WANGBin , ZHANGBenguo , GE Xuecai , LUO Fei , HAN Jisheng△
(Neuroscience Research Institute and De partment of Neurobiology , Peking University , Beijing 100083 ,China)
KEY WORDS　Morphine ; Heroin dependence ;Conditioned place preference ; Drug craving ; Transcuta2
neous elect ric nerve stimulation
SUMMARY　 Objective :To test the hypothesis that peripheral elect ric stimulation ( PES) may suppress
the reinstatement of morphine2induced conditioned place preference (CPP) in rat s as well as the drug
craving of detoxified heroin addict s in a f requency2dependent manner . Methods :CPP model of the rat
was const ructed with two compartment automatic CPP apparatus , and the craving of the heroin addict s
was assessed with a visual analogue scale (VAS) . Results :(1) PES of low f requency could prevent the
drug priming2 or foot shock2induced reinstatement of morphine CPP ; (2) this effect was naloxone2re2
versible , suggesting a possible involvement of endogenous opioid mechanisms ; and (3) PES of low f re2
quency could also accelerate the rate of natural decay of drug craving in heroin addict s af ter successful
abstinence. Conclusion :PES might serve as a therapeutic measure for the t reatment of heroin addic2
tion. ( J Peking Univ [ Heal th Sci ] , 2003 ,35 :2412247)
外周電刺激抑制大鼠**條件性位置偏愛的復(fù)發(fā)
及***成癮者的心癮
王 　玢 ,張本國 ,葛學(xué)采 ,羅 　非 ,韓濟(jì)生△
(北京大學(xué)神經(jīng)科學(xué)研究所 ,北京大學(xué)基礎(chǔ)醫(yī)學(xué)院神經(jīng)生物學(xué)系 , 北京 　100083)
[關(guān)鍵詞] **;***依賴;條件性位置偏愛;心癮;經(jīng)皮神經(jīng)電刺激
[摘 　要] 目的:探討外周電刺激(PES)是否能頻率依賴性地抑制**條件性位置偏愛(CPP)的復(fù)發(fā)和抑制***
成癮者脫毒后的心癮。方法:用二室自動(dòng) CPP箱記錄大鼠條件性 CPP ,用視覺模擬尺測量***成癮者的“心癮
(渴求)”。結(jié)果: (1)低頻 PES能抑制大鼠小劑量**點(diǎn)燃、或腳底電刺激誘發(fā)的** CPP ; (2)上述效應(yīng)可被小劑
量**受體拮抗劑納洛酮(1 mg· kg - 1
)翻轉(zhuǎn) ,提示有內(nèi)源性**機(jī)制參與; (3)低頻 PES還能使***成癮者脫毒
后對(duì)毒品“心癮”的自然消退過程加速。結(jié)論:外周神經(jīng)電刺激可能是*****成癮的一種有效方法。
[中圖分類號(hào)] R338 　　[文獻(xiàn)標(biāo)識(shí)碼] A　　[文章編號(hào)] 16712167X(2003) 0320241207
　　Drug addiction involves not only the acquisition
and maintenance of drug2using behavior , but also the
reinstatement of this behavior af ter successful absti2
nence. High rate of relapse af ter long period of absti2
nence characterizes the behavior of experienced users
of heroin and other drugs of abuse (J affe , 1990) .
Over the past several decades , most of the studies
were focusing on the neurobiological mechanisms of
drug reward , which was viewed as a cent ral factor in
drug abuse
[1 ,2 ]
1Only recently had scientist s begun to
focus on the neurobiology of relapse , a bet ter under2
standing of which could lead to more effective t reat2
ment st rategies for addictive disorders.
Drug craving is a subjective description that can
only be measured in human subject s. However , rein2
statement of an operant event can be directly mea2
sured when a laboratory animal reinitiates a particular
behavioral response. This reinstatement is thought to
be able to mimic the induction of drug seeking follow2
ing extinction f rom drug using. A model of the rein2
statement of self2administ ration had been built by
Shaham[3 ,4 ]
and co2workers using either a small
(priming) dose of morphine or a brief exposure to
foot shock st ress. Another putative animal model for
this purpose is the reinstatement of conditioned place
preference (CPP) , reported simultaneously by Parker
and McDonald[5 ]
, Lu and co2workers
[6 ]
and a team in
our lab[7 ]
. All these models can be employed for the
study of human relapse behavior . But until now , few
studies were reported aiming at the possible preven2
tion of relapse on these paradigms.
Our previous works had shown that peripheral
elect ric stimulation ( PES) , especially those with a
component of low f requency (2 Hz) , could block the
expression of CPP induced by morphine
[8 ]
in a nalox2
one2reversible manner . Other studies in our lab have
amply shown that PES of high f requency (100 Hz)
can suppress morphine withdrawal syndrome both in
rat s
[9 ,10 ]
and heroin addict s
[11 ,12 ]
. Hence it becomes
interesting to test the hypothesis whether PES could
also prevent the reinstatement of morphine2induced
CPP in laboratory animals , or more directly , to di2
minish drug craving in heroine addict s.
1 　Animal experiment
1. 1 　Materials and Methods
1. 1. 1 　Animals 　All experiment s were performed
on male Sprague2Dawley rat s provided by the Insti2
tute of Animal Research , Chinese Academy of Sci2
ence , weighing 180 - 200 g at the beginning of the
experiment . They were housed 6 per cage , with the
ent groups might not be equal .
11114 　Extinction and reinstatement of CPP 　Our
previous work has shown that morphine ( 4 mg·
kg - 1
, once a day for 10 days)2 induced CPP in rat s
extinguished 7 days af ter the last drug2 pairing ses2
sion[7 ]
. In the present study , observation was made 9
days af ter the last drug2 pairing session1Rat s were ei2
ther injected with a priming dose (0. 25 mg· kg - 1
,
s. c. ) of morphine 15 min before testing , or exposed
to intermit tent foot shock (with square waves of am2
plitude of 0. 5 mA , width of 0. 5 s , the off time ran2
domly dist ributed among 10 - 70 s with an average of
40 s) for 15 min immediately before testing. The test
procedure was the same as that mentioned above. Ap2
paratus generating the foot shock stimulation was
bought f rom Qinghua Elect ronic Product Company ,
Beijing , China.
11115 　Peripheral Elect ric stimulation for the rat
Rat s were kept in special holders , with their hind legs
and tails prot ruding[13 ]
. Two stainless steel needles of
0. 3 mm diameter were inserted into each hind leg ,
one in the acupoint ST36 (5 mm lateral to the anteri2
or tubercle of the tibia) , and the other in SP6 (3 mm
proximal to the medial malleolus , at the posterior
border of the tibia) . Constant current square2waveroom temperature maintained at (24 ±1) ℃, relative
humidity at 50 % , under a 12∶ 12 h light2dark cycle.
The experimental procedures were approved by the
Commit tee of Animal Care and Use of Peking Uni2
versity.
1. 1. 2 　Drugs 　Morphine hydrochloride was pur2
chased f rom the First Pharmaceutical Factory of
Shenyang , China. Naloxone HCl was purchased f rom
SIGMA Co. , Ltd. All drugs were dissolved in 0. 9 %
saline to their final concent rations.
1. 1. 3 　Conditioned place preference 　The methods
of CPP paradigm have been described in detail else2
where
[8 ]
. Briefly , conditioning took place in one of
the two distinct environment s , which differed in color
and texture and separated by a removable t ransparent
clapboard. The walls of one room were painted with
vertical black and white st ripes (width : 2 cm) , and
the floor comprised a layer of fiberboard bedding. In
the other room , the walls were painted with black
dot s ( diameter : 1. 5 cm) sprinkled on white back2
ground , and the flooring material was 1 cm thick of
sawdust. The later was used as the drug2paring
room. In the acquisition phase , rat s were t rained
once a day for 10 days by giving int raperitoneal
(i . p. ) injection of morphine (4 mg· kg - 1
) 5 min af2
ter the animals were put into the drug2pairing area ,
where they were kept for another 15 minutes. On the
11th day , the t ransparent clapboard was removed ,
and the rat s were scored during a 102min test session
with their time spared in drug2paring compartment .
Rat s were only considered as being in a compartment
when both forepaws were located in that environ2
ment . Rat s that did not show sufficient place prefer2
ence were discarded. Thus numbers of rat s in differ2
ent groups might not be equal .
11114 　Extinction and reinstatement of CPP 　Our
previous work has shown that morphine ( 4 mg·
kg - 1
, once a day for 10 days)2 induced CPP in rat s
extinguished 7 days af ter the last drug2 pairing ses2
sion[7 ]
. In the present study , observation was made 9
days af ter the last drug2 pairing session1Rat s were ei2
ther injected with a priming dose (0. 25 mg· kg - 1
,
s. c. ) of morphine 15 min before testing , or exposed
to intermit tent foot shock (with square waves of am2
plitude of 0. 5 mA , width of 0. 5 s , the off time ran2
domly dist ributed among 10 - 70 s with an average of
40 s) for 15 min immediately before testing. The test
procedure was the same as that mentioned above. Ap2
paratus generating the foot shock stimulation was
bought f rom Qinghua Elect ronic Product Company ,
Beijing , China.
11115 　Peripheral Elect ric stimulation for the rat
Rat s were kept in special holders , with their hind legs
and tails prot ruding[13 ]
. Two stainless steel needles of
0. 3 mm diameter were inserted into each hind leg ,
one in the acupoint ST36 (5 mm lateral to the anteri2
or tubercle of the tibia) , and the other in SP6 (3 mm
proximal to the medial malleolus , at the posterior
border of the tibia) . Constant current square2wave
elect ric stimulation produced by a programmed pulse
generator (HANS LH2800 , produced by Beijing Uni2
versity of Aeronautics and Ast ronautics ) was deliv2
ered via the two needles for a total of 30 min. The
f requency of stimulation used was 22 , 1002 , or
2/ 1002 Hz (2 Hz alternating automatically with 100
Hz , each lasting for 3 s) . The pulse width was 0. 6
ms in 2 Hz and 0. 2 ms in 100 Hz. The intensity of
the stimulation was increased stepwise f rom 1 mA to
2 and 3 mA with each step lasting for 10 min. In this
study , PES was administered to rat s 18 hours before
CPP reinstatement . The cont rol group received mock
PES , with the needles being inserted in si t u without
administering elect ric stimulation. The motor behav2
ior of the rat s was un2distinguishable between the
PES group and cont rol group before the CPP testing.
1. 2 　Result s
1. 2. 1 　Effect of PES on low dose morphine2 or foot
shock 2 induced reinstatement of CPP 　Seventy2six
rat s were randomly dist ributed into 8 groups , with
9 - 10 in each group. All groups were t rained with
i . p. injection of morphine (4 mg· kg - 1
) , once per
day for 10 days. Af ter that they were kept in home
cages for an 82day extinction. Each two groups of rat s
were then given normal saline (s. c. ) 15 min prior toPES of 22 , 1002 , or 2/ 1002 Hz , or needling without
elect rical stimulation as cont rol , respectively. Eigh2
teen hours af ter the PES , one of the two groups re2
ceived a priming injection of morphine ( 0. 25
mg· kg - 1
, s. c. ) . Fif teen minutes later , they were
tested in the t raining apparatus again. The other four
groups were tested immediately af ter an intermit tent
foot shock st ress of 15 min. The result s are shown in
Fig. 1 and Fig. 2. Training with 4 mg· kg - 1
mor2phine for 10 days established a stable CPP (original
CPP) . The extinguished CPP was successfully rein2
stated by our drug2 priming paradigm and foot shock
st ress ( reinstated CPP) . Treatment with 2 Hz or 2/
100 Hz PES blocked the reinstatement of the faded
CPP ( P < 0. 001) , while 100 Hz PES or needling
had no such effect ( P > 0. 05) .
3 P < 0. 001 , compared with cont rol .
Figure 1 　Effect of PES on drug2priming2induced CPP reinstatement
for the Diagnostic and Statistical Manual of Mental
Disorders , Fourth Edition ( SCID) . Their average
time period of heroin abuse was 3 years [ ( 2. 95 ±
1. 05) years ] , and the dose of heroin used per day was
estimated to be in the range of 0. 5 - 2. 0 g (purity
not identified , but estimated to be in the range of
30 % - 70 %) . The i . v. route of administ ration com2
prised about 30 % - 35 % of the cases , with the rest
being taken via inhalation. The research protocol has
been approved by the Department of Scientific Re2
search , Peking University Health Science Center ,
and the Department of Public Security of the Zhan2
J iang County , Guangdong Province. All the partici2
pant s had provided writ ten in3 P < 0. 001 , compared with cont rol .
Figure 2 　Effect of PES on foot shock2induced CPP reinstatement
1. 2. 2 　Naloxone blocks the effect s of PES on drug
priming 2 and foot shock2 induced CPP reinstatement
　Sixty2 four rat s were randomly dist ributed into 6
groups , with 10 - 11 in each group. The t raining ,
testing , and extinction procedures were exactly the
same as mentioned above. Rat s were injected with
naloxone ( 1 mg ·kg - 1
, s. c. ) 10 min prior to
needling or PES of 2 Hz or 2/ 100 Hz , respectively.
Eighteen hours later , they were either injected with
0. 25 mg· kg - 1
( s. c. ) morphine 15 min before CPP
test , or given intermit tent foot shock immediately be2
- 1
3 P < 0. 001 , compared with corresponding only saline + PES group ;
NS ,nornal saline ; NX , naloxone.
Figure 3 　Naloxone blocks the effects of PES on drug2priming2induced
CPP reinstatement
3 P < 0. 001 , compared with corresponding saline + PES group.
Figure 4 　Naloxone reversed the effects of PES on foot
shock2induced CPPreinstatementformed consent to par2fore the testing for CPP. Naloxone of 1 mg· kg - 1
completely reversed the effect of either 2 Hz 2 or
2/ 100 Hz2 PES on drug priming2( Fig. 3) and foot
shock2(Fig. 4) induced reinstatement of the extin2
guished CPP.
2 　Human observations
2. 1 　Materials and Methods
2. 1. 1 　Human subject s 　One hundred and seven2
teen heroin abusers , aged 18 - 41 years old , were in2
volved in this study. All but four of the subject s were
male. They have been diagnosed as heroin depen2
dence according to the St ructured Clinical Interview
2. 1. 2 　Transcutaneous elect rical nerve stimulation　
Constant current square2wave elect ric stimulations
were delivered f rom a HANS LH2800 stimulator
through two pairs of skin elect rodes. One pair was
put on the acupoint of Hegu (L I24 , on the back of
the hand , between the first and the second metacar2
pus , near the cent ral point of the second metacarpus)
and Laogong ( P28 ,on the palmer side of the hand ,
opposite to the Hegu point ) , and the other pair on
Neiguan (P26 , located at the palmer side of the fore2
arm, 2 inches above the palmer groove , in between
the two tendens) and Waiguan ( TE25 , at the oppo2
site side of the P26 point) . The f requency of elect ricalstimulation used was 22 , 1002 , or 2/ 1002 Hz. The
pulse width was 0. 6 ms in 2 Hz and 0. 2 ms in 100
Hz. The intensity of the stimulation was adjusted to
the maximal value that was tolerable by the subject
without causing any painful feeling. The threshold
value ( T) was 5 - 7 mA. The average intensity that
was tolerable for the subject in day 1 was 8 - 10 mA
(1. 5 ×T) ,followed by 10 - 15 mA (2 ×T) in day 2 ,
15 - 20 mA (3 ×T) or more in day 3 and thereaf ter .
The t reatment was administered once a day ( 30
minutes per session) for 10 days.
In the cont rol group , the skin pads were placed
on the same point s and connected to the HANS de2
vice. The f requency was set to 2/ 100 Hz AM (ampli2
tude modulation ) mode and intensity adjusted to
threshold level (5 - 7 mA) , remaining there for less
than 2 minutes , and switched to 1 mA thereaf ter .
This intensity has been shown to produce no signifi2
cant changes in pain threshold in humans
[14 ]
.
2. 1. 3 　Rating of craving scores　Self2rating of hero2
in craving scores were obtained with the use of a visu2
al analog scale (VAS) . The craving rate ranged f rom
0 (not at all , marked on the lef t ext reme of the scale)
to 10 ( the st rongest one could imaging , marked on
the right ext reme of the scale) . The assessment was
2. 2 　Result s
The effect of PES on subjective craving score in
the ex2heroin abusers was shown in Fig. 5. During
the pre2t reatment phase , the craving scores of pa2
tient s in these 4 groups decreased gradually in a simi2
lar rate. The rate of decay of the craving scores for
the cont rol , 100 Hz , 2 Hz , and 2/ 100 Hz group were
- 0. 084 ±0. 013 , - 0. 081 ±0. 005 , - 0. 093 ±
0. 008 , and - 0. 082 ±0. 013 , respectively. The rate
of decay remained constant in the cont rol group dur2
ing the t reatment phase ( - 0. 083 ±0. 004) and the
post t reatment phase ( - 0. 087 ±0. 004) . In the 100
Hz group , the rate of decay showed a moderate accel2
eration during the t reatment phase ( - 0. 102 ±
0. 006) and the post2t reatment phase ( - 0. 102 ±
0. 005) , but the difference as compared to the sham
group was statistically not significant . In cont rast ,
this rate was accelerated significantly in 2 Hz and
2/ 100 Hz groups ( - 0. 174 ±0. 005 and - 0. 220 ±
0. 007 , respectively ; P < 0. 05 and P < 0. 01 , com2
pared with the cont rol group , respectively) during
the t reatment phase , resulting in a net decrease of
1. 0 - 1. 5 scores as compared to the other two (con2
t rol and 100 Hz ) groups. In the post2t reatmentperformed in the morning by a t rained research assis2
tant , unaware of the grouping of the population.
2. 1. 4 　Treatment procedures 　The detoxification
procedure was conducted by the application of Han′ s
acupoint nerve stimulator ( HANS) , supplemented
with a small dose of methadone for the first 2 - 5 days
as described previously[15 ]
. The subject s were en2
rolled in the current research at least one month af ter
the successful detoxification protocol . A total of 117
heroin addict s were randomly assigned into 4 groups ,
i . e. ,cont rol ( n = 29) , 2 Hz2( n = 30) , 100 Hz2( n =
29) and 2/ 100 Hz2( n = 29 ) HANS group. The
whole study consisted of 3 phases : ( 1) Pre2HANS
Phase : f rom day 1 to day 10 , the craving scores were
accessed once a day , without HANS t reatment ; (2)
HANS t reatment phase : f rom day 11 to day 20 , the
subject s received HANS of different f requencies or
sham2t reatment (cont rol group) once a day , preceded
by the assessment of craving score ; (3) Post2HANS
phase : f rom day 21 to day 30 , the craving score was
assessed once a day , without HANS t reatment .
2. 1. 5 　Statistical analysis 　Data were processed by
commercially available sof tware GraphPad Prism 3. 0.
Result s were presented as ? x ±s?. In the animal
group , - 0. 131 ± 0. 008 ; 2/ 100 Hz group ,
- 0. 109 ±0. 002) midway between the pre2t reatment
and t reatment phase. These result s suggested that
PES containing a low f requency component could ac2
celerate the natural decay of heroin craving in human
subject s.
3 　 Discussion
The most important finding in the present study
is that PES containing a low2 f requency component
(either pure 2 Hz or 2/ 100 Hz) could effectively an2
tagonize the reinstatement of drug2seeking behavior
induced by drug2 priming or by foot shock st ress in ex2
perimental animals. This effect could be completely
prevented by pret reatment with a small dose (1 mg·
kg - 1
) of naloxone , the opioid receptor antagonist . In
the meantime , it could also accelerate the decay of the
craving score in the heroin abusers , with an identical
f requency dependency compared with that of the rat .
311 　The possible neural mechanism underlying the
inhibitory effect of low2 f requency PES on CPP rein2
statement induced by drug2 priming or foot shock
Although both st ress and re2exposure to drugsstudy , comparisons between means of groups were
made with two2way analysis of variance (ANOVA)
followed by Student s Newman2Keul′ s test . In the
human study , for each group of patient s and in each
t reatment phases , linear regression analysis was per2
formed , with the average craving score as the depen2
dent variable , and days af ter successful extinction as
independent variable. The slope of the regression line
was calculated as an indication of the rate of decay of
the craving score. The accepted level of statistical sig2
nificance was P < 0. 05.
are known to be important factors for inducing relapse
in humans
[16 , 17 ]
as well as in rodent s
[3 , 4 ]
, the mech2
anisms underlying the two relapse inducers seemed to
be different . While dopamine release in the nucleus
accumbens (NAc) is essential for opiate priming2in2
duced relapse , this was less important in st ress 2in2
duced CPP reinstatement
[3 , 4 ]
, despite the fact that
there is a greater induction of heroin2seeking behavior
by st ressors
[3 , 4 ]
. This discrepancy suggest s that
st ress 2induced relapse may involve certain dopamine2
independent mechanisms , including the corticot ropin
ticipate in the study.