7{) l~m.

z Rescar<'//, I)12 { I ~93 ) 7CJ ~

1993 Elsevier Science Publishers B.V. All rights rcsc~'cd 11t)i~6-~9q3,,'93 .'$~16.t!~

BRES 18827

Changes in nociception, arterial blood pressure and heart rate produced

by intravenous morphine in the conscious rat

C.L. Thurston a,b, A. Starnes b and A. Randich b

" Department o f Biomedical Sciences, UniL,ersity of South Alabama, Mobile, A L 36688 (USA) and t, Department of Psycholo,~,,
Unit,ersity of Alabama at Birmingham, Birmingham, A L 35294 (USA)

(Accepted 15 December 1992)

Key words: Morphine; Vagus; Antinociception; Blood pressure; Heart rate

The present study shows that intravenous (i.v.) administration of morphine produces dose-dependent increases in tail flick and hot plate latencies
in conscious rats. I.v. morphine also decreased heart rate, but had no significant effects on arterial blood pressure. Transection of the right vagus
at the cervical level or pre-treatment with the peripherally acting opioid receptor antagonist naloxone methobromide attenuated the increased
tail flick latency produced by either 1.75 or 2.5 mg/kg morphine. In addition, either right vagotomy or naloxone methobromide attenuated the
increased hot plate latency produced by 1.75 mg/kg of morphine but not by 2.5 mg/kg of morphine. Following pre-treatment with naloxone
methobromide, 1.75 and 2.5 mg/kg of morphine produced a small pressor response 1-3 min after injection. The bradycardia produced by 1.75
mg/kg of morphine was attenuated by naloxone methobromide, but not by right vagotomy. The bradycardia produced by 2.5 mg/kg of morphine
was attenuated by either naloxone methobromide or vagotomy. These data obtained in the conscious rat are similar to previous reports using
pentobarbital-anesthetized rats except for the following: (i) the dose-response function for inhibition of the tail flick was shifted to the right in
conscious rats, (ii) the depressor response to morphine observed in anesthetized rats was attenuated in conscious rats, (iii) following naloxone
methobromide, but not unilateral vagotomy, i.v. morphine produced a pressor response in the conscious rat, and (iv) unilateral vagotomy was not
as effective in attenuating the antinociception and bradycardia in conscious rats as bilateral vagotomy is in pentobarbital-anesthetized rats.

INTRODUCTION NMB, but these treatments had varied effects on the

Systemic administration of morphine is known to
produce antinociception by actions directly in the cen-
tral nervous system (see Refs. 6 and 12 for reviews).
However, Randich et al) 3:4 recently reported that
antinociception produced by intravenous (i.v.) adminis-
tration of morphine in the rat also depends, in part, on
the integrity of cervical vagal afferents. They showed
that inhibition of either the tail flick (TF) reflex or
heat-evoked responses of lumbosacral spinal dorsal
horn neurons produced by intravenous (i.v.) adminis-
tration of low doses of morphine in the pentobarbital-
anesthetized rat is attenuated by either bilateral cervi-
cal vagotomy (CVAG) or by i.v. administration of the
peripherally-acting opioid receptor antagonist nalox-
one methobromide (NMB). These low doses of mor-
phine also produced bradycardia and hypotension 13:4.
The bradycardia was attenuated by either CVAG or
hypotension depending on the dose of morphine 3'4:3'14.
These data suggest that morphine can produce
antinociception, bradycardia, and hypotension by ac-
tions at peripheral opioid receptors, and these effects
require the integrity of vagal afferents for their expres-
sion.
Since the experiments noted above were performed
in pentobarbital-anesthetized rats, the possible influ-
ence of the anesthesia in producing any of the ob-
served outcomes must be clarified. Specifically, blood
pressure responses to morphine administration vary
depending on whether the animal is anesthetized and
whether or not the vagi are intact 2:'~'w. Since both the
blood pressure responses and changes in nociception
produced by i.v. administration of morphine depend in
part on the integrity of vagal afferents 1314, and since it
has already been shown that the blood pressure re-
sponses to morphine can differ in the anesthetized

Correspondence: C.L. Thurston, Dept. of Biomedical Sciences, UCOM 6000, University of South Alabama, Mobile, AL 36688, USA.
 71

tration of either 1.75 or 2.5 m g / k g of morphine. TF latencies were

compared to conscious rat 2'8, the effects of anesthesia
on morphine-produced changes in nociception must be
determined.
Therefore, the present experiments evaluated the
effects of i.v. morphine administration on nociception,
mean arterial blood pressure (MAP) and heart rate
(HR) in conscious rats. Changes in nociception were
measured using both TF and hot plate tests. The TF
test was used for direct comparisons with previous data
obtained in the lightly-anesthetized rat. The hot plate
test was used as a second measure of nociception for
determining the effects of i.v. morphine on a
supraspinally mediated response. The role of vagal
afferents in i.v. morphine-produced changes in noci-
ception, MAP, and HR were evaluated in rats with
unilateral transection of the right cervical vagus since
rats do not generally recover following bilateral CVAG.
In addition, the role of peripheral opioid receptors in
i.v. morphine-produced changes in nociception, MAP,
and HR were tested by pretreating the rats with NMB
prior to i.v. morphine administration.
recorded 1 min after NMB to determine if NMB alone altered
baseline TF latencies. This value was used for comparisons with
post-morphine T F latencies.
On day 5, the effects of i.v. morphine were tested in the hot plate
test. Rats received the same drug or drug combination that they
received on day 3. The hot plate test consisted of placing the rat on a
copper surface maintained at 51 +_0.5°C by a water bath. The latency
for a rat to either lick a hind paw or j u m p with both hind paws off
the surface of the plate was measured before, and at 1 min and 20
min after, i.v. administration of morphine. A cutoff latency of 40 s
was imposed to prevent tissue damage, but all rats responded before
40 s. For those rats receiving 5.0 m g / k g of NMB prior to morphine,
the response latency was also measured 1 rain after administration of
NMB.
Data analysis
TF data were analyzed as the percent of the maximum possible
effect according to the following equation: [(test l a t e n c y - b a s e l i n e
l a t e n c y ) / ( 1 0 - b a s e l i n e latency)] × 100, where the test latency is the
TF latency measured after morphine, baseline latency is the TF
latency measured prior to morphine administration, and 10 is the
cutoff latency. Hot plate data were analyzed as the difference be-
tween the test latency after morphine and the baseline latency before
morphine. M A P and H R data were analyzed as the percent change
compared to baseline values.
Data were analyzed with A N O V A s and N e u m a n - K e u l s and
Scheffe post-hoc analyses. Alpha was 0.05 for all analyses.

Drugs
MATERIALS AND METHODS Morphine sulfate was obtained from Mallinckrodt. Naloxone
methobromide was a gift from Boehringer Ingelheim Pharmaceuti-
cals, Inc.
Male Sprague-Dawley rats (280-400 g) were singly housed, pro-
vided food and water ad libitum, and maintained on a 12:12 h
light/dark cycle. RESULTS
O n day 1, catheterization of the femoral artery (PE 60 connected
to silastic tubing) and jugular vein (silastic tubing) was performed in
rats deeply anesthetized with sodium pentobarbital (55 m g / k g , i.p.). Nociception
The catheters were externalized at the nape of the neck. Some rats Baseline TF latencies (Table I) did not differ be-
received unilateral transection of the right cervical vagus following
careful dissection from the aortic depressor nerve and sympathetic
tween any groups. I.v. administration of morphine pro-
chain. All wounds were sutured and covered with a local anesthetic duced dose-dependent increases in TF latency in intact
ointment. rats (Fig. 1A). An ANOVA of the TF latencies indi-
O n day 2, the rats were placed in cylindrical plexiglass restraining
tubes twice for 20 rain to adapt the rats to the restrainers. While the cated significant effects of morphine dose, F3,28 = 59.26,
rats were in the restrainers, the catheters were flushed with hep- and the test trial time, F8,224 = 2.79. Newman-Keuls
arinized saline.
post-hoc analyses indicated that all doses of morphine
On day 3, the effects of i.v. administration of either morphine or
saline on TF latency, MAP, and H R were tested. Rats were placed tested produced an increase in TF latency compared to
in the cylindrical plexiglass tubes. The femoral artery catheter was
connected to a Cobe Pressure Transducer interfaced with an IBM
computer for m e a s u r e m e n t of M A P and HR. The TF reflex was
evoked by applying noxious radiant heat to the tail. The intensity of TABLE I
heat was adjusted at the beginning of the experiment such that
pre-drug TF latencies were between 3 and 4 s. This intensity of heat Mean +_S.E.M. baseline TF and hot plate (HP) latencies for sham-op-
stimulation was maintained constant for all animals. A 10 s cutoff erated rats (SHAM), right vagotomized rats (CVAG), and rats pre-
latency was imposed to prevent tissue damage. At least 3 T F trials treated with 5.0 mg / kg NMB (NMB)
were administered prior to drug injection to establish a consistent
latency (variability of less than 0.5 s over 3 consecutive trials). The Group Morphine dose n TF (sec) HP (sec)
last TF latency measured was used as the baseline value. M A P and (mg / kg)
H R were measured for 20 s prior to morphine administration and SHAM 0 8 3.19_+0.11 8.49+_0.73
the values 2 s prior to morphine administration were used as base- 1.0 8 3.11+_0.20 8.57_+0.81
lines. 1.75 8 3.43 _+0.09 8.64 +_0.70
I.v. morphine (1.0, 1.75, or 2.5 m g / k g in 100 p,l/kg sterile saline) 2.5 8 3.69_+0.23 7.47_+0.52
was administered followed by a 100/xl sterile saline flush. A control
group received 100 p d / k g sterile saline. T F latencies were recorded CVAG 1.75 8 3.29 +_0.14 8.64 +_0.82
at 10 s, 1 min, 2 min, 3 rain, 4 rain, 5 min, 10 rain, 15 rain, and 20 rain 2.5 7/6 * 3.22+_0.08 8.48_+1.25
after morphine administration. M A P and H R were monitored con-
NMB 1.75 8 3.46 _+0.25 9.70 +_ 1.70
tinuously for the first 5 rain and then at 5 s prior to the TF tests at
2.5 8 3.65+_0.14 7.09_+0.36
10, 15, and 20 min. Two groups of rats received i.v. administration of
5 m g / k g NMB (10 m g / m l sterile saline) 5 min prior to i.v. adminis- * (n = 7 for tail flick and n = 6 for hot plate.
72

i' ~, i i + L1
saline controls. The mean effect of 1.0 m g / k g of mor-
phine was significantly greater than saline, but signifi- A
cantly less than the mean effects of 1.75 and 2.5 m g / k g 10(} TO ~o o- •
of morphine, which were not different from each other. , 0 ~ \ T [ _- . / +
8 () /T,,l , I '~ ~ ,
TF data for sham-operated, vagotomized, or NMB- L~J
3s 60
1~ [ I T t
treated rats administered 1.75 m g / k g of i.v. morphine ,/1"• i i i t .. "
• ~-~-i "• i" "
are presented in Fig. 2A. Sham-operated rats are the 40
2 Ii/ l/ ....... " !
same group of rats receiving 1.75 m g / k g of morphine ~ 20
CL
presented in the d o s e - r e s p o n s e function of Fig. 1. An I" j

A N O V A indicated significant effects of group, F2,21 =

20 - . . . . . . ~ . . . . . L J
12.48, and the time of test trials, F~.l~ ~ = 4.59. Post-hoc q f, I 0 15 hq

analyses indicated that the mean %MPE values of B

15
NMB-treated rats were significantly less than those of
sham-operated rats. The mean %MPE values of vago-
tomized rats were intermediate to those of sham-oper- LIJ
0 5
Z
<
ated and NMB-treated groups, but could not be shown I ©
L~
to differ from either group. The significant effect of

~j i 0
r~

(~ F, ] (:, l ! ," I
(

](}II~] [
• + t l-I-I . . . . . . .
"iIi+ \T T ~ J + - J I • ---I 2 C'

80 • • T
+
u
~. 60 + ] :Z'

i 1 1• ~ e iJ .... - °I- - - 2z
t+
~
+Jl
,~o +,
" ++ [ ' + + ' / ...... J
[ (,) 2C

m t J Ld

F
-\ ,> c, ~ " ~ 4o • CVAC
O" + "O" "'O -- - - -- '(> - ? LJ
C2 • 5.0 PTI£d/ ~ [] J ',/1[~
t. I l_ L + I
60
0 ~ 1C 1! :;' ,}

qt~E ~MF;)
18},
Fig. 2. Effects of 5.0 mg/kg of NMB or right vagotomy (CVAG) on
inhibition of the TF, changes in MAP, and decreases in HR pro-
hi duced by 1.75 mg/kg of i.v. morphine. A: TF latency presented as
the percent maximum possible effect (MPE). B: MAP data presented
as the percent change from baseline values. C: HR data presented as
the percent change from baseline values.

,", {: iC 15 20 time reflects an increase in %MPE values over test

trials in all groups.
20
TF data for sham-operated, vagotomized, or NMB-
L~
I treated rats receiving 2.5 m g / k g of i.v. morphine are
U
presented in Fig. 3A. An A N O V A indicated significant
Z
effects of groups, Fz,zo = 4.62, and the time of test
~ 2e
trials, F8,160 = 4.07. Post-hoc analyses indicated that
LJ
( '
c~ 40
the mean %MPE values of vagotomized and NMB-
Ld
Cl treated rats were significantly less than those of sham-
6 ',') operated rats. The mean %MPE values of vagotomized
3 ~ 10 15 2C
rats were also significantly less than the mean %MPE
TME ,M 14b

Fig. 1. Dose response functions for i.v. administration of morphine.

values of rats treated with NMB.
A: TF latency presented as the percent maximum possible effect Hot plate data for all groups of rats are presented in
(MPE). B: MAP data presented as the percent change from baseline Fig. 4. Baseline hot plate latencies did not differ signif-
values. C: HR data presented as the percent change from baseline
values. icantly between any groups (Table I). In intact rats, an
 73

1 MIN POST INJECTION

A N O V A of the dose-response effects of i.v. morphine
on hot plate latency indicated a significant effect of Ld
12
Uq SHAM
morphine dose, F3,28 = 5.16. Scheffe post-hoc analyses NMB
:#

indicated that the mean difference in hot plate laten- bz 9

CVAG
Ld
cies for the 1.75 and 2.5 m g / k g groups did not signifi- 6
cantly differ, but were significantly greater than the
Z
mean difference in hot plate latency for the saline 5

control group. The mean difference in hot plate la- 3

Ld
O~
o m
tency for the 1.0 m g / k g group did not differ from
L~
either the combined means of the 1.75 and 2.5 m g / k g -5

groups or the saline control group. Although there was SAL 1.0 175 25

no significant effect of time, 1.0 m g / k g of morphine MORPHINE DOSE ( m g / k g )

clearly had no effect on the hot plate latency at 1 min

after injection. Therefore, A N O V A s were run at the
two time points. At 1 min post-morphine administra- 20 MIN POST-INJECTION

I SHAM
Ld
Ca [~] NMB
2.5 mg/kg IV M O R P H I N E
b 9 KS~ CVAO

5
A
1 O0 • .T r,e-e-• ? - - ? ~ |

z;

~
8O

6O

4o
/ "/
z

U
z
w
c~
3

0
"r n
DA
(D
O2
U~ 20
,1,it a 5
SAL 10 1.75 2.5
O_

0 MORPHINE DOSE ( m g / k g )

--20
Fig. 4. Effects of i.v. morphine on the hot plate latency in sham-oper-
0 5 10 15 20 ated (SHAM) rats, rats with right vagotomies (CVAG), or rats
pre-treated with NMB at 1 min (top panel) and 20 min (bottom
B
15
panel) post-injection. SAL = saline. * = P < 0.05 compared to saline
controls, t = P < 0.05 compared to sham-operated controls receiving
same dose of morphine.
T •

z
I 0
(D T/tT j ~ ~ I I tion, 1.0 m g / k g of morphine had no significant effect
on hot plate latency whereas 1.75 and 2.5 m g / k g
DA
Q) • "+.~ I 1
O/
w --10
morphine significantly increased hot plate latencies
13_
compared to both saline controls and the 1.0 m g / k g
--15
0 5 10 15 2O dose. At 20 rain post-morphine administration, all 3
C doses of morphine produced a significant increase in
20
hot plate latency compared to saline controls with no
significant differences between doses.
~ o
An A N O V A comparing the effects of 1.75 m g / k g of
2-
C) --20
-~_¥, - * "r U i.v. morphine on hot plate latency in intact, vago-
/f • SHAM tomized, and NMB-treated rats indicated a significant
IJ
~ 4o ,~ • CVAG effect between groups, F2,21 = 4.35. Post-hoc analyses
kd
Q_ • 5.0 mg/kg NMB
indicated that the mean increase in hot plate latencies
--60 - - following i.v. administration of morphine in vago-
0 5 10 15 20
TIME (MIN)
tomized rats or rats pre-treated with N M B were not
Fig. 3. Effects of 5.0 mg/kg of NMB or right vagotomy(CVAG) on significantly different, but were significantly less than
inhibition of the TF, changes in MAP, and decreases in HR pro- the mean increases in hot plate latencies in sham-oper-
duced by 2.5 mg/kg of i.v. morphine. A: TF latency presented as the ated rats.
percent maximum possible effect (MPE). B: MAP data presented as
the percent change from baseline values. C: HR data presented as An A N O V A comparing the effects of 2.5 m g / k g of
the percent change from baseline values. i.v. morphine on hot plate latency in intact, vago-
74
tomized, and NMB-treated rats indicated a significant
effect across time of testing only, k'~.~,) -- 6.55. indicat-
ing the greater increase in hot plate latency at 20 min
compared to 1 rain.
Arterial blood pressure
Baseline MAP differed significantly between groups
in studies of the morphine dose response function,
F<> = 4.46 (Table II). The mean baseline MAP of rats
to receive 1.0 m g / k g of morphine was significantly less
than the mean baseline MAP of rats to receive either
saline or 2.5 m g / k g of morphine.
l.v. administration of morphine in sham-operated
rats had no significant effect on MAP compared to
saline controls, with only a significant effect of the time
of testing, Fs,2~la = 6.09. There was a tendency for a
small depressor response to occur during the first 2
rain after injection (Fig. 1B), but one-way ANOVAs at
each time trial fail to indicate any significant effects.
Fig. 2B presents the MAP data for sham-operated,
NMB-treated, and vagotomized rats given 1.75 m g / k g
of morphine. An ANOVA indicated significant effects
of the time of testing, Fs,~,0 = 2.11; and the group ×
time interaction, F~,.t6 . = 2.45. One-way ANOVAs and
Neuman-Keuls post-hoc analyses indicated that at 1
and 2 rain, NMB-treated rats showed a significant
increase in MAP compared to sham-operated and
vagotomized rats. At 4 min, the percent change in
MAP of vagotomized rats was significantly different
from NMB-treated rats, although neither group dif-
fered significantly from sham-operated rats.
In sham-operated, NMB-treated, and vagotomized
rats given 2.5 m g / k g of morphine (Fig. 3B), an ANOVA
indicated significant effects of group, F2,2o = 3.95; the
time of testing, Fs.t6 o = 4.43; and the group × time
interaction, F~<~,o= 2.71. One-way ANOVAs and
Neuman-Keuls post-hoc analyses indicated that fol-
T A B L E 11
Mean + S.E.M. baseline mean arterial blood pressure (MAP) and heart
rate (HR) of sham-operated rats (SHAM), right t,agotomized rats
(CVAG), and rats pre-treated with 5.0 mg / kg NMB (NMB)
Group Morphine dose n MAP (mmHg) HR (bpm)
(rag / kg)
SHAM 0 8 132.1+2.0 478.5+14.1
1.0 7 121.4+ 1.4" 416.6+ 6.6 "
1.75 8 126.1+3.(/ 458.6_+ 13.5
2.5 8 133.8+-2.7 457.5+ 11.3
CVAG 1.75 8 134.8+- 3.6 469.1 + 16.5
2.5 7 139.4+-3.7 448.7+14.2
NMB 1.75 8 131.6_+_3.5 477.6+- 7.1
2.5 8 130.5+3.3 498.5+- 11.2 b
~' Significantly different compared to saline controls.
b Significantly different compared to SHAM-operated rats receiving
the same dose of morphine.
lowing NMB, 2.5 m g / k g of i.v. morphine produced a
significant increase in MAP from 1--3 min post-injec-
tion compared to sham-operated and vagotomized rats.
At 10 s post morphine administration, MAP was signif-
icantly decreased in sham-operated rats compared to
both NMB and vagotomized rats, with the percent
change in MAP in vagotomized rats significantly less
than that in NMB-treated rats.
Heart rate
In studying the dose-response function of mor-
phine, baseline H R differed significantly between
doses, Fs,27 = 4.07 (Table II). The mean baseline HR
of rats to receive 1.0 m g / k g of morphine was signifi-
cantly less than the mean baseline HR of rats to
receive saline.
Neither vagotomy nor NMB had a significant effect
on baseline H R in rats to receive 1.75 m g / k g of
morphine. However, there was a between groups effect
in rats to receive 2.5 m g / k g of morphine, F2, m = 4.15.
Post-hoe analyses indicated that rats pre-treated with
NMB had significantly greater baseline HRs than
sham-operated or vagotomized rats. However, a
within-subject comparison of baseline HRs before and
after NMB administration failed to indicate any signifi-
cant difference.
l.v. administration of morphine in intact rats pro-
duced a pronounced bradycardia (Fig. IC), with signifi-
cant effects of morphine dose, F~2~, = 14.98; the time
of testing, Fs,2o s -- 46.65; and the dose × time interac-
tion, F24,2~~ = 7.31. One-way ANOVAs and post-hoe
analyses indicated that all 3 doses of morphine pro-
duced a significant bradycardia compared to saline-in-
jected rats during the first 5 min post-injection with no
differences between doses. At 15 min post-injection,
1.75 and 2.5 m g / k g of morphine produced a significant
bradycardia compared to saline and 1.0 m g / k g of
morphine. By 20 min, HR had returned to baseline
levels. During the first 10 s following morphine admin-
istration, rats frequently showed arrhythmias and occa-
sionally complete block of the heart lasting 2-4 s. This
occurred at all 3 doses of morphine tested.
NMB completely blocked the bradycardia produced
by 1.75 m g / k g of i.v. morphine (Fig. 2C), with signifi-
cant effects of g r o u p , F2,20 = 8.55; the time of testing,
F8,160); and the group × time interaction, F~<~,, = 14.59.
One-way ANOVAs and post-hoc analyses indicated
that NMB significantly attenuated morphine-produced
bradycardia for the first 3 min following injection com-
pared to both vagotomized and sham-operated rats. At
4 min post-injection, both NMB and vagotomy signifi-
cantly attenuated morphine-produced bradycardia
compared to sham-operated rats.

Fig. 3C shows the effects of NMB or vagotomy on
the bradycardia produced by 2.5 mg/kg of i.v. mor-
phine. An ANOVA indicated significant effects of the
time of testing, F8,160 = 38.69; and the group x time
interaction, F16,160 = 8.59. One-way ANOVAs and
Neuman-Keuls post-hoc analyses indicated that NMB
significantly attenuated the bradycardia compared to
both vagotomized and sham-operated rats at 10 s post
injection, whereas both NMB and vagotomy attenuated
the bradycardia at 1 min post-injection. There were no
significant differences at later time points.
DISCUSSION
Nociception
Intravenous administration of morphine produced
dose-dependent antinociception as measured with ei-
ther the TF or hot plate assays. In the TF test, 1.0
mg/kg of morphine produced a small increase in TF
latency compared to saline controls, whereas both 1.75
and 2.5 mg/kg of morphine increased TF latencies to
near the cut-off value of 10 s. A comparison with
previous data obtained in the pentobarbital-anesthe-
tized rat indicates that the dose-response function for
i.v. morphine-produced inhibition of the TF is shifted
to the right in conscious rats 13'14. Specifically, 1.0 mg/kg
of morphine increased TF latencies to 4.5-6.5 s in the
conscious rat, whereas both 0.5 and 1.0 mg/kg of
morphine increased the TF latencies in the lightly-
anesthetized rat to approximately 8 s, using the identi-
cal apparatus as the present study.
In the present study, all 3 doses of morphine pro-
duced an increase in TF latency as early as 10 s
post-injection. However, morphine produced greater
increases in the TF latency at later time points (10-20
min). Similar observations were made in the lightly-
anesthetized rat 13'14, and as noted below, this may
reflect a peripherally-mediated component that occurs
shortly after injection (within 10 s) and a centrally-
mediated component that occurs at later time points.
Right vagotomy or NMB-treatment significantly at-
tenuated increases in TF latency produced by either
1.75 or 2.5 mg/kg of i.v. morphine. The 1.0 mg/kg
dose of morphine was not tested with either NMB or
vagotomy due to the relatively small increase in TF
latency produced by this dose. The attenuation was
greatest within 5 min of morphine administration, and
the pattern of attenuation was nearly identical to that
observed in the lightly-anesthetized rat 13,14 Not sur-
prisingly, unilateral vagotomy was less effective in at-
tenuating morphine-produced increases in the TF la-
tency than bilateral vagotomy. These data suggest that
morphine-produced changes in TF latency observed
75
shortly after injection (within 5 min) require peripheral
opioid receptors and the integrity of vagal afferents.
However, after 5 min, the antinociceptive effects of
morphine are primarily due to non-vagal actions, pre-
sumably direct central or spinal antinociceptive actions
that may mask the ability to detect peripheral actions.
In the hot plate test, 1.75 and 2.5 mg/kg of mor-
phine produced increases in hot plate latencies at 1
min, whereas 1.0, 1.75, and 2.5 mg/kg increased hot
plate latencies at 20 min. Both NMB and vagotomy
attenuated the increases in hot plate latency produced
by 1.75 mg/kg morphine at 1 but not 20 min, whereas
neither NMB nor vagotomy affected the increases in
hot plate latencies produced by 2.5 mg/kg morphine.
These data suggest that the 2.5 mg/kg dose of mor-
phine may be able to access the central nervous system
within 1 min post-injection, and that both the 1.75 and
2.5 mg/kg doses of morphine acted centrally by 20
min.
Interestingly, the antinociception produced by 2.5
mg/kg of morphine seemed to be more susceptible to
attenuation by unilateral vagotomy than NMB in both
the TF and hot plate assays during the first 5 min post
injection. However, the dose of NMB may not have
been great enough to block all peripheral actions of
morphine (see also the heart rate data for 2.5 mg/kg
morphine). Use of a higher dose may have produced
problems in interpretation due to possible access to the
CNS ~. In the lightly-anesthetized rat, bilateral vago-
tomy and NMB were equally effective in attenuating
morphine-produced inhibition of the TF reflex 13
Previous studies with the peripherally-acting mor-
phine analogue, N-methyl morphine reported that
there are little peripheral effects of this drug in pro-
ducing antinociception in the hot plate test. Specifi-
cally, subcutaneous 15 or intraperitoneal 7 administration
of N-methyl morphine had no effect on hot plate
latencies whereas lesser doses of morphine increased
hot plate latencies. The lack of an effect of N-methyl
morphine in the above studies may be attributed to the
route of administration or time of testing. In the study
by Smith et al. 15, the hot plate latency was measured 45
min after subcutaneous administration of N-methyl
morphine. Peripheral actions of this opiate may be
short-lived. In the study by Foster et al. 7, procedures
followed are not clearly stated and data not clearly
presented, although the authors do state that time
response curves for the hot plate test were generated.
In addition, Bianchi et al. 1 failed to find an effect of
1-8 mg/kg NMB on increases in hot plate latency
produced by i.v. administration of 5 mg/kg morphine.
There are two important comments to be made regard-
ing this study. First, 5 mg/kg of morphine produced a
76
ceiling effect (i.e. all animals failed to respond prior to
the cutoff latency of 30 s) in the study of Bianchi et al. 1
that may have precluded detection of a peripheral
action. Second, the hot plate test was the assay used by
Bianchi et al. as the measure of central actions of
NMW. According to their reasoning, if NMB is found
to attenuate i.v. morphine-produced antinociception,
then NMB is presumed to be accessing the CNS.
However, had a d o s e - r e s p o n s e function for morphine
been performed, they may have found attenuation with
doses of NMB less than 8 m g / k g when doses of
morphine of 2.5 m g / k g or less are used, as in the
present study.
Arterial blood pressure
I.v. administration of morphine had no significant
effect on MAP compared to saline controls, although a
small depressor response was evident at all 3 doses
during the first 3 min post-injection. Although this
varies from the large depressor responses observed in
our previous studies in the lightly-anesthetized rat 13'14,
the lack of a significant depressor effect was not sur-
prising since anesthesia has been reported to have
profound effects on opiate-produced changes in blood
pressure (see Refs. 5, 8, and 11 for reviews). For
example, G o m e s et al. 8 reported that 10 m g / k g of
morphine in the conscious rat produces an increase in
blood pressure and heart rate whereas the same dose
in the anesthetized rat produces a decrease in blood
pressure and heart rate. The effect of anesthesia may
be to produce a rightward shift in the dose response
MAP function similar to what we observed with the TF
test since lower doses of morphine, 7.5 m g / k g infused
over 10 rain, produces a decrease in blood pressure
and heart rate in conscious rats within the first minute
of infusion 16. In addition, intramuscular administration
of 10 m g / k g of morphine in conscious rats produced
an initial depressor response that lasted less than 5 min
and was followed by a pressor response 2.
Following pre-treatment with NMB, 1.75 and 2.5
m g / k g of i.v. morphine produced a pressor response at
1-3 rain following injection. These data suggest that
morphine produces a central pressor response that is
normally masked by a peripheral depressor response.
This pressor response following NMB was not ob-
served in lightly-anesthetized rats t3'14, suggesting that
the pressor effects of systemic morphine are attenu-
ated in the anesthetized rat. Further, attenuation of
the pressor response by anesthesia may allow detection
of a peripherally-mediated depressor response that is
not observed in the conscious rat.
In the anesthetized rat preparation, higher doses of
morphine (1-100 m g / k g ) produce a depressor re-
sponse in intact rats and a pressor response in rats with
bilateral vagotomy, whereas lower doses of morphine
produce a depressor response in both intact and bilat-
eral vagotomized rats 3'4'13. Although morphine did not
produce a pressor response in rats with unilateral
vagotomy, the inability to test bilateral vagotomy in
conscious rats precludes strong conclusions on negative
data.
Heart rate
All 3 doses of morphine produced an equivalent
decrease in H R that was greatest at 10 s following
injection. The bradycardia is similar to that produced
in lightly anesthetized rats 13'14. Arrhythmias also oc-
curred at all 3 doses, with occasional cardiac arrest
occurring within the first 20 s after injection. NMB, but
not right vagotomy, attenuated the bradycardia pro-
duced by 1.75 m g / k g of morphine. NMB or vagotomy
significantly attenuated the bradycardia produced by
2.5 m g / k g of morphine at 1 min and NMB also attenu-
ated the bradycardia at 10 s post-injection. Further,
none of the rats pre-treated with NMB showed ar-
rhythmias or cardiac arrest. The inability of right vago-
tomy to attenuate the bradycardia is most likely at-
tributed to the use of unilateral transection since stud-
ies in the anesthetized animal have repeatedly shown
that morphine-produced bradycardia is attenuated by
bilateral vagotomy 3'4"13'14. The residual bradycardia in
NMB-treated rats suggests that the dose of NMB was
insufficient to adequately block peripheral opiate re-
ceptors with the high dose of morphine administered.
Kiang et al. 1° previously showed that NMB produces a
shift in the ED50 for morphine-produced bradycardia
in the anesthetized rat. However, it is also possible that
morphine has central bradycardic actions more easily
observed in the conscious compared to anesthetized
rat.
In conclusion, the present data obtained in con-
scious rats confirms our previous data obtained in
lightly anesthetized rats 13 that either vagotomy or i.v.
administration of NMB attenuates i.v. morphine-pro-
duced antinociception. NMB and vagotomy also atten-
uated the cardiovascular responses of morphine.
Acknowledgements. This work was supported by NIH Grant 22966
to A. Randich.
REFERENCES
1 Bianchi, G., Fiocchi, R., Tavani, A. and Manara, L., Quaternary
narcotic antagonists' relative ability to prevent antinociception
and gastrointestinal transit inhibition in morphine-treated rats as
an index of peripheral selectivity, Life Sci., 30 (1982) 1875-1883.
2 Conway, E.L., Brown, M.J. and Dollery, C.T., Plasma cate-
cholamine and cardiovascular responses to morphine and D-ala~-

d-leuS-enkephalin in conscious rats, Arch. Int. Pharmacodyn., 265
(1983) 244-258.
3 Delle, M., Thoren, P., Skarphedinsson, J.O., Hoffman, P., Cads-
son, S. and Rickstein, S., Differentiated responses of renal and
adrenal sympathetic nerve activity to intravenous morphine ad-
ministration in anesthetized rats, J. Pharm. Exp. Ther, 253 (1990)
655 -660.
4 Fennessy, M.R. and Rattray, J.F., Cardiovascular effects of intra-
venous morphine in the anaesthetized rat, Eur. J. Pharrnacol., 14
(1971) 1-8.
5 Feuerstein, G., The opioid system and central cardiovascular
control: analysis of controversies, Peptides, 6 (1985) 51-56.
6 Fields, H.L. and Basbaum, A.I., Brainstem control of spinal
pain-transmission neurons, Ann. Rev. Physiol., 40 (1978) 217-248.
7 Foster, R.S., Jenden, D.J. and Lomax, P., A comparison of the
pharmacologic effects of morphine and N-methyl morphine, J.
Pharrn. Exp. Ther., 157 (1967) 185-195.
8 Gomes, C., Svensson, T.H. and Trolin, G., Effects of morphine
on central catecholamine turnover, blood pressure and heart rate
in the rat, Naunyn-Schmiedeberg's Arch. Pharmacol., 294 (1976)
141-147.
9 Holaday, J,W., Cardiovascular effects of the endogenous opiate
system, Annu. Rev. PharmacoL Toxicol., 23 (1983) 541-594.
10 Kiang, J.G., Dewey, W.L. and Wei, E.T., Tolerance to morphine
bradycardia in the rat, J. Pharm. Exp. Ther., 226 (1983) 187-191.
11 Lang, R.E., Bruckner, U.B., Kempf, B., Rascher, W., Sturm, V.,
77
Unger, Th., Speck, G. and Ganten, D., Opioid peptides and
blood pressure regulation, Clin. Exper. Hyper. Theory Pract., A4
(1982) 249-269.
12 MiUan, M.J., Multiple opioid systems and pain, Pain, 27 (1986)
303-347.
13 Randich, A., Thurston, C.L., Ludwig, P.S., Timmerman, M.R.
and Gebhart, G.F., Antinociception and cardiovascular responses
produced by intravenous morphine: the role of vagal afferents,
Brain Res., 543 (1991) 256-270.
14 Randich, A., Thurston, C.L., Ludwig, P.S., Robertson, J.D. and
Rasmussen, C., Intravenous morphine-induced activation of vagal
afferents: peripheral, spinal and CNS substrates mediating inhibi-
tion of spinal nociception and cardiovascular responses, J. Neuro-
physiol., 68 (1992) 1027-1045.
15 Smith, T.W., Buchan, P., Parsons, D.N. and Wilkinson, S., Pe-
ripheral antinociceptive effects of N-methyl morphine, Life Sci.,
31 (1982) 1205-1208.
16 Thornhill, J.A., Townsend, C. and Gregor, L., Intravenous mor-
phine infusion (IMF) to drug-naive, conscious rats evokes brady-
cardic, hypotensive effects, but pressor actions are elicited after
IMF to rats previously given morphine, J. Physiol. Pharmacol., 67
(1989) 213-222.
17 Willette, R.N. and Sapru, H.N., Peripheral versus central car-
diorespiratory effects of morphine, Neuropharmacology, 21 (1982)
1019-1026.