- Email: [email protected]
Failure of the gamma-aminobutyric acid (GABA) derivative, baclofen, to stimulate growth hormone secretion in heroin addicts
Life Sciences, Vol. P r i n t e d in the U S A
51, pp.
247-251
Pergamon
Press
FAILURE OF THE GAMMA-AMINOBUTYRIC ACID (GABA) DERIVATIVE, BACLOFEN, TO STIMULATE GROWTH HORMONE SECRETION IN HEROIN ADDICTS. R. Volpi, G. Gerra, S. Vourna, P.P. Vescovi, D. Maestri**, P. Chiodera*, V. Coiro Chair of Medical Clinic, Endocrinology*, University of Parma, RIA Laboratory**, Hospital of Parma, Parma, Italy. (Received
in final
form May 15,
1992)
S ummary.
In order to establish possible alterations in the gamma aminobutyric acid (GABA)ergic control of growth hormone (GH) secretion in heroin addicts, ten patients (age, 25.8 + 1.07 yr (mean + SE); duration of heroin addiction, range 3-8 yr; weight, 67.3 + 0.87 kg body weight), and ten age (29.1 + 0.84 yr)- and weight (69.7 + 0.87 kg)-matched normal controls were tested with the GABAergic Breceptor agonist baclofen (10 mg p.o. at 09.00 h) (experimental test) or a placebo (control test). Blood samples for GH assay were taken every 15 rain for the next 150 min. Normal controls underwent one control and one experimental test. Heroin addicts were submitted to both baclofen and placebo test twice, once around the time of their admission to a recovery community for drug abusers, when they were still assuming heroin, and again after two months of permanence in the community. From the time of their admission to the community, the patients were forbidden to use heroin. For two weeks after admission they were treated with clonidine and acetylsalicilic acid to attenuate withdrawal symptoms. Thereafter, the patients underwent a period of wash-out of pharmacological treatments for at least 6 weeks before being retested. Basal GH levels were similar in normal controls and heroin addicts in all tests and remained unmodified during control tests in all subjects. The administration of baclofen increased four times the serum GH levels within 120 minutes in the normal controls, whereas it did not modify serum GH concentrations in heroin addicts either during the period of drug abuse or after two months of abstinence. These data show that the control of GH secretion mediated by GABAergic B-receptors is impaired in heroin addicts. It is hypothesized that this neuroendocrine alteration might represent a trait marker of heroin addiction, or more likely, that it was a consequence of a long addiction to heroin persisting after two months of abstinence. Alterations of hypothalamic-pituitary function have been described in narcotic addicts (1). Particularly, these patients show altered GH secretory patterns in various experimental conditions (2, 3). In the present study, we focused our attention on the gamma aminobutyric acid (GABA)ergic control of GH secretion in heroin addicts. Complex and still unclear interactions between opioid and GABAergic pathways have been described at various levels in the central nervous system (4, 5, 6). Since both GABA and opioids influence GH secretion, we wondered whether heroin addiction modifies the GABAergic control of GH secretion. In order to answer this question, the GH response to the GABAergic B-receptor agonist baclofen (7, 8, 9) was evaluated in male heroin addicts during drug abuse. The same subjects were retested after two months of abstinence, when the effects of acute heroin assumption or the possible neuroendocrine changes associated to drug withdrawal had ceased. Correspondence: R. VOLPI, MD - Chair of Medical Clinic, Via Gmmsci 14 - 43100 PARMA, Italy
0024-3205[92 $ 5 . 0 0 + . 0 0 Copyright
© 1992 Pergamon
Press Ltd
All rights
reserved.
248
GH Response
to B a c l o f e n
in H e r o i n A d d i c t s
Vol.
51, No.
4, 1992
Materials and methods Ten normal men (age: 29.1 + 0.84 yr (mean + SE); weight: 69.7 + 0.87 kg) and ten heroin addicts (age: 25.8 + 1.07 yr (mean + SE); weight: 67.3 + 0.87 kg) volunteered to participate in the study. All the subjects were informed of the purpose of the study and gave their informed consent. The study was in accordance with the Helsinki II declaration. The heroin addicts were recruited when they came to the center for health control of drug addicts of Parma (Italy) and they asked to be admitted to a recovery community for drug abusers. According to availability of admission into the community, subjects waited 5-7 days before taking shelter. These subjects had been heroin abusers for at least 5 years (mean + SE: 5+ 1.2 yr). At the time of the study, they were assuming heroin at doses ranging between 1 and 1,5 gr/day. Heroin was taken off as soon as the patients entered the community. After endocrine tests, patients were treated for 2 weeks with clonidine and acetylsalicilic acid because of symptoms of heroin withdrawal. Thereafter, the patients were not allowed to assume any drug. While they were in the community, patients were under careful control. Furthermore, urine samples were obtained twice a week and were used for toxicological assays. Subjects with positive urinary tests for heroin catabolites were excluded from the study. All men were within 10% of their ideal body weight. All the subjects were well nourished and without clinical or laboratory evidence of renal, cardiovascular or endocrine abnormalities. In all subjects the laboratory parameters of liver function were within the normal range at the time of the study, even though some of the heroin addicts had been affected by hepatitis 2-4 years before this study. All subjects were assessed with the Hospital Anxiety and Depression scale (10), the Research Diagnostic Criteria (11) and the Hamihon depression rating scale (12) with negative results. T¢~tS. Normal men were tested twice, once with baclofen (Lioresal, Ciba-Geigy, Italy; experimental test) and another time with a placebo (control test). Tests were carried out in random order with an interval of at least five days. Heroin addicts were tested with baclofen or placebo on the morning of the day following their first coming to the center for health control of drug addicts and 5-7 days later on the morning of their admission to the community center. On these occasions patients were in similar conditions of drug abuse, as evidenced by positive results of urinary tests for opiate catabolites. Tests were performed double blind in random order. After two months of permanence in the community, the patients were retested with baclofen or placebo in random order on two different days at 5 days intervals. On the morning of the tests urine samples were taken for toxicological assay. Experimental procedt, re. At 09.00 h of the experimental day, a 19-gauge intravenous indwelling needle was inserted into an antecubital vein of subjects lying in the recumbent position and fasting from the previous evening. The needle was kept patent with a slow infusion of normal saline. Blood specimens were taken at time 0 (30 rain. after the insertion of the cannula), and at time 30, 60, 90, 120 and 150 minutes. Ten nag baclofen or a placebo was administered p.o. at time 0just after blood sampling. Blood pressure and heart rate were measured with a chronometer and a mercury sphygmomanometer respectively, at each sampling time during tests. Blood samples were centrifuged after each experiment; serum was separated and stored at 20 ° C until assayed. Serum GH concentrations were measured with a specific radioimmunoassay, using the double antibody tecnique (13). The intra- and inter-assay coefficients of variation were respectively 4.2% and 4% and the lower limit of sensitivity of the assay was 0.24 ng/ml. Plasma LH (14) and FSH (15),testosterone (T) (16) and 17-[3-estradiol (E2) (17) concentrations were measured in samples taken at time 0 of the tests using specific RIA methods. All samples from the same subjects were measured in the same assay in duplicate and in random order. Toxicological assays in urine consisted of measurements of opiate catabolites in 24 h samples, utilizing an immunoenzymatic method with reagents supplied by Bracco (Milan, Italy). Statistics. Results were analyzed by two-way or three-way analysis of variance (ANOVA) with repeated measures and Wilcoxon's matched-pair rank sum test, as appropriate. Values are expressed as mean +SE.
Vol.
51, NO.
4, 1992
GH R e s p o n s e
to B a c l o f e n
in H e r o i n A d d i c t s
249
Table 1 S e r u m G H concentrations (ng/ml) d u r i n g placebo and baclofen test in n o r m a l controls and heroin addicts. ° M e a n + S E o f l 0 observations. * p<0.01 vs baseline. TIME(min)
-15
0
30
60
120
150
1.87i-0.06 6.1 l!O.07 ~'
1.83i-0.07 7.01:t0.07"
1.86:t0.05 6.05:k0.07"
2.43+0.21 2.25+0.17
2.52:t0.20 2.45±0.27
2.44±0.19 2.42._+0.24
2.41i-0.21 2,36i'0.20
2.20±0.06 2.73:1:0.31
2.39:1-0.36 2.59i-0.24
2.52!-0.17 2.51i0.20
2,18:L'0.08 2.60:£0.20
Normal Controls Placebo ° Baclofen test °
1.67_+0.08 1.77_+0.09
1.63±0.09 1.73i-0.11
1.64_+0.07 2.81_+0.07
137±0.08 5.21:t:0.09"
tleroin Addicts before abstinence Placebo ° Baclofen test °
2.20:~0.09 2.58+1.07
2.12±0.09 1.98±O.08
2.29i-0.16 2.11±0.09
Heroin Addicts after abstinence Placebo ° Baclofcn leSt °
2.15_+0.07 2.38_-,20.22
2.36:t0.09 2.35_+0.20 2.43_+0.26 2.51_+0.25
90
Results Normal subjects and heroin addicts showed similar plasma T, E 2, LH and FSH levels (heroin addicts: time 0 of the first control test: T= 9.48 + 0.35 ng/ml (mean _+ SE), E2= 21.50 + 2.05 pg/ml, LH= 5.34 +_ 0.20 mIU/ml, FSH= 2.66 +_0.38 mlU/ml; second control test (after two months of abstinence): T= 9.05 + 0.21 , E2= 20.1 + 1.85, LH= 5.57 -+ 0.23, FSH= 3.05 + 0.40; normal controls: T= 8.54 _+ 0.35, E2= 19.6 +- 1.6, LH= 4.73 _+ 0. 16, FSH--- 2.89 + 0.17).
800,//////////
600~/////////~
X
._. E
Z/////X//~ ~/U///////~
--1---
400-
X ZJ/4//Z/ZZ/
e
E
200-
(4////////~ '/////////2"/. ~///Z//Z/z/.g Z/////Z//~
~/////~//, z l l l ~ z z / t l l
NORMALS
HEROIN ADDICTS BEFORE ABSTINENCE
HEROIN ADDICTS AFTER ABSTINENCE
Fig. 1 GH secretory areas during placebo (empty bars) and baclofen (full bars) tests in controls and heroin addicts. Each bar represents the mean + SE of the observations. *p < 0.01 Baclofen vs Placebo test.
In heroin addicts the urinary assays were positive for the presence of heroin catabolites at the time of both first control and first experinaental test; in contrast, they gave always negative results during the two months of abstinence and when they were performed on the morning of both second control and second experimental test. The basal concentrations of GH were similar in normal controls and heroin addicts (Table 1). During all control tests, GH levels remained unmodified in normal men and heroin addicts (Table 1). The administration of baclofen in the normal subjects induced a significant increase in serum GH levels (Fig. 1), with a mean peak response at 120 min (p<0.01 vs basal value; F= 9.89,
250
GH Response
to B a c l o f e n
in H e r o i n A d d i c t s
Vol.
51, No.
4, 1992
p<0.01 vs control test (Table 1). In contrast, baclofen was unable to increase the circulating concentrations of GH in heroin addicts both before and after two months of abstinence (NS vs time 0 and vs control tests) (Table 1 and Fig. 1). Baclofen did not produce side effects in any subjects. Blood pressure and heart rate remained unmodified during all tests. Discussion Studies in heroin addicts have provided evidence of wide alterations in hypothalamicpituitary function. Various hormonal systems appear to change in heroin addicts; in fact, reduced PRL (1) and ACTH/cortisol (2) responses to releasing stimulations have been described. Narcotic addicts have been also shown to have subnormal GH responses to various GH releasing stimuli, such as arginine- and insulin-induced hypoglycemia (2). In addition, in the present study we found that baclofen is unable to stimulate GH secretion in heroin addicts. Morphine has been shown to suppress the GH rise in response to surgical stress (3); however, the possibility that the blunted GH response to baclofen observed in our heroin addicts was because of recent heroin assumption is unlikely, since GH remained unresponsive to stimulation with the GABAergic Breceptor agonist even when the subjects were tested after two months of abstinence. After such a long period of abstinence, the possible interferences attributable to drug withdrawal can be excluded. Furthermore, the possibility that the use of clonidine [a wellknown GH stimulator (18)] during the first two weeks of abstinence might have influenced the GH response to baclofen is unlikely, since the baclofen tests were carried out six weeks after clonidine withdrawal. Baclofen is known to stimulate GH secretion in men, but not in women (19), suggesting that sex steroids modulate the GH response to GABAergic B-receptor stimulation. The finding of normal plasma sex steroid concentrations in our subjects excludes the possibility that baclofen was unable to stimulate GH because of heroin-induced hypogonadism. The results of this study suggest GABAergic B-receptor alterations in the control of GH secretion in heroin addicts. This GABAergic alteration might be supposed to represent a trait marker of subjects addicted to heroin, but this hypothesis needs to be substantiated. More likely, the derangement of GABAergic B receptor-mediated neurotransmission was the consequence of a long term addiction to heroin. Even though there is no conclusive evidence about the mechanism underlying the GABAergic control of GH secretion, GABA and GABAergic drugs are supposed to produce their GH releasing effect through inhibition of somatostatin release from the hypothalamus (20, 21); in contrast, a direct effect of GABA on the pituitary somatotrophs has been excluded by in vitro studies (22). Also opioids show stimulatory effects on GH secretion (23-27). Interestingly, both GABA (28, 29) and opioids (18) inhibit the hypothalamic dopaminergic tone, and hence probably reduce somatostatin release (30, 31). In view of the possible interactions between GABAergic and opioid neurotransmission, the effect of GABA on GH has been supposed to be mediated by opioid receptors (4). However, in contrast with this hypothesis, naloxone has been found unable to change the GH releasing effect of baclofen (32). On the other hand, it is likely that opiates interact with GABAergic receptors because the opiate alkaloids are known to displace 3HGABA bound to synaptic membranes (33, 34). Furthermore, GABA B-receptor agonists (5) and antagonists (35) have been found to modify opioid actions in the central nervous system. Therefore, a possible involvement of GABA B-receptors in the mechanism underlying the effect of opioids on GH secretion may be supposed. A long term addiction to morphine is known to decrease endorphin and methionine enkephalin levels in the brain and pituitary (36). These changes are not reversed after protracted abstinence from morphine (36). We suppose that in our patients a chronic addiction to heroin induced opioid alterations at the hypothalamic level, and thus produced GABAergic B receptor changes, which were unmasked by the absent GH response to baclofen. These neuroendocrine changes appear to persist after two months of abstinence.
Vol.
51, No. 4, 1992
GH Response to Baclofen
in Heroin Addicts
251
R~ferences 1.
M.L. WILLEMBRING , J.E. MORLEY , D.D. KRAHAN , G.A. CARLSON , A.S. LEVINE,R.B. SHAPER, Psychoneuroendocrinology 14 371-391 (1989). 2. P. CUSHMAN, J. Clin. Endocrinol. Metab. 35 352-358 (1972). 3. J.M. GEORGE, C.E. RIER, R.A.LANESE, J.M. ROWER, J. Clin. Endocrinol. Metab. 38 736-741 (1976). 4. F. MORONI D.L. CHENEY, E. PERALTA, E. COSTA, J. Pharmacol. Exp. Ther. ~Q7 870-874 (1978) 5. J. SAWYNOK, F.S. LABELLA, Eur. J. Pharmacol 70 103-110 (1981). 6. H. Katakami, Y. Kato, N. Matsushita, A.Shimatsu, H. Imura, Endocrinology 109 10331036 (1981). 7. P.H. KELLY, K.E. MOORE, Neuropharmacology 17 169-174 (1978). 8. N.E. ANDEN and H. WACHTEL, Acta Pharmacol. Toxicol. (Kbh) 40 310-320 (1977). 9. H.U. FEHR and H.J. BEIN, J. Int. Med. Res.2 36-39 (1974). 10. A.S. ZIGMOND and R.P. SNAITH, Acta Psychiatr. Scand. 67 361-370 (1983). 11. R.L. SPITZER, J. ENDICOTT, E. ROBINS, Arch. Gen. Psychiatry 35 773-782 (1978). 12. M. HAMILTON, J. Neurol. Neurosurg. Psychiatry 23 56-62 (1960). 13. W.D. ODELL, P.L. RAYFORD, G.T. ROSS, J. Lab. Clin. Med. 70 973-980 (1967). 14. M.G. F O R E S T , A.M. CATHIARD, J.A. BERTRAND, J. Clin. Endocrinol. Metab. 36 1132-1142 (1973). 15. R. DE HERTOG, K. THOMAS, Y. BIETLOT I. VANDERHEYDEN, J. FERIN, J. Clin. Endocrinol. Metab. 40 93-101 (1975). 16. A.R. MIDGLEYjr, Endocrinology 79 10-18 (1966). 17. A.R. MIDGLEYjr, J. Clin. Endocrinol. Metab. 27 295-299 (1967). 18. J. TUOMISTO, P. MANNISTO, Pharmacol. Rev. 37 288-293 (1985). 19. P. MONTELEONE, M. MAJ, M. IOVINO L. STEARDO, Acta Endocrinologica 119 353357 (1988). 20. J. TAKAHARA, S. YUNOKI, H. HOSOGI W. YAKUSHIJI, J. KAGEYAMA, T. OFUJI, Endocrinology 106 343-347 (1980). 21. T.D. STRYKER, T. CONLIN, S. REICHLIN, Brain Res. 362 339-343 (1986). 22. R.A. ANDERSON and R. MITCHELL, J. Endocrinol. 108 1-7 (1986). 23. A. DUPONT, L. CUSAN, M. GARON, C.H. LI, Proc. Natl. Acad. Sci. (USA) 74 358359 (1977). 24. C. RIVIER, W. VALE, N. LING, M. BROWN, R. GUILLEMIN, Endocrinology 100 238-241 (1977). 25. Y. KATO, Y. IWASAKI, H. ABE, S. OHGO, H. IMURA, Proc. Soc. Exp. Biol. Med. 158 431-436 (1978). 26. B.VON GRAFFENRIED, E. DEL POZO, J. ROUBICEK, E. KREBS, W. POLDINGER, P. BURMEISTER, L. KERP, Nature 272 729-730 (1978). 27. Y. KATO, H. KATAKAMI, H. 1MURA, In: Growth ~and Growth Factor~. K. Shizume, K. Takano (eds), 159-167, University Tokyo Press, Tokyo, (1980). 28. F. CASANUEVA, J.A. APUD A. LOCATELLI, A. MARTINEZ-CAMPOS, C. CIVATI, G. RACAGNI, D. COCCHI, E.E. MULLER, Endocrinology |09 567-575 (1981). 29. J.R. WALTERS, R.H. ROTH, G.K.AGHAJANIAN, J. Pharmacol. Exp. Ther. 18(; 630639 (1973). 30. K. CHIHARA, A. ARIMURA, Abstract of the 60~h Anmlal Meetin~ of the Endocrin~ Society, 195 (1978). 31. G. RACAGNI, J.A. APUD, D. COCCHI, V. LOCATELLI, E.E. MULLER, Life Sci 31 823-838 (1982). 32. P. CHIODERA, R. VOLPI, V. COIRO, L. BARILLI, G. ROSSI, E. ROTI, J. Endocrinol. Invest. 6 381-384 (1983). 33. R. DINGLEDINE, L.L. IVERSEN, E. BREUKER, Europ. J. Pharmacol. 47 19-27 (1978). 34. M.A. WERZ, K. BAUM, A.B. YOUNG, R.L. MACDONALD, Neurosci. Abstr. 7 478 (1981). 35. E. ARENAS, J. MARSAL, J. ALBERCH, Neurosci. Lett. 120 201-204 (1980). 36. P.G. KRISHNAMURTHY, G.A TEJWANI, H.N. BHARGAVA, Brain Res. 553 284290 (1991).
51, pp.
247-251
Pergamon
Press
FAILURE OF THE GAMMA-AMINOBUTYRIC ACID (GABA) DERIVATIVE, BACLOFEN, TO STIMULATE GROWTH HORMONE SECRETION IN HEROIN ADDICTS. R. Volpi, G. Gerra, S. Vourna, P.P. Vescovi, D. Maestri**, P. Chiodera*, V. Coiro Chair of Medical Clinic, Endocrinology*, University of Parma, RIA Laboratory**, Hospital of Parma, Parma, Italy. (Received
in final
form May 15,
1992)
S ummary.
In order to establish possible alterations in the gamma aminobutyric acid (GABA)ergic control of growth hormone (GH) secretion in heroin addicts, ten patients (age, 25.8 + 1.07 yr (mean + SE); duration of heroin addiction, range 3-8 yr; weight, 67.3 + 0.87 kg body weight), and ten age (29.1 + 0.84 yr)- and weight (69.7 + 0.87 kg)-matched normal controls were tested with the GABAergic Breceptor agonist baclofen (10 mg p.o. at 09.00 h) (experimental test) or a placebo (control test). Blood samples for GH assay were taken every 15 rain for the next 150 min. Normal controls underwent one control and one experimental test. Heroin addicts were submitted to both baclofen and placebo test twice, once around the time of their admission to a recovery community for drug abusers, when they were still assuming heroin, and again after two months of permanence in the community. From the time of their admission to the community, the patients were forbidden to use heroin. For two weeks after admission they were treated with clonidine and acetylsalicilic acid to attenuate withdrawal symptoms. Thereafter, the patients underwent a period of wash-out of pharmacological treatments for at least 6 weeks before being retested. Basal GH levels were similar in normal controls and heroin addicts in all tests and remained unmodified during control tests in all subjects. The administration of baclofen increased four times the serum GH levels within 120 minutes in the normal controls, whereas it did not modify serum GH concentrations in heroin addicts either during the period of drug abuse or after two months of abstinence. These data show that the control of GH secretion mediated by GABAergic B-receptors is impaired in heroin addicts. It is hypothesized that this neuroendocrine alteration might represent a trait marker of heroin addiction, or more likely, that it was a consequence of a long addiction to heroin persisting after two months of abstinence. Alterations of hypothalamic-pituitary function have been described in narcotic addicts (1). Particularly, these patients show altered GH secretory patterns in various experimental conditions (2, 3). In the present study, we focused our attention on the gamma aminobutyric acid (GABA)ergic control of GH secretion in heroin addicts. Complex and still unclear interactions between opioid and GABAergic pathways have been described at various levels in the central nervous system (4, 5, 6). Since both GABA and opioids influence GH secretion, we wondered whether heroin addiction modifies the GABAergic control of GH secretion. In order to answer this question, the GH response to the GABAergic B-receptor agonist baclofen (7, 8, 9) was evaluated in male heroin addicts during drug abuse. The same subjects were retested after two months of abstinence, when the effects of acute heroin assumption or the possible neuroendocrine changes associated to drug withdrawal had ceased. Correspondence: R. VOLPI, MD - Chair of Medical Clinic, Via Gmmsci 14 - 43100 PARMA, Italy
0024-3205[92 $ 5 . 0 0 + . 0 0 Copyright
© 1992 Pergamon
Press Ltd
All rights
reserved.
248
GH Response
to B a c l o f e n
in H e r o i n A d d i c t s
Vol.
51, No.
4, 1992
Materials and methods Ten normal men (age: 29.1 + 0.84 yr (mean + SE); weight: 69.7 + 0.87 kg) and ten heroin addicts (age: 25.8 + 1.07 yr (mean + SE); weight: 67.3 + 0.87 kg) volunteered to participate in the study. All the subjects were informed of the purpose of the study and gave their informed consent. The study was in accordance with the Helsinki II declaration. The heroin addicts were recruited when they came to the center for health control of drug addicts of Parma (Italy) and they asked to be admitted to a recovery community for drug abusers. According to availability of admission into the community, subjects waited 5-7 days before taking shelter. These subjects had been heroin abusers for at least 5 years (mean + SE: 5+ 1.2 yr). At the time of the study, they were assuming heroin at doses ranging between 1 and 1,5 gr/day. Heroin was taken off as soon as the patients entered the community. After endocrine tests, patients were treated for 2 weeks with clonidine and acetylsalicilic acid because of symptoms of heroin withdrawal. Thereafter, the patients were not allowed to assume any drug. While they were in the community, patients were under careful control. Furthermore, urine samples were obtained twice a week and were used for toxicological assays. Subjects with positive urinary tests for heroin catabolites were excluded from the study. All men were within 10% of their ideal body weight. All the subjects were well nourished and without clinical or laboratory evidence of renal, cardiovascular or endocrine abnormalities. In all subjects the laboratory parameters of liver function were within the normal range at the time of the study, even though some of the heroin addicts had been affected by hepatitis 2-4 years before this study. All subjects were assessed with the Hospital Anxiety and Depression scale (10), the Research Diagnostic Criteria (11) and the Hamihon depression rating scale (12) with negative results. T¢~tS. Normal men were tested twice, once with baclofen (Lioresal, Ciba-Geigy, Italy; experimental test) and another time with a placebo (control test). Tests were carried out in random order with an interval of at least five days. Heroin addicts were tested with baclofen or placebo on the morning of the day following their first coming to the center for health control of drug addicts and 5-7 days later on the morning of their admission to the community center. On these occasions patients were in similar conditions of drug abuse, as evidenced by positive results of urinary tests for opiate catabolites. Tests were performed double blind in random order. After two months of permanence in the community, the patients were retested with baclofen or placebo in random order on two different days at 5 days intervals. On the morning of the tests urine samples were taken for toxicological assay. Experimental procedt, re. At 09.00 h of the experimental day, a 19-gauge intravenous indwelling needle was inserted into an antecubital vein of subjects lying in the recumbent position and fasting from the previous evening. The needle was kept patent with a slow infusion of normal saline. Blood specimens were taken at time 0 (30 rain. after the insertion of the cannula), and at time 30, 60, 90, 120 and 150 minutes. Ten nag baclofen or a placebo was administered p.o. at time 0just after blood sampling. Blood pressure and heart rate were measured with a chronometer and a mercury sphygmomanometer respectively, at each sampling time during tests. Blood samples were centrifuged after each experiment; serum was separated and stored at 20 ° C until assayed. Serum GH concentrations were measured with a specific radioimmunoassay, using the double antibody tecnique (13). The intra- and inter-assay coefficients of variation were respectively 4.2% and 4% and the lower limit of sensitivity of the assay was 0.24 ng/ml. Plasma LH (14) and FSH (15),testosterone (T) (16) and 17-[3-estradiol (E2) (17) concentrations were measured in samples taken at time 0 of the tests using specific RIA methods. All samples from the same subjects were measured in the same assay in duplicate and in random order. Toxicological assays in urine consisted of measurements of opiate catabolites in 24 h samples, utilizing an immunoenzymatic method with reagents supplied by Bracco (Milan, Italy). Statistics. Results were analyzed by two-way or three-way analysis of variance (ANOVA) with repeated measures and Wilcoxon's matched-pair rank sum test, as appropriate. Values are expressed as mean +SE.
Vol.
51, NO.
4, 1992
GH R e s p o n s e
to B a c l o f e n
in H e r o i n A d d i c t s
249
Table 1 S e r u m G H concentrations (ng/ml) d u r i n g placebo and baclofen test in n o r m a l controls and heroin addicts. ° M e a n + S E o f l 0 observations. * p<0.01 vs baseline. TIME(min)
-15
0
30
60
120
150
1.87i-0.06 6.1 l!O.07 ~'
1.83i-0.07 7.01:t0.07"
1.86:t0.05 6.05:k0.07"
2.43+0.21 2.25+0.17
2.52:t0.20 2.45±0.27
2.44±0.19 2.42._+0.24
2.41i-0.21 2,36i'0.20
2.20±0.06 2.73:1:0.31
2.39:1-0.36 2.59i-0.24
2.52!-0.17 2.51i0.20
2,18:L'0.08 2.60:£0.20
Normal Controls Placebo ° Baclofen test °
1.67_+0.08 1.77_+0.09
1.63±0.09 1.73i-0.11
1.64_+0.07 2.81_+0.07
137±0.08 5.21:t:0.09"
tleroin Addicts before abstinence Placebo ° Baclofen test °
2.20:~0.09 2.58+1.07
2.12±0.09 1.98±O.08
2.29i-0.16 2.11±0.09
Heroin Addicts after abstinence Placebo ° Baclofcn leSt °
2.15_+0.07 2.38_-,20.22
2.36:t0.09 2.35_+0.20 2.43_+0.26 2.51_+0.25
90
Results Normal subjects and heroin addicts showed similar plasma T, E 2, LH and FSH levels (heroin addicts: time 0 of the first control test: T= 9.48 + 0.35 ng/ml (mean _+ SE), E2= 21.50 + 2.05 pg/ml, LH= 5.34 +_ 0.20 mIU/ml, FSH= 2.66 +_0.38 mlU/ml; second control test (after two months of abstinence): T= 9.05 + 0.21 , E2= 20.1 + 1.85, LH= 5.57 -+ 0.23, FSH= 3.05 + 0.40; normal controls: T= 8.54 _+ 0.35, E2= 19.6 +- 1.6, LH= 4.73 _+ 0. 16, FSH--- 2.89 + 0.17).
800,//////////
600~/////////~
X
._. E
Z/////X//~ ~/U///////~
--1---
400-
X ZJ/4//Z/ZZ/
e
E
200-
(4////////~ '/////////2"/. ~///Z//Z/z/.g Z/////Z//~
~/////~//, z l l l ~ z z / t l l
NORMALS
HEROIN ADDICTS BEFORE ABSTINENCE
HEROIN ADDICTS AFTER ABSTINENCE
Fig. 1 GH secretory areas during placebo (empty bars) and baclofen (full bars) tests in controls and heroin addicts. Each bar represents the mean + SE of the observations. *p < 0.01 Baclofen vs Placebo test.
In heroin addicts the urinary assays were positive for the presence of heroin catabolites at the time of both first control and first experinaental test; in contrast, they gave always negative results during the two months of abstinence and when they were performed on the morning of both second control and second experimental test. The basal concentrations of GH were similar in normal controls and heroin addicts (Table 1). During all control tests, GH levels remained unmodified in normal men and heroin addicts (Table 1). The administration of baclofen in the normal subjects induced a significant increase in serum GH levels (Fig. 1), with a mean peak response at 120 min (p<0.01 vs basal value; F= 9.89,
250
GH Response
to B a c l o f e n
in H e r o i n A d d i c t s
Vol.
51, No.
4, 1992
p<0.01 vs control test (Table 1). In contrast, baclofen was unable to increase the circulating concentrations of GH in heroin addicts both before and after two months of abstinence (NS vs time 0 and vs control tests) (Table 1 and Fig. 1). Baclofen did not produce side effects in any subjects. Blood pressure and heart rate remained unmodified during all tests. Discussion Studies in heroin addicts have provided evidence of wide alterations in hypothalamicpituitary function. Various hormonal systems appear to change in heroin addicts; in fact, reduced PRL (1) and ACTH/cortisol (2) responses to releasing stimulations have been described. Narcotic addicts have been also shown to have subnormal GH responses to various GH releasing stimuli, such as arginine- and insulin-induced hypoglycemia (2). In addition, in the present study we found that baclofen is unable to stimulate GH secretion in heroin addicts. Morphine has been shown to suppress the GH rise in response to surgical stress (3); however, the possibility that the blunted GH response to baclofen observed in our heroin addicts was because of recent heroin assumption is unlikely, since GH remained unresponsive to stimulation with the GABAergic Breceptor agonist even when the subjects were tested after two months of abstinence. After such a long period of abstinence, the possible interferences attributable to drug withdrawal can be excluded. Furthermore, the possibility that the use of clonidine [a wellknown GH stimulator (18)] during the first two weeks of abstinence might have influenced the GH response to baclofen is unlikely, since the baclofen tests were carried out six weeks after clonidine withdrawal. Baclofen is known to stimulate GH secretion in men, but not in women (19), suggesting that sex steroids modulate the GH response to GABAergic B-receptor stimulation. The finding of normal plasma sex steroid concentrations in our subjects excludes the possibility that baclofen was unable to stimulate GH because of heroin-induced hypogonadism. The results of this study suggest GABAergic B-receptor alterations in the control of GH secretion in heroin addicts. This GABAergic alteration might be supposed to represent a trait marker of subjects addicted to heroin, but this hypothesis needs to be substantiated. More likely, the derangement of GABAergic B receptor-mediated neurotransmission was the consequence of a long term addiction to heroin. Even though there is no conclusive evidence about the mechanism underlying the GABAergic control of GH secretion, GABA and GABAergic drugs are supposed to produce their GH releasing effect through inhibition of somatostatin release from the hypothalamus (20, 21); in contrast, a direct effect of GABA on the pituitary somatotrophs has been excluded by in vitro studies (22). Also opioids show stimulatory effects on GH secretion (23-27). Interestingly, both GABA (28, 29) and opioids (18) inhibit the hypothalamic dopaminergic tone, and hence probably reduce somatostatin release (30, 31). In view of the possible interactions between GABAergic and opioid neurotransmission, the effect of GABA on GH has been supposed to be mediated by opioid receptors (4). However, in contrast with this hypothesis, naloxone has been found unable to change the GH releasing effect of baclofen (32). On the other hand, it is likely that opiates interact with GABAergic receptors because the opiate alkaloids are known to displace 3HGABA bound to synaptic membranes (33, 34). Furthermore, GABA B-receptor agonists (5) and antagonists (35) have been found to modify opioid actions in the central nervous system. Therefore, a possible involvement of GABA B-receptors in the mechanism underlying the effect of opioids on GH secretion may be supposed. A long term addiction to morphine is known to decrease endorphin and methionine enkephalin levels in the brain and pituitary (36). These changes are not reversed after protracted abstinence from morphine (36). We suppose that in our patients a chronic addiction to heroin induced opioid alterations at the hypothalamic level, and thus produced GABAergic B receptor changes, which were unmasked by the absent GH response to baclofen. These neuroendocrine changes appear to persist after two months of abstinence.
Vol.
51, No. 4, 1992
GH Response to Baclofen
in Heroin Addicts
251
R~ferences 1.
M.L. WILLEMBRING , J.E. MORLEY , D.D. KRAHAN , G.A. CARLSON , A.S. LEVINE,R.B. SHAPER, Psychoneuroendocrinology 14 371-391 (1989). 2. P. CUSHMAN, J. Clin. Endocrinol. Metab. 35 352-358 (1972). 3. J.M. GEORGE, C.E. RIER, R.A.LANESE, J.M. ROWER, J. Clin. Endocrinol. Metab. 38 736-741 (1976). 4. F. MORONI D.L. CHENEY, E. PERALTA, E. COSTA, J. Pharmacol. Exp. Ther. ~Q7 870-874 (1978) 5. J. SAWYNOK, F.S. LABELLA, Eur. J. Pharmacol 70 103-110 (1981). 6. H. Katakami, Y. Kato, N. Matsushita, A.Shimatsu, H. Imura, Endocrinology 109 10331036 (1981). 7. P.H. KELLY, K.E. MOORE, Neuropharmacology 17 169-174 (1978). 8. N.E. ANDEN and H. WACHTEL, Acta Pharmacol. Toxicol. (Kbh) 40 310-320 (1977). 9. H.U. FEHR and H.J. BEIN, J. Int. Med. Res.2 36-39 (1974). 10. A.S. ZIGMOND and R.P. SNAITH, Acta Psychiatr. Scand. 67 361-370 (1983). 11. R.L. SPITZER, J. ENDICOTT, E. ROBINS, Arch. Gen. Psychiatry 35 773-782 (1978). 12. M. HAMILTON, J. Neurol. Neurosurg. Psychiatry 23 56-62 (1960). 13. W.D. ODELL, P.L. RAYFORD, G.T. ROSS, J. Lab. Clin. Med. 70 973-980 (1967). 14. M.G. F O R E S T , A.M. CATHIARD, J.A. BERTRAND, J. Clin. Endocrinol. Metab. 36 1132-1142 (1973). 15. R. DE HERTOG, K. THOMAS, Y. BIETLOT I. VANDERHEYDEN, J. FERIN, J. Clin. Endocrinol. Metab. 40 93-101 (1975). 16. A.R. MIDGLEYjr, Endocrinology 79 10-18 (1966). 17. A.R. MIDGLEYjr, J. Clin. Endocrinol. Metab. 27 295-299 (1967). 18. J. TUOMISTO, P. MANNISTO, Pharmacol. Rev. 37 288-293 (1985). 19. P. MONTELEONE, M. MAJ, M. IOVINO L. STEARDO, Acta Endocrinologica 119 353357 (1988). 20. J. TAKAHARA, S. YUNOKI, H. HOSOGI W. YAKUSHIJI, J. KAGEYAMA, T. OFUJI, Endocrinology 106 343-347 (1980). 21. T.D. STRYKER, T. CONLIN, S. REICHLIN, Brain Res. 362 339-343 (1986). 22. R.A. ANDERSON and R. MITCHELL, J. Endocrinol. 108 1-7 (1986). 23. A. DUPONT, L. CUSAN, M. GARON, C.H. LI, Proc. Natl. Acad. Sci. (USA) 74 358359 (1977). 24. C. RIVIER, W. VALE, N. LING, M. BROWN, R. GUILLEMIN, Endocrinology 100 238-241 (1977). 25. Y. KATO, Y. IWASAKI, H. ABE, S. OHGO, H. IMURA, Proc. Soc. Exp. Biol. Med. 158 431-436 (1978). 26. B.VON GRAFFENRIED, E. DEL POZO, J. ROUBICEK, E. KREBS, W. POLDINGER, P. BURMEISTER, L. KERP, Nature 272 729-730 (1978). 27. Y. KATO, H. KATAKAMI, H. 1MURA, In: Growth ~and Growth Factor~. K. Shizume, K. Takano (eds), 159-167, University Tokyo Press, Tokyo, (1980). 28. F. CASANUEVA, J.A. APUD A. LOCATELLI, A. MARTINEZ-CAMPOS, C. CIVATI, G. RACAGNI, D. COCCHI, E.E. MULLER, Endocrinology |09 567-575 (1981). 29. J.R. WALTERS, R.H. ROTH, G.K.AGHAJANIAN, J. Pharmacol. Exp. Ther. 18(; 630639 (1973). 30. K. CHIHARA, A. ARIMURA, Abstract of the 60~h Anmlal Meetin~ of the Endocrin~ Society, 195 (1978). 31. G. RACAGNI, J.A. APUD, D. COCCHI, V. LOCATELLI, E.E. MULLER, Life Sci 31 823-838 (1982). 32. P. CHIODERA, R. VOLPI, V. COIRO, L. BARILLI, G. ROSSI, E. ROTI, J. Endocrinol. Invest. 6 381-384 (1983). 33. R. DINGLEDINE, L.L. IVERSEN, E. BREUKER, Europ. J. Pharmacol. 47 19-27 (1978). 34. M.A. WERZ, K. BAUM, A.B. YOUNG, R.L. MACDONALD, Neurosci. Abstr. 7 478 (1981). 35. E. ARENAS, J. MARSAL, J. ALBERCH, Neurosci. Lett. 120 201-204 (1980). 36. P.G. KRISHNAMURTHY, G.A TEJWANI, H.N. BHARGAVA, Brain Res. 553 284290 (1991).