Marks, J. (1978). The benzodiazepines; use,overuse, misuse,abuse. Lancaster, England: M.T.P.

Press.

Mendelson, G. (1978). Withdrawal symptoms after oxazepam. Lancet,I, 565.

Möhler, H. & Okada, T . (1977). Benzodiazepine receptors demonstrated in the central nervous

system . Science, 108,849-851.

Pevnick, J. S.,Jasinski, D. R. & Haertzen, C. A. (1978). Abrupt withdrawal from therapeutically

administered diazepam. Arch.gen. Psychiat ., 35,995-8.

Preskom, S. H. & Denner, L. J. (1977). Benzodiazepines and withdrawal psychosis; report of

three cases. J. Am. med. Ass., 237,36-38

Rickels, K., Downing, R. W. & Winokur, A. (1978). Anti-anxiety drugs; c1inical use in

psychiatry. In Handbook ofPsychopharmacology, Vol. 13, ed. Iversen, L. L., Iversen , S. D.

& Snyder, S. H., pp. 395-430. New York: Plenum.

Rifkin , A., Quitkin, F. & Klein , D. F. (1976). Withdrawal reaction to diazepam. J. Am. med.

Ass., 236,2172-3.

Roth, M., Gumey, c., Garside, R. F. & Kerr, T. A. (1972). Studies in the classification of

affective disorders. The relationship between anxiety states and depressive iIInesses. Brit. J.

Psychiat., 121,147-162 .

Rutherford, D. M., Okoko, A. & Tyrer, P. J. (1978). Plasma concentrations of diazepam and

desmethyldiazepam during chronic diazepam therapy. Brit. J. c1in. Pharmac., 6,69-73.

Skegg, D. C. G., Doll, R. & Perry, J. (1977). Use ofmedicines in general practice. Brit. med.J..I,

1561-3.

Trethowan, W. (1975). Pills for personal problems. Brit. med. J ., 3, 749-751.

Tyrer, P. (1978). Drug treatment of psychiatric patients in general practice, Brit. med: J., 2,

1008-1010.

Tyrer, P. J. & Lader, M. H. (1974). Response to propranolol and diazepam in somatic and

psychic anxiety. Brit. med. J., 2, 14-16.

Winokur, A., Rickels, K., Greenblatt, D. J., Snyder, P. J. & Schatz, N. J. (1980). Withdrawal

reaction from long term, low dosage, administration of diazepam. Arch.gen. Psychiat., 37,

101-105.

World Health Organisation Expert Committee on Drug Dependence (1974). 20th Report.

W.H.O. Report Series, No. 551.W.H.O.: Geneva.

RELATIONSHIPS OF ß-ENDORPHIN

IN PHYSICAL DEPENDENCE

E.L. WAY

Departm ent of Pharmacology.

Sch ool ofMedicine,

University of California,

San Francisco,

California 94143, USA

There is mounting evidence that many opiate-like peptides do not always exh ibit

some of the classic signs of opiate action possessed by morphine and its alkaloidal

agonist surrogates. The explanation offered by most investigators is that a multiplicity of opiate receptors exist with varying degrees of affmity for va rious opioid

agonists. This variation in activity extends not only to different organs and tissues but

also to different sites within the brain itself. Furthermore, there is considerable

species and strain difference in this regard. This obviously means that many of the

effects noted with the opiate-like peptides in experimental animals cannot always be

extrapolated for accurate prediction ofpharmacological effects in humans.

Ofthe peptides that have been isolated, ß-endorphin is the most potent in relieving

pain and most generally resembles morphine with respect to pharmacological

profile, The isolation ofthe peptide was first reported independently by Li & Chung

(1976) and by Bradbury, Smyth & Snell (1976). In vitro ß-endorphin can be demonstrated to bind with high affin ity to opiate binding sites in the brain and to inhibit the

contractile response to electrical stimulation in the guinea pig ileum (Cox , Goldstein

& Li, 1976). In vivo also ß-endorphin has been shown to produce many ofthe classic

opiate-like effects in rodents. In mice it can be shown to produce analgesia (Loh,

Tseng, Wei & Li, 1976b) and hyperthermia after acute administration (HuidobroToro & Way, 1979). ß-endorphin also exhibits cross tolerance to morphine with

respect to analgesia in m ice and rats , (Tseng, Loh & Li, 1976; Huidobro-Toro, Li &

Way, 1978) and inhibitory effects on the electrically stimulated ileum from guinea

pigs (Huidobro-Toro & Way, unpublished data). Cross dependence to morphine can

also be demonstrated in rodents as evidenced by the ability of ß-endorphin to

suppress the abstinence signs following discontinuance of morphine after chronic

administration (Wei & Loh , 1976; Tseng et al., 1976). Primary dependence on

p-endorphin could also be established in the rat after its sustained administration

intraventicularly by osmotic minipumps (Wei & Loh, 1976). It would be reasonable,

therefore, to expect that ß-endorphin should produce similar effects in humans.

Some experiments on the use of ß-endorphin to suppress abstinence signs and

symptoms in heroin addicts will be described as weil as some animal studies with

respect to the effects of ß-endorphin on Cav flu x upon which an hypothesis for

opiate action has been proposed .

374 E. L. WAY

Suppression ofthe abstinence syndrome in heroin addicts by p-endorphin

A Taipei study reported by Su, Lin, Wang, Li, Hung, Lin & Lin (1978), indicates that

p-endorphin can effectively attenuate the signs and symptoms of heroin addicts

undergoing severe acute withdrawal. The study was conducted double-blind on eight

subjects shortly after they were arrested for drug use and understandably the amount

of heroin the addicts were dependent on could not be accurately determined,

although some indication was obtained by relating the stated size of the daily habit

with the severity of the withdrawal signs and symptoms, To overcome this shortcoming, two follow-up experiments on four subjects which were performed under

conditions where physical dependence could be unequivocally established and

quantified (Su, Lin, Peng, Cheng, Loh, Li & Way, 1980)will be described .

In the first study, the subjects were stabilized on a regimen of 60 mg of morphine

daily , administered in three divided doses for two to three weeks. Abrupt withdrawal

was initiated after this period by substituting saline for the morphine. An evaluation

of the ability of p-endorphin to suppress the abstinence signs and symptoms was

carried out between 28-32 h after discontinuance of morphine. At this time the

addicts exhibited the c1assic signs of opiate withdrawal; yawning, lacrimation, rhinorrhea and goose flesh were generally in evidence. The symptomatic complaints were

also typical in that there were frequent complaints of chilis, muscular and skeletal

aching. Moreover, restlessness and anorexia were prominent as well as increasing

hostility. The inten sity ofthe abstinence syndrome experienced by each subject was

recorded on a double-blind basis using a modification of a well-established scoring

system (Kolb Himmelsbach, 1938; Quock, Cheng, Chan & Way, 1968). Assessments

were made following intravenous infusion of saline and then comparing on a blind

basis the effects of deliberate injection of p-endorphin or saline over aperiod of ten

minutes into the infusion-system.

All subjects reported alleviation in physical discomfort with the institution of

saline infusion but began complaining after 30-60 min that the drug was too weak.

The tests to evaluate thesuppressant action of p-endorphin on the withdrawal signs

and symptoms were then commenced by slow injection of saline or ß-endorphin

(8-16 mg) into the infusion system in full view ofthe subject over a 30-60 min interval. Again, all subjects reacted positively to the injection but , with the exception of

one subject, the relief afforded was reported to be temporary and lessthan that afforded by the initial priming injection with saline . Six to eight hours later the subjects

were very hostile and aggressive in demanding relief.

The failure to discriminate effectively between the test and the control substances

required a change in the protocol. The morphine injections were continued in two

subjects for an additional two days after which time are-evaluation of p-endorphin

was performed. However, the injection procedure was modified in that the subjects

were not permitted to watch the proceedings, that is, the injection of the saline or

p-endorphin was made into the infusion system placed behind a screen. Also, the

control or test substance was administered rapidly into the infusion system instead of

by slow infusion. As previously, the assessments were made immediately after an

initial injection of saline in both subjects to obtain baseline data . This was followed

by injection of either p-endorphin (12 mg) or saline on a double-blind basis. One

hour later a eross-over study was performed.

Under these conditions, the effects of saline were not perceptible when given either

before or after p-endorphin. On the other hand, drug effects after p-endorphin were

noted within a minute after its injection when given prior to or following saline . After

p-endorphin, both subjects first reported a feeling of constriction in the ehest and

abdomen. This was followed shortly thereafter by more pleasant effects which they

identified as opiate-like. With in frve to ten minutes muscular and bone aches were

stated to be disappearing and marked improvement in attitude became readily

ß -ENDO RPHIN IN PHYSI CAL DEPE ND ENCE 375

apparent. Neither subject reported euphoria after 8-endorphin but within one hour

after the injection both siept soundly and upon awakening ate weil.

In a second study on three subjects, two were stabi lized for ten days on three 30 mg

doses of morphine daily and a third on three 20 mg doses daily. Abrupt withdrawal

was initiated on the tenth day as before by substituting saline for the midnight

morphine dose and continuing with the saline for two more times; tests were initiated

30-32 h after discontinuance of morphine. The abstinence signs and symptoms in

these subjects were less than those encountered in the first study when subjects were

stabilized on 60 mg daily for a longer period.

Before initiating the actual tests, the three subject s were conditioned first by an

intravenous bolus inject ion of saline followed by interrogation for drug effects and

measurements of vital signs. All three subjects stated that the 'drug' was beneficial

but was very weak or short-Iasting . Thirty minutes after the saline injection the

subjects were prepared for intravenous infusion of saline . After 20 min , a bolus

injection ofsaline into the infusion system was made from behind the screen and the

subject was again queried concerning his state . When no indication ofawareness that

an injection had been made , double-blind testing with ß-endorph in (15 mg) and

saline was initiated .

The fmdings were in agreement with the first study. In essence, a feeling of tightness in the ehest and/or abdomen after ß -endorphin was noted frrst, followed by

rapid and increasing relief of morphine withdrawal signs and symptoms over a 5-30

min period. Two of the subjects , however, complained about symptorns after 6-8 h

and again were afforded reliefby a repeat injection of ß-endorphin.

The study essentiall y not only confirmed previous clinical and an imal fmdings

with respect to the ab ility of ß-endorphin to suppress the signs and symptoms of

opiate withdrawal but also provides some useful information concerning the dosage

and adverse effects of ß-endo rphin in humans, Th e effectiveness of ß-endorph in on

att enuating morphine abstinence was clearly established on a doubl e-blind basis.

Preconditioning the patient by bolus injections or infusing with saline appeared to

eliminate the positive elements ofsuggestibility evoked by these manoeuvres.

Doses between 8-16 mg at a concentration of 10 mg mi-I yielded positive effects

when the total dose was injected within a Imin period but similar doses were

ineffective when diluted tenfold and infused over a 10 min period. It was concluded

that the latter procedure did not allow for the attainment of sufftcient amounts ofthe

drug at receptor sites for mediating pharmacological effects because ofrap id biotransformation in the body.

The study prov ides range-fmding information concerning the dosage level of

ß-endorphin that can be used in humans. lt appears that 15 mg can be safely

administered even when injected fairly rapidly. Other than producing a temporary

feeling of constriction in the ehest and mild abdominal cramping the adverse effects

ofthis dosage appear to be min imal. lt has been indicated that 30 mg of ß-endorphin

infused over a 30 min period is well-tolerated and alleviates signs and symptoms of

methadone withdrawal (Catlin , Gorelick, Gerner, Hui & Li, 1980).

The fact that ß -endorphin can .suppress morphine abstinence suggests that

ß-endorphin can transverse the blood /brain barrier but for the present it can not be

excluded that the effects could have been mediated peripherally.

Ca++ hypothesis for the mechanism of

opiate tolerance and physicaI dependence

The deve!opment of tolerance to and physical dependence on opiate drugs after

repeated and frequent administrations are strong reinforcing factors that contribute

to their addiction liability. Despite intensive work, the mechanisms responsible for

376 E. L. WAY

these phenomena have not been established. Several general theories oftolerance and

physical dependence have been proposed, but the specific biochemical processes that

might be involved have not been delineated. As an approach, the interactions

between Car' and opioids in the brain have been investigated.

There are cogent reasons for attempting to establish a causaI relationship between

Ca++ and opiate action. In the brain, Ca++ is a vital ion that not only is

concerned with regulating the activity of the neurone and its excitable state but also

with controlling the release of neurotransmitters. Considerable evidence is at hand

indicating that Ca++ and opiate drugs can interact to produce important physiological and biochemical changes. Although there are many agents which can affect

Ca++ disposition the evidence to date indicates that opiate effects are highly

selective. The possible link between opiates and Ca++ is provided by evidence

indicating that morphine inhibits the release of acetylcholine (Paton, 1957;

Schauman , 1957), noradrenaline (Henderson, Hughes & Kosterl itz, 1975) dopamine

(Celsen & Kuschinsky, 1974) and the endorphins may do likew ise (Loh , Brase,

Sampath-Khanna, Mar & Way, 1976a; Jhamandas, Sawynok & Sutak, 1977). That

opiates may produce their pharmacological effects by altering the disposition of'Catt

in the neurone is supported by three main pieces ofevidence:

I) The pharmacological effects ofopiates are reduced by Ca++ and augmented by Ca++

antagonists.

2) The Ca++ content in enriched nerve endings (synaptosomes) is reduced by acute

opiate administration.

3) Synaptosomal Ca++ content is gradually increased by continuous opiate

administration and the increase is proportional to the degree of tolerance and

physical dependence development.

Based on the assessment of these Catt-morphine interactions, a hypothesis to

explain acute and chronic opiate action has been proposed. The thesis is that there

are two opposing effects of opiates on neuronal Cart, that is, an immediate effect to

lower Ca++ levels and a delayed effect which reflects a counter adaptive response to

reverse the acute lowering effect on Ca++. These two opposing actions of morphine

can be utilized as a framework to explain its classic effects, namely, analgesia,

tolerance and physical dependence.

The operational hypothesis is that the excitable state is regulated by the Car" level

within the neurone, a lowering effects analgesia, while an elevation in Ca++ results in

hyperalgesia. The lowering of neuronal Ca++ levels induced by acute opiate

administration is opposed by a homeostatic mechanism which tends to reverse the

reduction in Ca++ levels . This latter process is cumulative so that with

continuous opiate administration there is a gradual build up of neuronal Ca++. The

consequence would be tolerance development since more opiate would be

required to lower the elevated neuronal Ca++ levels to effect analgesia. Under such

conditions, a new elevated steady state for Ca++ becomes established whereby

lowering of Ca++ content becomes more difficult and the ability to retain Ca++

becomes enhanced but requires the presence ofthe opiate (physical dependence). The

abstinence syndrome would then reflect a supersensitive state to opiate lack or a

hyperirritable response to Ca++ excess that is ordinarily inhibited by the presence of

opiate. These conjectures have gained substance by the evidence that has been

collected with ß-endorphin.

Acute effects ofmorphine and ß-endorphin on brain Ca++

It has been demonstrated previously that acute administration of morphine reduces

the Ca++ content in the nerve ending fract ion (synaptosomes) of brain homogenates

from the rat by 20-30% (Harris, Yamamoto, Loh & Way, 1977; Yamamoto, Harris,

Loh & Way, 1978; Guerrero-Munoz, Cerreta, Guerrero & Way , 1979a). Fraction-

ß-ENDO RPHIN IN PHYSICAL DEPE NDE NCE 377

ation of the subcellular components of mice synaptosomes revealed a selective

depletion ofCa " in the synaptic vesicles and the synaptic plasma mernbrane (Harris

et al., 1977). In vitro binding studies performed with 4SCa++ indicated an increase in

availability of ca- low affm ity binding sites in synaptic vesicles and increased high

affmity binding sites in synaptic plasma membrane (Yamamoto, Harris, Loh & Way,

1978).

Morphine also reduced brain Car' by inhibiting Ca++ uptake. Synaptosomes

derived from mice given an acute injection ofmorphine exhibited decreased ability to

take up 4SCa++ and decreased 45Ca++ uptake could also be demonstrated in

synaptosomes from untreated animals preincubated with IO...{, M morphine

(Guerrero-Munoz et al., 1979a). Decreased uptake of 4SCa++ was also noted in Iysed

synaptosomes pretreated with morphine and the process was found to be ATPdependent (Guerrero-Munoz, Guerrero & Way, 1979b). Thus, the decrease in brain

Catt content after acute morphine administration can be attributed in part to

inhibition of synaptosomal uptake and to an inhibition of binding by synaptic

vesicles and synaptic plasma membrane. That these responses to morphine on brain

Ca++ are selective opiate agonist effects are indicated by the fact that all of the above

described effects can be completely reversed in vivo or in vitro by the specific narcotic

antagonist, naloxone. Moreover, the acute opiate effects are seen with the active

opiate agonist levorphanol but not by its inactive enantiomer dextrorphan (Cardenas

& Ross , 1976; Guerrero-Munoz et al., 1979a).

ß-endorphin, similar to the opioid agonist alkaloids, also reduced 4SCa++ uptake in

vivo and in vitro (Guerrero-Munoz, Guerrero & Way, 1979c). To demonstrate the

effect of ß-endorphin on synaptosomal Car' uptake in vivo, unanaethetized mice

were injected intracerebrally with 7 pmol (25 ng in 5 lJ1 of 0.9% saline), a dose that

produced significant analgesia after 30 min by the tail-flick procedure. Immediately

after this interval, the mice were killed and the uptake of 4SCa++ was determined.

ß-endorphin was found to decrease the synaptosomal 4SCa++ uptake by about 20%

and the effect was apparent within 2 min . Although the effect was dose-dependent,

the changes in response to increasing doses of ß-endorphin became less apparent

when maximum inhibition was attained at 10 min. Administration ofnaloxone ( 2

mg kg-I

) subcutaneously antagonized the blockade of 4SCa++ uptake induced by

ß-endorphin. Furthermore, the inhibitory effect of ß-endorphin on 4SCa++ uptake

was reflected by a decrease in synaptosomal Ca++ content.

In vitro ß-endorphin at a concentration of IO- M also reduced significantly th e

45Ca++ uptake; thi s con cent ration is 100 times lower than the concent ration of

morphine required to block uptake. The specificity of op iate agonist action was

sho wn by the fact that the inh ibition of 4sCa++ uptake by ß-endorphin was reversed

by naloxone. When 10-7 M ß-endorphin was incubated with synaptosomes in the

presence of 1.9 x 10-8 M naloxone , the initial and maximum 45Ca++ upt ake was nearl y

identical to that of the control group. Naloxone by itself did not affect the

synaptosomal 4SCa++ content.

AIterations inneuronal Ca++ and acute opiate action

To support that lowering of cellular Ca++ results in hypoalgesia and raising cellular

Ca++ results in hyperalgesia, it has been demonstrated that manipulations to lower

neuronal Ca++, other than with opiates, also produced analgesia. Thus chelating

available Ca++ with EGTA or preventing neuronal Ca++ entry with La+++ not

only produces antinociceptive effects but also enhances the action of opiates (Harris,

Loh & Way, 1975a; Schmidt & Way, 1980). Their potency is less than that ofopiates,

likely because they are less selective in their mechanism of decreasing neuronal Ca++

content. Opiates appear to act at selective Ca++ pools that include the inner synaptic

378 E. L. WAY

plasma membrane and synaptic vesicles (Harris et a/., 1977b; Yamamoto et al., 1978).

The similarities between morphine and La+++ induced antinociception extended to

common subcortical sites of action. Unilateral microinjection of morphine or La+++

into the mesencephalic periaqueductal grey region (PAG) of thc rat which were

reduced after PAG microinjection of Cart (Harri s, Iwarnoto, Loh & Way, 1975b;

lwarnoto, Harris, Loh & Way, 1978).

In contrast, opposite etTects are produced by manoeuvres that increase neuronal

Ca++. High neuronal Ca++ should produce hyperalgesia and antagonize opiate and

La+++ action. In agreement with this supposition, the intraventricular injection of

Car' not only produced hyperalgesia but also increased the amount of opiates or

La+++ needed to produce analgesia (Schmidt & Way, 1979). Earlier, it was reported

(Harris et al., 1975a) that facilitating Ca++ entry with the ionophore X53 7A, enhanced

the anti morphine action of Ca++. Also, antagonism of opiate analgesia can be

obtained by X537 A and Ca++ in combination at a dose ofeach agent that is ordinarily

inetTectivewhen either is used alone.

Chronic effects ofmorphine and ß-endorphin on brain Ca++

It has been shown that subcutaneous implantation of a morphine pellet in mice to

induce tolerance and physical dependence, results in a selective increase in Car'

content. The elevation in Ca++ was found in synaptic vesicles and Iysed synaptic

plasma membranes but not in the crude nuclear, myelin , microsomal or extrasynaptosomal mitochrondrial fractions. The increase in synaptosomal Ca++

paralleled the time course of tolerance development and was prevented by the

implantation of a naloxone pellet together with the morphine pellet to block tolerance and dependence development. In vitro bind ing studies with 4SCa++ revealed no

change in binding with intact synaptosomes but less 4SCa++ could be found on

synaptic plasma membrane and synaptic vesicles. Presumably with the increased

occupancy of endogenous Car' at these loci, less sites were available for binding 4SCa

(Harris et a/., 1977; Yamamoto et al., 1978).

The increase in synaptosomal Ca++ after tolerance and physical dependence

development was also due in part to enhanced uptake. The initial rate of

synaptosomal uptake of<Ca increased linearly with increas ing duration ofmorphine

pellet implantation (Guerrero-Munoz et a/., 1979a). When 4SCa++ uptake was studied

in Iysed synaptosomes after tolerance had been induced, an enhancement of 4SCa++

uptake was observed in the presence but not in the absence ofATP indicating that an

active process is involved in the etTect. The increase in uptake was proportional to the

degree oftolerance and physical dependence and could be prevented by the implantation of a naloxone pellet together with the morphine pellet (Guerrero-Munoz et al.,