stimuli to long-term changes in neuronal function [24]. Our findings, at the same time, indicate that rather than simply serving as an indicator for neuronal activation, expression of the inducible c-fo.s gene in the NTS may represent an early step in the cascade of intracellular events that leads to long-term modulation of baroreflex control of blood pressure.

Acknowledgements This work was supported in part by research grants NSC-85-2331-B075-077 from the National Science Council and VGH-266 from the Veterans General HospitalTaipei, Taiwan, Republic of China to J.Y.H. Chan. References [1] Cavali& A., Berninger, B., Haas, C.A., Garcia, D.E., Lindholm, D. and Lux, H.D,, Constitutive upregulation of calcium channel currents in rat phaeochromocytoma cells: roIe of c-@ and c-jun, J. Physiol., 479 (1994) 11-27. [2] Chiasson, B,J., Armstrong, J,N., Hooper, ML,, Murphy, P.R. and Robertson, H.A., The application of antisense oligonucleotide technology to the brain: some pitfalls, Cell. Mol. Neurobiol., 14 (1994) 507-521. [3] Chiasson, B.J., Hooper, M.L., Murphy, P.R. and Robertson, H.A., Antisense oligonucleotide eliminates in vivo expression of c-fos in mammafian brain, Eur. J, Pharmacol., 227 (1992) 451–453. [4] Ciriello, J., Brainstem projections of aortic baroreceptor afferent fibers in the rat, Neurosci. Lett., 36 (1983) 37-42. [5] Crooke, S.T., Therapeutic applications of oligosmcleotides, Annu. Reu. Pharmacol. Toxicol., 32 (1992) 329-376, [6] Curmn, T., Gordon, M.B., Rubino, K.L. and Sambucetti, L.C., Isolation and characterization of the c-fos (rat) cDNA and arraIysis of post-translational modification in vitro, Oncogene, 2 (1987) 79– 84. [7] Dean, C. and Seagard, J.L,, Expression of c-fos protein in the nucleus tractus solitaries in response to physiological activation of carotid baroreceptors, Neuroscience, 69 (1995) 249–257. [8] Dragunow, M, and Faull, R.L.M., The use of c-fos as a metabolic marker in neuronal pathway tracing, J. Neurosci. Meth., 29 (1989) 261-265. [9] Erickson, J.T. and Millhorn, D.E., Fos-like protein is induced in neurons of the medulla oblongata after stimulation of the carotid sinus nerve in awake and anesthetized rats, Brain Res., 567 (1991) 11-24. [101Heilig, M., Engel, J.A. and Soderpalm, B., c-fos antisense in the nucleus accumbens blocks the Iocomotor stimulant action of cocaine, Eur. J. Pharmacol., 236 (1993) 339–340. [11] Hoffman, B.B. and Lefkowitz, R.J,, Catecholamines and sympathomimetic drugs. In A.G. Gilman, T,W. Rail, A.S. Nies and P. Taylor (Eds.), The Pharmacological Basis of Therapeutics, Pergamon Press, New York, 1991, pp. 187-220. [12] Hooper, M.L., Chiasson, B.J. and Robertson, H.A., Infusion into the brain of an antisense oligonucleotide to the immediate–early gene c-fos suppresses production of Fos and produces a behavioral effect, Neuroscience, 63 (1994) 917-924. [13] Hughes, P. and Dragunow, M., Induction of immediate-early genes and the control of neurotransmitter-regulated gene expression within the nervous system, Pharmacol. Reu., 47 (1995) 133–178. [14] Hunter, A.J., Leslie, R.A., Gloger, 1.S, and Lawrence, M., Probing the function of novel genes in the nervous system: is antisense the answer? Trends Neurosci., 18 (1995) 329–331. [15] Kanter, R.K., Erickson, J.T. and Millhorn, D.E., Activation of the c-fos gene in prodynorphin- and proenkephalin-expressing cells of nucleus tractus solitaries after seizures, E.xp. Neurol., 129 (1994) 290–298. [16] Konradi, C., Cole, R.L., Heckers, S. and Hyman, S.E., Amphetamine regulates gene expression in rat striatum via transcription factor CREB, J. Neurosci., 14 (1994) 5623-5634. [17] Konradi, C. and Heckers, S., Haloperidol-induced Fos expression in striatum is dependent upon transcription factor cyclic AMP response element binding protein, Neuroscience, 65 (1995) 1051–1061.

C.-D. Shih et al./Brain Research 738 (1996) 39-47 [18] Krukoff, T.L., Morton, T.L., Harris, K.H. and Jhamandas, J.H., Expression of c-jios protein in rat brain elicited by electrical stimulation of the pontine parabrachial nucleus, J. Neurosci., 12 (1992) 3582-3590. [19] Len, W.-B., Tsou, M.-Y., Chan, S.H.H. and Chan, J.Y.H., Substance P suppresses the activity of az-adrenoceptors of the nucIeus reticularis gigantocellularis involved in cardiovascular regulation in the rat, Brain Res., 238 (1994) 227–234. [20] Li, Y.-W. and Dampney, R.A.L., Expression of c-~os protein in the medulla oblongata of conscious rabbits in response to baroreceptor activation, Neurosci. Lett., 144 (1992) 70–74. [21] McKitrick, D.J., Kmkoff, T.L. and Calaresu, F.R., Expression of c-~os protein in rat brain after electrical stimulation of the aortic depressor nerve, Brain Res., 599 (1992) 215-222. [22] Met., C.B., Schneider, S.P. and Fyffe, R.E.W., Selective suppression of endogenous peroxidase activity: application for enhancing appearance of HRP-labeled neurons in vitro, ./. Neurosci. Meth., 26 (1989) 181-188. [23] Miura, M., Takayama, K. and Okada, J., Neuronal expression of Fos protein in the rat brain after baroreceptor stimulation, J. Auton. Neru. Syst., 50 (1994) 31-43. [24] Morgan, J.I. and Curran, T., Stimulus-transcription coupling in the nervous system: involvement of the inducible proto-oncogenes fos and jun, Annu. Reu. Neurosci., 14 (1991) 421-451. [25] Murphy, A.Z., Ermis, M., Shipleyj M.T. and Behbeharri, M.M., Directionally specific changes in arterial pressure induce differential patterns of Fos expression in discrete areas of the rat brainstem: a double-Iabeling study for Fos and catecholamines, J. Comp. NeuroL, 349 (1994) 36-50. [26] Narv6ez, J.A., Cove?ias, R., de Le6n, M., Aguime, J.A., Cintra, A., Goldstein, M. and Fuxe, K., Induction of c-fos immunoreactivity in tyrosine hydroxylase and phenylethanolamine-N-methyltmnsferase

-.

47

imrrnrnoreactive neurons of the medulla oblongata of the rat after phosphate-buffered saline load in the urethane-anesthetized rat, Brain Res., 602 (1993) 342–349. [27] Sagar, S.M., Sharp, F.R. and Curran, T., Expression of c-jos protein in brain: metabolic mapping at the cellular level, Science, 240 (1988) 1328-1331. [28] Shih, C.-D., Chan, S.H.H. and Chan, J.Y.H., Participation of hypothalamic paraventricular nucleus in locus ceruleus-induced baroreflex suppression in rats, Am. J. Physiol., 269 (1995) H46-H52. [29] Stein, C.A., Does antisense exist? Namre Med., 1 (1995) 1119-1121. [30] Stemberger, L.A., Zmnrunocytochemistry,Prentice-Hall, Englewood Cliffs, 1974. [31] Suzuki, S., Pilowsky, P., Minson, J., Amolda, L., Llewellyn-Smith, I.J. and Chalmers, J., c-~os antisense in rostral ventral medulla reduces arterial blood pressure, Am. J. Physiol., 266 (1994) R1418– R1422. [32] Sved, A.F., Mancini, D.L., Graham, J.C., Schreihofer, A.M. and Hoffman, G.E., PNMT-containing neurons of the Cl cell group express c-~os in response to changes in baroreceptor input, Am. J. Physiol., 266 (1994) R361-R367. [33] Takayama, K., Suzuki, T. and Miura, M., The comparison of effects of various anesthetics on expression of Fos protein in the rat brain, Neurosci. L.ett., 176 (1994) 59-62. [34] Whitesell, L., Geselowitz, D., Chavany, C., Fahmy, B., Walbridge, S., Alger, J.R. and Neckers, L.M., Stability, clearance, and disposition of intraventricularly administered oligodeoxynucleotides: implications for therapeutic application within the central nervous system, Proc. Natl. Acad. Sci. USA, 90 (1993) 4665-4669. [35] Yee, F., Ericson, H., Reis, D.J. and Wahlestedt, C., Cellular uptake of intracerebroventricularly administered biotin- or digoxigeninIabeled antisense oligodeoxynucleotides in the rat, Cell. Mol. Neurobiol., 14 (1994) 475-486.

..—.