Involvement of cyclic AMP at the level of the nucleus reticularis pontis caudalis in the acoustic startle response. Thereza C.M. de Lima 1, Michael Davis *.
BRAIN RESEARCH ELSEVIER
Brain Research 700 (1995) 59-69
Research report
Involvement of cyclic AMP at the level of the nucleus reticularis pontis caudalis in the acoustic startle response Thereza C.M. de Lima 1, Michael Davis * Yale University School of Medicine, Department of Psychiatry, 34 Park Street, New HaL,en, CT 06508, USA Accepted 21 June 1995
Abstract
Rats were implanted with cannulas in the nucleus reticularis pontis caudalis (PnC), an obligatory part of the neural pathway that mediates the acoustic startle reflex. Following at least 1 week of recovery, rats were tested for acoustic startle amplitude before or after infusion of compounds known to alter the second messenger, adenosine cyclic Y,5'-monophosphate (cAMP). Local infusion into the PnC of the cAMP analog, 8-bromo cAMP (0.125-1.0 /zg), increased the amplitude of the acoustic startle response in a dose-dependent manner. In addition, local infusion of a phosphodiesterase inhibitor, rolipram (10 /xg) or the water soluble adenylate cyclase activator, forskolin-DHA (2.5 /zg), produced a significant enhancement of startle amplitude. These effects probably resulted from intracellular actions because cAMP itself, which does not readily penetrate lipid membranes, had no effect. Moreover, the effects seemed somewhat specific because the precursor of cAMP, ATP or 8-bromo cGMP, also failed to alter startle at doses where 8 bromo-cAMP did. The fact that a phosphodiesterase inhibitor elevated startle suggests that cAMP serves to tonically elevate startle at this level of the pathway. Hence, treatments that either increase (fear, sensitization) or decrease (habituation, pre-pulse inhibition) startle at the level of the PnC may do so via release of neurotransmitters either positively or negatively coupled to cAMP, which in turn may alter either sound evoked transmitter release, excitability of PnC neurons or both. Keywords: Reticular formation; Fear; cAMP; Startle; Amygdala
1. Introduction
The acoustic startle response is a short-latency reflex elicited by sudden, loud auditory stimuli, which can be elicited in all mammals [15]. This reflex is being used increasingly to study the effects of drugs on behavioral reactivity to sensory stimulation [13], attention [23,63], sensitization [7,16,39] and fear and anxiety [17,27]. The acoustic startle reflex is mediated by an elementary neural pathway within the brainstem and spinal cord which we now believe involves projections of the auditory nerve to cochlear root neurons [49], lesions of which eliminate acoustic startle in rats [43]. These cochlear root neurons project directly to the nucleus reticularis pontis caudalis [46,47] which then projects to motoneurons in the spinal cord [46]. Hence, the nucleus reticularis pontis caudalis
* Corresponding author. Fax: (1) (203) 562-7079. 1 Present address: Department of Pharmacology, CCB, Universidade Federal de Santa Catarina, Rua Ferreira Lima, 82, Florianopolis, SC, Brazil, 88015-420. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved
SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 8 3 7 - 3
(PnC) plays a critical role in relaying acoustic input to motoneurons [10,20,24,28,29,37,43-46,58,64,68]. In addition, the PnC may be an important point along the acoustic startle circuit where other parts of the brain project so as to modulate this basic reflex. For example anatomical and physiological data indicate a direct excitatory pathway from the medial part of the central nucleus of the amygdala to the part of the PnC where lesions eliminate the acoustic startle response [35,57]. Electrical stimulation of the amygdala increases tone-evoked single unit activity in the PnC [35] and increases the amplitude of the acoustic startle reflex in freely moving rats [54], probably via modulation at the PnC [56]. Elevation of acoustic startle by conditioned fear or footshock is blocked by lesions of this direct neural pathway [30,31] and both conditioned fear and sensitization of startle by footshocks appear to ultimately modulate startle at the level of the PnC [4,7,39]. At the present time, a detailed anatomical description of the connection between the amygdala and neurons in the PnC is not yet available, so that it is not clear how activation of the amygdala facilitates the acoustic startle
60
T.C.M. de Lima, M. Davis/Brain Research 700 (1995) 59-69
reflex. The amygdala might directly increase the excitability of reticulospinal neurons by a direct projection to these cells in the PnC. Consistent-with this 'post-synaptic mechanism', Koch and Ebert [35] found that electrical stimulation of the central nucleus of the amygdala in anesthetized rats induced spikes in 60% of PnC neurons and increased the responsiveness of 87% of these cells to a startle-eliciting stimulus. Using intracellular recording in anesthetized rats, Lingenhohl and Friauf [46] found that electrical stimulation of the central nucleus of the amygdala increased the excitability of PnC neurons at currents which also produced excitatory post-synaptic potentials in these reticulospinal neurons. In addition, Yeomans and Pollard [70] report that electrical stimulation of the amygdala at high currents elicits a jerk which they interpret as a startle response. Another, not mutually exclusive, possibility is that the central nucleus of the amygdala projects to terminals in the PnC originating from neurons in earlier parts of the startle pathway. If so, activation of these axonal-axonal connections could increase transmitter release from these terminals and thereby increase acoustic startle amplitude. Consistent with this 'pre-synaptic mechanism' are the findings that: (a) a conditioned fear stimulus has never been reported to elicit a startle reflex itself based on behavioral [42] or electromyographic recordings [11] and (b) electrical stimulation of the amygdala can markedly increase startle at currents which produce no behavioral [54] or electromyographic [55] effects by themselves. Concerning the neurotransmitters involved in facilitation of startle at the PnC, it is important to distinguish between the neurotransmitter that actually mediates startle at this level of the pathway vs. the neurotransmitter released from the amygdala which then alters the activity of the former transmitter. It is probable that an excitatory amino acid like glutamate may mediate startle at the level of the PnC [22,38,50]. Tone-evoked activation of PnC neurons can be inhibited by iontophoretic application of the glutamate antagonist 6-cyano-7-nitroquinoxaline-2,3dione (CNQX [22]) and local infusion of either CNQX or the NMDA antagonist D,L-2-amino-5-phosphonopentanoic acid (AP5) in the PnC in freely moving rats essentially eliminates the acoustic startle reflex [38,50]. In some brain areas release of glutamate can be enhanced by adenosine cyclic 3',5'-monophosphate (cAMP) [21]. In other instances, cAMP analogues have been shown to increase ionic currents mediated by glutamate receptors [26,66,67] which are substrates for cAMP-dependent phosphorylation [6]. If the amygdala released a neurotransmitter positively coupled to cAMP in the PnC, then activation of the amygdala could increase startle by increasing the release of glutamate in terminals from auditgry neurons which project to the PnC a n d / o r by increasing excitability of PnC neurons via phosphorylation of glutamate receptors. If so, then direct infusion of cAMP analogues into the PnC might increase the amplitude of the acoustic startle
reflex. The purpose of the present study was to test this latter hypothesis by locally infusing into the PnC drugs that affect the cAMP second messenger system.
2. Materials and methods
2.1. Animals Seventy naive male albino Sprague-Dawley rats (Charles River, Kingston) weighing approximately 320400 g were used. Following surgery, the animals were individually housed in metal cages in a large colony room maintained on a 12/12 light-dark cycle (lights on at 07.00 h) with water and food continuously available.
2.2. Surgery Surgery was carried out under sodium pentobarbital (Nembutal, 50 m g / k g i.p.) anesthetic using methods developed previously [50]. The rats were placed in a skull-flat orientation in a Kopf stereotaxic instrument. Unilateral (22 gauge, 11 mm length; Plastic Products n = 20) or bilateral guide cannulas (n = 50) with a 3.0 mm separation and fitted with internal cannulas (28 gauge) extending 1.0 mm beyond the guide tip were held in the stereotaxic electrode carrier. The carrier was then angled back 12 ° so that the cannulas entered the brain moving from caudal toward rostral, at an angle of 78 ° with respect to the skull, rather than the more standard 90 ° entry. With the carrier in this position, the coordinates, with respect to bregma, were: - 1 2 . 0 mm AP, 1.5 mm ML and - 9 . 6 mm DV from the skull surface over the PnC [52]. This oblique approach was used to avoid the mid-saggital sinus at the level of the lambdoid suture, which bleeds excessively when punctured and also to avoid passing the cannulas through the parabrachial nuclei, which, if damaged bilaterally, produces respiratory distress. The implanted cannulas were held in place using jewelers screws attached to the skull and a crown of dental acrylic. Following surgery, animals recovered for 1 week before behavioral testing.
2.3. Drugs and infusions 8-Bromo adenosine cyclic 3',5' hydrogen phosphate monosodium salt (8-bromo cAMP; Sigma), forskolin hydrochloride 7b-deacetyl-Tb-[g-(morpholino) butyryl] (forskolin-DHA - - Calbiochem), 8-Bromo guanosine cyclic 3',5' hydrogen phosphate monosodium salt (8-bromo cGMP; RBI), adenosine cyclic 3',5' hydrogen phosphate monosodium salt (cAMP, Calbiochem) and adenosine triphosphate disodium salt (ATP, RBI) were dissolved in artificial cerebrospinal fluid (ACSF). ACSF was made up with 5 mM potassium chloride; 130 mM sodium chloride, 1.25 mM sodium phosphate, 15 mM D-glucose, 24 mM sodium carbonate, 2 mM calcium chloride and 2 mM
T.C.M. de Lima, M. Davis/Brain Research 700 (1995) 59-69
magnesium sulfate. Rolipram was dissolved in a 100% DMSO. All solutions were adjusted to pH 7.0. A Harvard infusion pump was used to infuse the compounds through removable injectors inserted into the guide cannulas, in a volume of 0.5 /zl over a 2-min period. The injectors remained in the guide cannulas for an additional minute after infusion and then were replaced by stylets.
61
2501 "~
200-
r~ 150-
I~
100-
1,~ t-
O-
2.4. Apparatus The apparatus used to measure startle has been described previously [9]. Briefly, five separate stabilimeters were used to record the amplitude of the startle response. Each stabilimeter consisted of an 8 × 15 × 15 cm high Plexiglas and wire mesh cage suspended between compression springs within a steel frame. Cage movement resulted in displacement of the accelerometer, the analog output of which was amplified (Endevco Model 104) and digitized via a MacAdios II board interfaced to a Macintosh II computer. Startle amplitude was defined as the maximum accelerometer voltage that occurred during the first 200 ms after the startle stimulus was delivered. The stabilimeters were housed in a dark, ventilated, sound attenuating chamber. Each cage was located 10 cm from a high frequency speaker (Radio Shack Supertweeter). The startle stimulus was a 50-ms burst of white noise having a rise-decay time of 5 ms. Background white noise, provided by a white noise generator, was 55 dB. Sound level measurements were made within the cages using a Bruel and Kjaer 2235 sound level meter (A-scale).
8-Bromo cAMP
550 -
1.0
0.75
"o o,-m ii ~-
E
,~
500-
450 400350 -
ACSFO ~
0
250-
2ooI~
150 -
ACSF 100500
.15
0
15
30
45
60
75
90
105
120
i
135
Time (minutes) Fig. 1. M e a n amplitude startle d u r i n g the last 5 min o f the pre-infusion baseline a n d over subsequent 20-min periods f o l l o w i n g infusion o f A C S F , 0.125, 0.25, 0.5, 0.75 and 1.0 /xg 8 - b r o m o c A M P in eight rats or repeated infusions o f A C S F only in another eight rats. A r r o w s indicate times o f infusion.
-50-
-100-
0./25 Dose
0.~5 of
0'.5 8-Bromo
0.~s
i
cAMP
(~g)
Fig. 2. Percent c h a n g e in startle, relative to an initial infusion of A C S F , after infusion of various doses o f 8 - b r o m o c A M P in eight rats or repeated infusions o f A C S F in another eight rats. The mean startle amplitude over a 20-min period after an initial infusion o f A C S F w a s first subtracted f r o m the m e a n startle amplitude over the 10-min prior to infusion. This difference w a s then subtracted f r o m the m e a n startle amplitude over a 2 0 - m i n period after e a c h o f the following infusions and converted into a percent c h a n g e relative to the initial A C S F difference.
2.5. Behavioral procedures Because of large individual differences in baseline startle amplitudes, but consistent startle amplitudes across days, the animals were divided into groups which had equivalent mean baseline startle amplitudes during a matching procedure carried out before and 1 week after surgery. In this matching procedure rats were placed in the startle test cages and 5 min later presented with the first of 30 noise bursts (10 at each of three intensities: 95, 105 and 115 dB) at a 30-s interstimulus interval (ISI). During the test session, 20 rats implanted with unilateral cannulas into the PnC were placed in the startle cages and 5 min later presented with 20 startle stimuli at 30-s intervals. The last ten stimuli of this pre-infusion test period provided a baseline startle level against which to evaluate subsequent drug effects. Each animal was then removed from the cage, infused with ACSF and returned to the startle cage. A total of 40 startle stimuli were then presented at 30-s intervals. Half of the animals were then removed from the test cages, infused with the lowest dose of 8-bromo cAMP (0.125 /zg) and then presented with another 40 startle stimuli. This procedure was repeated using successively higher doses of 8-bromo cAMP (0.25, 0.5, 0.75 and 1.0 /xg) with each infusion. The other half of the animals were treated identically except that ACSF was infused each time. About 1 week later, rats previously infused with ACSF were now tested in the same way after
T.C.M. de Lima, M. DaL,is / Brain Research 700 (1995) 59-69
62
x\ OclM
DCIC //
/
/I1
\\X
"\
I
f'
/
\
-9.30
/
\//
~
//
x % CIC
X"
I
I
"/
n ,[
', "~%
Plh
\
,,,.
PFi
• }"7 /PtSVL
t PnC
%,
Fni
I I I
/
,.---
11/~,
Oc I ~
paS/// %%
~-
l/#
,
~',,
I
%%%
fJ~
-9.68 PLF
',
,/
i
PFI
",
.--,..1
-.
;_;7;.
-9.80 ~.),,/-/I
•
%~.,,,"
[
/
r.c
% -"~
~,,¢>
...~,,,,,.. ,.,;',, ._ ,,,-
-10.04 /
PLF
//
PFI
-10.30 )
,--.7 -10.52
,/ ,'k~'i-, ",,
/ A,
,.":,
"" 45,"
I_l /
I
T.C.M. de Lima, M. Davis~Brain Research 700 (1995) 59-69
63
repeated infusions with a single dose of 8-bromo cAMP
(0.25 ~g). The 50 animals implanted with bilateral cannulas were divided in six groups with equivalent mean startle amplitudes, based on the matching data. After the pre-infusion period and the vehicle period, as described above, they were infused with either 8-bromo cAMP (0.5 /~g/side, n = 10), rolipram (5 /xg/side, n = 10), forskolin-DHA (2.5 p,g/side, n = 7), 8-Br cGMP (0.5 /zg/side, n = 8), cAMP (0.5 /.Lg/side, n = 8) or ATP (0.5 /xg/side, n = 7) at a rate of 0.5 /zl/min. Immediately after infusion all animals were presented with 100 startle stimuli at 30-s intervals.
:t
"•
Forskolin-DHA 2.5 I~g
L.
...............
• .........................
r~ 100-
ACSF
50-
2.6. Histology At the end of the experimental procedures the animals were sacrificed by an overdose of chloral hydrate and perfused intracardially with physiological saline followed by 10% buffered paraformaldehyde solution. The brains were removed and fixed by immersion with 30% sucrose paraformaldehyde solution and cut on a freezing microtome. Sections (45 ~m) were mounted and stained with Cresyl violet. Infusion sites were localized under a light
0
4
-15
;
, 115
30
Time
450
715
-
4oo,~
.,.~-
.
