- Email: [email protected]
Effect of opioid antagonism on β-endorphin processing and proopiomelanocortin-peptide release in the hypothalamus
BRAIN RESEARCH ELSEVIER
Brain Research 648 (1994) 24-31
Research Report
Effect of opioid antagonism on -endorphin processing and proopiomelanocortin-peptide release in the hypothalamus Sharon B. Jaffe, Sylwester Sobieszczyk, Sharon L. Wardlaw * Departments of Medicine and Obstetrics and Gynecology, Columbia Unicersity College of Physicians and Surgeons, New York, NY 10032, USA
Accepted 22 February 1994
Abstract Previous studies have shown that chronic opioid receptor blockade has significant effects on POMC gene expression and peptide levels in the hypothalamus. We have now examined the effects of the opioid antagonist naltrexone on fl-EP processing in the hypothalamus and on the release of 2 POMC-derived peptides,/3-EP and 73-MSH, from the perifused hypothalamus in vitro. The /3-EP immunoactivity in the medial basal hypothalamus (MBH) of 7 rats infused for 1 week with naltrexone by osmotic minipump, was individually analyzed by HPLC and compared to 7 control rats. The mean ratio of /3-EP~_3~ compared to fl-EPl_27 plus fl-EP 1_26 was 2.34 +_0.41 in the naltrexone treated rats, significantly higher than the ratio of 1.26 +_ 0.09 in the control rats (P < 0.02). Thus in the setting of chronic opioid antagonism although /3-EP content decreases, there is relatively more /3-EPl_31, the biologically active opioid form of the peptide, compared to the C-terminally cleaved forms of fl-EP which have reduced biological activity. To study the effects of naltrexone on/3-EP and y3-MSH release, hypothalami were perifused in vitro with 10 6M naltrexone. Basal release of y3-MSH was significantly higher from the naltrexone treated brains compared to the controls (221 _+ 20 pg/60 min vs. 161 _ 6.7 pg/60 min) (P < 0.0l); KCI stimulated y3-MSH was also significantly higher in the naltrexone group (951 _ 94 vs. 543 _+ 85 pg/60 min) (P < 0.005). Basal release of fl-EP was 136 _+ 45 pg/60 min in the naltrexone treated brains compared to 93 _+ 15 pg in the controls, but this difference was not significant; KC1 stimulated release of/3-EP, however was significantly higher in the naltrexone group (558 _+ 103 vs. 275 +_49 pg/60 min) (P < 0.02). To study the acute and chronic effects of naltrexone in vivo on /3-EP and y3-MSH release, rats were either injected with naltrexone and sacrificed 40-60 min later or were infused with naltrexone for 7 days. Baseline Y3-MSH release was significantly higher in rats treated with naltrexone 40-60 min prior to the perifusion (P < 0.01). Baseline /3-EP release was below the limit of assay detection. No differences were noted in the responses of y3-MSH or /3-EP to KCI in either group. In contrast after chronic treatment with naltrexone for 1 week, baseline peptide release was not different from the control animals despite a more than 50% fall in peptide content. The 73-MSH and /3-EP responses to KCI stimulation, however, were significantly less in the naltrexone treated animals. Thus there is an increase in POMC peptide release acutely after treatment with naltrexone in vitro and in vivo. After 1 week of naltrexone, baseline POMC peptide release continues unchanged despite the fall in peptide content, however, the response to KC1 is blunted possibly reflecting the decrease in peptide content after chronic stimulation with naltrexone. We conclude that naltrexone has significant effects on POMC peptide release and on fl-EP processing in the hypothalamus. These results further demonstrate that the brain POMC system can respond to opioid blockade at several levels and are consistent with inhibitory feedback mechanisms for the autoregulation of the POMC system by endogenous/3-EP. Key words." Proopiomelanocortin; fl-Endorphin; y3-Melanocyte-stimulating hormone; Opioid antagonism; Naltrexone; Hypothalamus
1. Introduction
* Corresponding author. Department of Medicine, Columbia University College of Physicians & Surgeons, 630 West 168th Street, New York, NY 10032, USA. Fax: (1) (212) 305-2131. 0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0006-8993(94)00259-F
/ 3 - E n d o r p h i n (/3-EP), o n e of the three distinct classes of e n d o g e n o u s opioids, is synthesized in the arcuate n u c l e u s of the h y p o t h a l a m u s a n d has n u m e r o u s biological effects. In clinical m e d i c i n e fl-EP has b e e n proposed to play a role in a variety of states r a n g i n g from
S.B. Jaffe et aL / Brain Research 648 (1994) 24-31
drug addiction to the neuroendocrine control of pituitary function [1,7,9,27]. Regulation of brain /3-EP and its precursor proopiomelanocortin (POMC) by exogenous opiates has been studied with respect to mechanisms of opiate tolerance and dependence [3,16]. In addition studies with the short-acting opioid antagonist naloxone have demonstrated on a short-term basis that there is feedback regulation of /3-EP by endogenous opioids [18]. Little is known, however, about how brain /3-EP and its precursor POMC adapt to chronic opioid blockade. This is of importance since the long-acting opiate receptor antagonist naltrexone is being used on a chronic basis to treat opiate addiction [5]. In addition chronic treatment with naltrexone has been suggested for a variety of pathological conditions associated with high endogenous opioid levels [2,25]. We have previously shown in the rat that chronic treatment with naltrexone had significant effects on POMC gene expression and peptide levels in the hypothalamus [12,20]. After 1-4 weeks of treatment there was a significant decline in the concentration of/3-EP in the hypothalamus; a parallel decline in the concentrations of several other POMC-derived peptides, 3'3MSH, a-MSH and CLIP, was also noted. After 3 weeks of naltrexone administration there was also a significant increase in the concentration of POMC mRNA in the medial basal hypothalamus (MBH). This increase in POMC gene expression associated with a fall in peptide levels was felt to be consistent with a compensatory increase in brain fl-EP synthesis and release in the setting of chronic opioid receptor blockade. This study, however, did not measure POMC peptide release and did not characterize the immunoreactive forms of /3-EP to determine if there were specific effects of naltrexone on /3-EP processing. In the hypothalamus /3-EP~_3~ can be further processed to 13-EP1_27 and fl-EPl_26. The C-terminally cleaved forms of/3-EP have reduced biological activity and may also function as /3-EP antagonists [17]. In the present study we have therefore examined the effects of treatment with naltrexone on /3-EP processing and POMC peptide release in the hypothalamus of the rat.
2. Materials and methods 2.1. Animals and treatment schedules Adult male Sprague-Dawley rats, weighing 200-225 g, were obtained from Charles River (Wilmington, MA) and used in all studies. Naltrexone HC1 was kindly donated by Du Pont Pharmaceuticals (Wilmington, DE). Methohexital sodium was used as the anesthetic for the subcutaneous placement of osmotic minipumps (Alza, Palo Alto, CA) for naltrexone infusion. Two sets of studies were performed. The first study examined the effect of naltrexone on /3-EP processing in vivo in the hypothalamus. In the second series of studies a hypothalamic perifusion system was set up and validated in
25
order to examine the acute and chronic effects of naltrexone on /3-EP and y3-MSH release from the hypothalamus in vitro.
2.1.1. Effect of naltrexone on [3-EP processing Seven rats were infused with naltrexone, 2.4 mg/day, via subcutaneously implanted osmotic minipumps for a total of 7 days. Naltrexone was dissolved in 8.5% lactic acid for infusion. Seven rats with empty Silastic implants served as controls. After 7 days the hypothalamus of each rat was individually extracted and the /3-EP immunoactivity analyzed by HPLC and RIA as described below.
2.1.2. Effects of naltrexone on fl-EP and T3-MSH release (a) In an initial experiment hypothalami from 12 rats were sliced with a Mcllwain mechanical tissue chopper as described below and perifused (2 hypothalami per chamber including the MBH and AH) using the Acusyst-S perifusion system to determine basal fl-EP and y3-MSH release and response to depolarization with 56 mM KC1 at the beginning and end of a 5 hr perifusion period. (b) To study the effect of naltrexone on fl-EP and y3-MSH release in vitro 12 chambers containing 3MBH each were perifused with control media and 11 chambers with 3MBH each were perifused with 10 6M naltrexone dissolved in control medium. Hypothalami were equilibrated and perifused as described below. Peptide release was measured at 20 min intervals over a 60 min period prior to depolarization with 56 mM KCI and for a 60 rain period following KCI. (c) To study the acute and chronic effects of naltrexone injection on /3-EP and y3-MSH release in vitro, rats were treated in vivo with naltrexone and sacrificed within 40-60 min after one injection or after 7 days of chronic treatment. For the acute study 24 rats were injected subcutaneously with naltrexone (4 mg/kg) and were sacrificed 40-60 min later, 22 rats injected with saline served as controls. To study the chronic effects of naltrexone on fl-EP and y3-MSH release, 12 rats were infused with naltrexone 2.4 rag/day by subcutaneously implanted osmotic minipump for 7 days; 12 rats with empty silastic implants served as controls. Animals were sacrificed and the hypothalami (2MBH per chamber) were perifused as described below.
2.2. Tissue dissection and extraction of peptides Rats were sacrificed by decapitation between 10.00 and 11.00 h and the brain was quickly removed and placed in ice cold Gibco 199 with Hank's salts, and L-glutamine. The brain was then placed in a lucite brain holder, ventral side up, with razor blades positioned at 2-mm intervals for cutting sections. The medial basal hypothalamus (MBH) was dissected from a 2-mm coronal section cut immediately caudal to the optic chiasm. The dissection was limited laterally by the hypothalamic sulci and dorsally by the mammillothalamic tract. In one experiment the anterior hypothalamus (AH) was dissected from the next rostral 2-mm coronal section and was bordered dorsally by the anterior commissure. For HPLC analysis, each MBH was homogenized immediately in 1 ml of 0.2 N HC1 and processed as described below. At the end of the perifusion period in some experiments, the hypothalami were removed from the perifusion chambers and homogenized in 2 ml of 0.2N HCI. The 4000 × g supernatant was used for peptide RIA.
2.3. In vitro perfusion of rat hypothalami To measure in vitro release of POMC-derived peptides hypothalami were perifused, essentially as described by Nikolarakis et al., [18] using the Acusyst-S perifusion system (Endotronics, Coon Rapids, MN). Hypothalami were sliced with a Mcllwain mechanical tissue chopper in the sagittal plane at 250/xm intervals. Two or three
26
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31
hypothalami were placed in each perifusate chamber which was preincubated at 37 C with preoxygenated Gibco 199 with Earle's salts, L-glutamine and NaHCO 3 2200 m g / l supplemented with an additional 0.8 g / l glucose and 1 g / l BSA. The system was continuously gassed with O2-CO 2 (95:5). A multiple channel peristaltic pump delivered the media to the tissues at a defined rate. Initially, the chambers were perifused at 1 m l / m i n x 20-30 min followed by a 20-30 min period at 0.1 m l / m i n to achieve a stable rate of peptide release. The perifusion continued at 0.1 m l / m i n during sample collection. Samples were collected every 20 min into tubes containing 0.4 ml of 0.2N HCI for acidification and inactivation of proteases. Samples were frozen and subsequently assayed for/3-EP and y3-MSH by RIA. In an initial study release of /3-EP and y3-MSH from the MBH and AH was determined over a 5 hr period and the response to 56 mM K C I x 3 0 rain was studied at the beginning and end of this perifusion period. Subsequently all experiments were performed using two or three MBH per chamber and peptide release was measured over a 60 rain period before depolarization with 56 mM KCI and for a 60 min period after KCI.
2.4. Radioimmunoassays The fl-EP concentrations in the MBH extracts and in the acidified perifusates were assayed with an antiserum to human /3-EP raised in this laboratory which crossreacts equally with camel /3-EP, N-acetyl /3-EP and fl-EP I 27 and 30% on a molar basis with human /3-1ipotropin [24]. Synthetic camel /3-EP (Peninsula Laboratories, Belmont, CA) was used as the standard. For the/3-EP measurements shown in Fig. 4, samples were evaporated in a Speed Vac Concentrator prior to RIA to increase assay sensitivity. The effective level of detection for fl-EP per 20 min fraction was 12 pg. We have previously shown that the /3-EP immunoactivity in hypothalamic homogenates measured with this RIA elutes upon gel filtration in the same position as synthetic fl-EP standard [24]. Further characterization by ion exchange chromatography or HPLC revealed that the fl-EP immunoactivity in the hypothalamus consists primarily of flEPt_31 and to a lesser extent fl-EPl_27 and fl-EPI_26 [23,24]. The 73-MSH concentrations in the MBH extracts and in the perifusates were measured with an antiserum provided by Drs. Meador-Woodruff and Huda Akil (University of Michigan) directed against the midportion of rat Y3-MSH; rat y3-MSH synthesized by Dr. Larry Taylor from the same group was used as the standard [14]. The y3-MSH immunoactivity in the hypothalamic homogenates and perifusates was characterized by gel filtration as described below.
Cls reverse phase column (Vydac, Hesperia, CA) with the Waters 510 and 590 solvent delivery system, the 680 automated gradient controller and the U6K Universal Injector. Samples were eluted with a gradient of 80% acetronitrile containing 0.1% TFA (solvent B) against 0.1% TFA in water (solvent A) at a flow rate of 1 ml/min. The concentration of solvent B rose in a linear fashion from 30 to 35% over the first 5 min and then from 35 to 45% over the next 45 rain and from 45 to 50% over the next 5 min. The column was then washed with 100% solvent B for 10 min and returned to the starting condition of 30% B. 0.5 ml fractions were collected, evaporated in a Speed Vac Concentrator and dissolved in buffer for/3-EP RIA. The column was calibrated each day with 5 ng each of synthetic camel 13-EPI_31, fl-EPl_27 and fl-EPl_26.
3. Results
3.1. Effect o f naltrexone on f l - E P processing When
the /3-EP immunoactivity
noted HPLC
compared
to 7 control
profile of a naltrexone
of 7
rats. An example treated
of the
rat compared
to
a c o n t r o l r a t is s h o w n in F i g . 1. I n t h e c o n t r o l a n i m a l ~
1-31
1-27
1-26
1~0
8OO
Control 6OO
& c~.
2OO
0
i
40
2.5. Gel filtration Perifusate samples from several control perifusions were pooled, dried in a Speed Vac Concentrator and then resuspended in 0.05 N HCI containing 0.1% BSA. For comparison the extract from a control hypothalamic homogenate was also used for gel filtration. Samples were chromatographed on a Sephadex-G-50 (superfine) column (0.9x50 cm) in 0.05 N HCI and 0.1% BSA with a flow rate of 0.2 ml/min. Fractions (1 ml) were collected and assayed in duplicate by RIA. Both y3-MSH and fl-EP were measured after gel filtration of the hypothalamic extract but the entire sample was used for y3-MSH RIA after gel filtration of the perifusates. The column was calibrated with blue dextran, fl-EP, and fl-LPH.
in the MBH
rats, infused for 1 week with naltrexone, was individually a n a l y z e d b y H P L C , s i g n i f i c a n t d i f f e r e n c e s w e r e
50
60
70
i 8O
8OO
o
Naltrexone
600
"~ ~oo & 2OO
0
2.6. High-performance liquid chromatography
40
50'
60'
? '0
80
traction
Individual aliquots (0.5 ml) of the 4000x g supernatant from each MBH were evaporated in a Speed Vac Concentrator and dissolved in 0.2 ml of 0.1% trifluoroacetic acid (TFA) containing 24% acetronitrile and subjected to high-performance liquid chromatography (HPLC). HPLC was carried out on a 5-mm, 300 A pore,
Fig. l. Reverse phase HPLC elution profiles of the/3-EP immunoactivity in extracts of the MBH obtained from a control rat (upper panel) and from a rat treated with naltrexone for 7 days (lower panel). Arrows indicate the elution positions of the synthetic peptides fl~-EPl_31, flc-EPi 27 and flc-EPi 26.
27
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31
although much of the fl-EP-immunoactivity eluted in the position of fl-EPt_3~, a considerable portion also eluted in the positions of fl-EPl_27 and fl-EPl_26. In contrast after naltrexone almost all the immunoactivity could be accounted for by fl-EPI_ar The mean ratio of fl-EPt_3] compared to fl-EPl_27 plus fl-EPt_26 was 2.34 _+0.41 in the naltrexone treated rats, significantly higher than the ratio of 1.26 _+0.09 in the control rats (P < 0.02). The mean ratio of ~-EPI_31 to /3-EP,_27 was 3.03 + 0.46 in the naltrexone treated animals; this was also significantly higher than the ratio of 1.95 + 0.13 in the control animals (P < 0.05).
i
1000
KCl
KCI i P,
!
i
..~ 800
E 600
400 m
200
3.2. [3-EPand y3-MSH release during hypothalamic perifusion
IL
0
i
i
6O
120
i
160
i
240
i
300
300
When 6 chambers were perifused containing slices of AH and MBH from 2 rats each, there was stable peptide release over a 5 h period. In addition there was a good response to depolarization with KC1 at the beginning and at the end of the 5 h period (Fig. 2). Subsequently only MBH slices were used as peptide release appeared to be more consistent between chambers for this dissection.
250 /
200
~I
E
150
"'b
j ~
100 e~
3.3. Characterization of peptide immunoactivity in the perifusates compared to hypothalamic homogenates
50 0
The y3-MSH immunoactivity in the hypothalamic perifusates eluted upon gel filtration in 2 distinct peaks (Fig. 3). The second peak eluted in the same position as the fl-EP standard which corresponds to what has been reported for glycosylated 73-MSH; [6] the larger molecular weight peak eluted just after the /3-LPH standard. A similar pattern of 73-MSH immunoactivity was seen when the hypothalamic homogenate was chromatographed. This corresponds with a previous report demonstrating 2 peaks of y3-MSH related immunoactivity in the rat hypothalamus with the first peak corresponding to a molecule of an apparent molecular weight of 7100 [6]. As reported previously all of the/3-EP immunoactivity in the homogenate eluted in the position of the synthetic /3-EP standard [24].
3.4. Effect of perifusion with naltrexone in uitro on y3-MSH and [3-EP release Baseline and KC1 stimulated released of 73-MSH was significantly higher from hypothalami perifused with naltrexone compared to the control hypothalami. Baseline y3-MSH release was 161 _+ 6.7 pg/60 min in the control group compared to 221 _+20 pg/60 min in the naltrexone group (P < 0.01). Following stimulation with KCI, y3-MSH release increased to 543 _+85 pg/60 min in the control group compared to 950 _+94 pg/60 min in the naltrexone group (P < 0.005) (Fig. 4). The
6O
i
i
i
i
120
180
240
300
time (mln) Fig. 2. Individual in vitro release profiles of y3-MSH (upper panel) and/3-EP (lower panel) when 3 chambers were perifused containing hypothalamic slices (MBH plus AH) from 2 rats each. Fractions were collected at 15 rain intervals over a 5 h period. Response to depolarization with KCI was determined after 1 h and after 4 h of perifusion.
baseline release of fl-EP from the naltrexone group of 136 _+45 pg/60 min was not significantly higher than the baseline release from the control group of 93 + 15 pg/60 min. However following KCI stimulation, fl-EP release from the naltrexone group was 558 _+ 103 pg/60 min, significantly higher than the release of 275 _+49 pg/60 min from the control group (P < 0.02) (Fig. 4).
3.5. Acute and chronic effects of naltrexone treatment in vivo on 73-MSH and fl-EP release Baseline T3-MSH release from the perifused hypothalamus was significantly higher in rats treated in vivo with naltrexone 40-60 min prior to the perifusion. Mean y3-MSH release was 194 + 29 pg/60 min in the saline-treated animals compared to 367 + 47 pg/60 min in the naltrexone-treated animals (P < 0.01) (Fig. 5). Response to KC1 stimulation was similar in both groups 651 + 87 vs. 656 _+84 pg/60 min. Baseline/3-EP
28
S.B. Jaffe et al. / Brain Research 648 (1994) 24-31 Vo
LPH
[~-EP
41-
41.
41,
I000
Vt
800
600
& 600
5OO 400
o °~
"~
[]
saline
•
naltrexone
400
~
200
;~"
<60 0 baseline stimulated baseline s t i m u l a t e d
2OO
Acute
Chronic
250
100
w r 0
2
•
I
I
30
40
•
I 50
200 ~D
fraction
150
Fig. 3. Sephadex G-50 chromatography of the -x3-MSH immunoactivity in pooled hypothalamic perifusates. The arrows indicate the void volume, the elution positions of /3-LPH and /3-EP and the total volume.
100 ~..
<60
CQ.
baseline
.E
1000 "
E
800"
[] Control • Nattrexone
o~ 4oo p--
stimulated b a s e l i n e
Acute
stimulated
Chronic
Fig. 5. Mean release per hour of y3-MSH_+ SEM (upper panel) and /3-EP +S.E.M. (lower panel) during hypothalamic perifusion (2 MBH per chamber) after either acute or chronic treatment, with naltrexone in vivo as compared to saline treatment. Samples were collected at 20 min intervals. Peptide release is depicted over a 60 rain period before and after stimulation with KCI. The dashed line indicates the cumulative level of peptide detection over a 60 min period. * P < 0.01 compared to saline treated animals.
200
Baseline
Stimulated
700' "~
600 ¸
-~
500-
T
400-
300"
200100
Baseline
Stimulated
Fig. 4. Mean release per hour of y3-MSH _+S.E.M. (upper panel) and /3-EP + S.E.M. (lower panel) during hypothalamic perifusion (3 MBH per chamber) with either 1 0 - 6 M naltrexone or control media. Samples were collected at 20 rain intervals. Peptide release is depicted over a 60 rain period before and after stimulation with KCI. * P < 0.02 vs. control.
release was below the limit of assay detection; no difference was noted in the response to KC1 in either group. In contrast after chronic treatment with naltrexone for 1 week baseline 73-MSH release was not different from the control animals; basal /3-EP release was again below the limit of assay detection. The -,/3-MSH and /3-EP responses to KC! stimulation, however, were significantly less in the naltrexone treated rats (Fig. 5). Mean ,/3-MSH release after KC1 was 860 + 117 p g / 6 0 min in the control rats compared to 234 + 38 p g / 6 0 min in the naltrexone-treated rats ( P < 0.001). Mean ~ - E P release after KCI was 170 + 45 p g / 6 0 min in the controls compared to 74.2 + 6.7 p g / 6 0 min after naltrexone ( P < 0.01). At the end of the perifusion the hypothalami from each chamber were homogenized and assayed for 73-MSH and /3-EP peptide content. As expected there was a significant decrease in peptide content in the naltrexone group. 73-MSH was reduced from 24,950 + 690 to 11,700 + 400 p g / c h a m b e r and /3-EP was reduced from 14,030 + 600 to 6420 +_ 190 p g / c h a m b e r ( P < 0.001). Thus there is
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31
an increase in POMC peptide release shortly after an initial dose of naltrexone. After 1 week of naltrexone, when the hypothalamic content of several POMC-derived peptides is decreased, baseline POMC-peptide release is unchanged from control animals, however, the response to KC1 is blunted.
4. Discussion
Evidence accumulated from several studies using opioid receptor antagonists supports the concept that there is feedback regulation of POMC by endogenous opioids. We have previously reported that treatment with naltrexone for 1 month caused a fall in the concentration of fl-EP in the hypothalamus where POMC is synthesized as well as in several other brain regions where POMC neurons project [20]. The fall in the concentration of /3-EP was paralleled by a fall in the concentrations of a-MSH, CLIP and y3-MSH, the other major POMC-derived peptide products in the hypothalamus [12]. A decline in peptide content was noted as early as 2 days of treatment. Naltrexone was also noted to have a significant effect on POMC gene expression in the MBH [4,12]. After 3 weeks of naltrexone, the concentration of POMC mRNA in the MBH was higher in naltrexone treated animals compared to the controls. This increase in the concentration of POMC mRNA was seen in both intact and in castrated animals and was thus unrelated to any potential effect naltrexone may have had on LH and testosterone levels. Although the increase in POMC gene expression associated with a fall in peptide levels was interpreted as being consistent with a compensatory increase in brain /3-EP synthesis and release in the setting of chronic opioid receptor blockade, it was unknown if in fact POMC peptide release was altered. In addition it was unclear if the immunoactive form of fl-EP that was being measured was in fact the biologically active opioid peptide ]3-EPl_3V /3-EP which results from the posttranslational processing of POMC can be further processed through N-terminal acetylation or C-terminal cleavage to yield fl-EPl_26 and fl-EPl_27 [6,21,29]. These modifications greatly reduce or abolish opioid activity. N-acetyl /3-EP which is inactive in opiate receptor binding assays, is not a major form of/3-EP in the hypothalamus. In contrast, fl-EPl_26 and 1-27 which are present in the hypothalamus, have opioid receptor binding activity but have little analgesic activity. In addition fl-EPl_27 has been shown to antagonize the analgesic effects of/3-EP1_31 [17]. We therefore determined the effect of naltrexone on the relative amounts of the major forms of /3-EP in the hypothalamus: ~-EPI_31 , fl-EPl_27 and fl-EPL_26. We have shown that in the setting of chronic opioid antagonism although all 3 forms of /3-EP decrease, there is relatively more
29
/~-EPI_31, the form of the peptide with opioid bioactivity, compared to fl-EPl_27 ' which has opioid antagonist activity, and /3-EP1_26. This is in agreement with the results of Bronstein and colleagues who have shown by gel filtration of pooled hypothalamic samples that naltrexone treatment reduces the peak coeluting with fl-EPl_27 and fl-EPI_26 relative to the /3-EPl_31 peak [4]. This change in peptide ratios could result from a specific effect on processing enzymes responsible for the C-terminal proteolysis of /3-EP or could be a reflection of increased/~-EP turnover and release with less time available for further processing. Additionally there may be relative changes in peptide degradation. Dopaminergic agents have been shown to influence /3-EP processing in the neurointermediate lobe of the pituitary [8,15]. For example in addition to causing changes in /3-EP synthesis and release dopamine antagonism has been shown to affect a-N-acetyltransferase activity in the intermediate lobe [15]. Changes in /3-EP processing have also been reported in the hypothalamus associated with sexual maturation and aging [13,26]. There is also some data to suggest that in morphine treated rats there is an increase in the opiate inactive forms of /3-EP in the hypothalamus [3]. We now show in the setting of chronic opioid blockade that there is a relative decrease in the same opiate inactive forms of/3-EP in the hypothalamus. In the present study, when naltrexone was administered either in vitro or in vivo, significant effects on POMC peptide release were noted from the perifused hypothalamus. In a previous study, Nikolarakis et al. have shown that naloxone (10-6M) produced a calcium-dependent increase in the non-stimulated release of fl-EP in the perifused hypothalamus in vitro [18]. In our study 10-6M naltrexone in vitro increased basal T3-MSH release and KCI stimulated y3-MSH and/3-EP release. There was an increase in the non-stimulated release of y3-MSH from the hypothalamus of animals treated with naltrexone 40-60 min prior to sacrifice. Since non-stimulated /3-EP release was below the level of assay detection in these studies no statement can be made about /3-EP release but in general the pattern of release of /3-EP and y3-MSH have paralleled each other in vitro. After treatment with naltrexone for 1 week baseline y3-MSH release, which had been elevated acutely, was not significantly different from control. KCl-stimulated y3-MSH, however, was significantly less than the controls; a similar effect was seen with /3-EP release. These responses occurred at a time when the hypothalamic concentrations of y3-MSH and /3-EP were shown to be decreased by 53% and 54% respectively. Non-stimulated peptide release however appeared to be normal despite the low peptide content. In fact, if normalized to the decreased MBH peptide content, basal y3-MSH release could be interpreted as relatively increased after chronic naltrexone
30
S.B. Jaffe et al. / Brain Research 648 (1994) 24-31
treatment. These findings are consistent with a decrease in peptide content associated with increased peptide turnover after chronic stimulation with naltrexone. Opioid agonists and antagonists have also been shown to have electrophysiological effects on POMC neurons and to affect opioid receptor binding. When intracellular recordings were made from arcuate neurons in the guinea pig, mu opioid agonists were found to induce membrane hyperpolarization and to decrease the spontaneous firing of POMC neurons [10]. This is consistent with the hypothesis that opioid receptors function as autoreceptors which respond to endogenous/3-EP release. Chronic exposure to opioid antagonists including naltrexone has also been shown to increase opioid receptor binding sites in brain [11,19,22, 28]. Opioid receptor upregulation is seen after 8 or more days of treatment. An increase in the analgesic potency of opiates was also noted after a period of pretreatment with naltrexone [22,28]. Functional supersensitivity therefore parallels the increase in receptor binding after chronic exposure to naltrexone. Thus the changes in POMC peptide release and gene expression that are measured after opioid receptor blockade [12,18] are accompanied by an enhanced ability to bind endogenous /3-EP. In summary the current studies demonstrate acute and chronic effects of naltrexone on POMC peptide release from the hypothalamus in vitro. There is also a change in peptide processing as shown by a shift in the forms of/3-EP found in the hypothalamus resulting in a relative increase in the biologically active opioid form of/3-EP. These results are consistent with an inhibitory feedback mechanism for the autoregulation of POMC by /3-EP. When taken together with previous studies our results provide further support for a compensatory increase in POMC peptide release and synthesis in the setting of opioid receptor blockade.
Acknowledgements The technical assistance of Ms. Marion Richman and Mr. Samuel Yum is greatly appreciated. This work has been supported by NIDA Grant DA 07732.
References [1] Akil, H., Watson, S.J., Young, E, Lewis, M.E., Khachaturian, H. and Walker, J.M., Endogenous opioids: biology and function, Annu. Ret:. Neurosci., 7 (1984) 223-255. [2] Atkinson, R.L., Endocrine and metabolic effects of opiate antagonists, Z Clin. Psychiatry, 45 (1984) 20 24. [3] Bronstein, D.M., Przewlocki, R. and Akil, H., Effects of morphine treatment on pro-opiomelanocortin systems in rat brain, Brain Res., 519 (1990) 102-11.
[4] Bronstein, D.M., Pre- and posttranslational regulation of/3-endorphin biosynthesis in the CNS: effects of chronic naltrexone treatment, J. Neurochem., 60 (1993) 40-49. [5] Crabtree, B.L., Review of naltrexone, a long-acting opiate antagonist, Clin. Pharm., 3 (1984) 273-280. [6] Emeson, R.B. and Eipper, B.A., Characterization of proACTH/endorphin-derived peptides in rat hypothalamus, J. Neurosci., 6 (1986) 837-849. [7] Ferin, M., Van Vugt, D. and Wardlaw, S., The hypothalamic control of the menstrual cycle and the role of endogenous opioid peptides, Recent Prog. Horm. Res., 40 (1984) 441-485. [8] Ham, J. and Smyth, D.G., Regulation of bioactive /3-endorphin processing in rat pars intermedia, FEBS Lett., 175 (1984) 407411. [9] Howlett, T.A. and Rees, L.H., Endogenous opioid peptides and hypothalamo-pituitary function, Annu. Reu. Physiol., 48 (1986) 527-536. [10] Kelly, M.J., Loose, M.D. and Ronnekleiv, O.K., Opioids hyperpolarize /3-endorphin neurons via 1-receptor activation of a potassium conductance, Neuroendocrinology, 52 (1990) 268-275. [11] Lahti, R.A. and Collins, R.J., Chronic naloxone results in prohinged increases in opiate binding sites in brain, Eur. J. Pharmacol., 51 (1978) 185-186. [12] Markowitz, C.E., Berkowitz, K.M., Jaffe, S.B. and Wardlaw, S.L., Effect of opioid receptor antagonism on proopiomelanocortin peptide levels and gene expression in the hypothalamus, Mol. Cell. Neurosci., 3 (1992) 184 190. [13] Martensz, N.D., Changes in the processing of /3-endorphin in the hypothalamus and pituitary gland of female rats during sexual maturation, Neuroscience, 16 (1985) 625-640. [14] Meador-Woodruff, J.H., Pellerito, B., Bronstein, D., Lin, H-L., Ling, N. and Akil, H., Differential effects of haloperidol on the rat pituitary: decreased biosynthesis, processing and release of anterior lobe pro-opiomelanocortin, Neuroendocrinology, 51 (199(/) 294-303. [15] Millington, W.R., O'Donohue, T.L., Chappell, M.C., Roberts, J.L. and Mueller, G.P., Coordinate regulation of petide acetyltransferase activity and proopiomelanocortin gene expression in the intermediate lobe of the rat pituitary, Endocrinology, 118 (1986) 2024-2033. [16] Mocchetti, 1, Ritter, A. and Costa, E., Down-regulation of proopiomelanocortin synthesis and beta-endorphin utilization in hypothalamus of morphine-tolerant rats, J. Mol. Neurosci., 1 (1989) 33 38. [17] Nicolas, P. and Li, C.tt., /3-Endorphin-(l-27) is a naturally occurring antagonist to etorphine-induced analgesia, Proc. NatZ Acad. Sci. USA, 82 (1985) 3178-3181. [18] Nikolarakis, K.E., Almeida O.F.X. and Herz, A., Feedback inhibition of opioid peptide release in the hypothalamus of the rat, Neuroscience, 23 (1987) 143-148. [19] Pert, C., B. and Snyder, S.H., Opiate receptor binding enhancement by opiate administration in vivo, Biochem. Pharmacol., 25 (1975) 847-853. [20] Ragavan, V.V., Wardlaw, S.L., Kreek, M.J. and Frantz, A.G., Effect of chronic naltrexone and methadone administration on brain immunoactive ,6- endorphin in the rat, Neuroendocrinology, 37 (1983) 266-268. [21] Smith, A.I. and Funder, J.W., Proopiomelanocortin processing in the pituitary, central nervous system, and peripheral tissues, Endocr. Rel~., 9 (1988) 159-179. [22] Tempel, A., Gardner, E.L. and Zukin, R.S., Neurochemical and functional correlates of naltrexone-induced opiate receptor upregulation, J. Pharmacol. Exp. Ther., 232 (1984) 439-444. [23] Wardlaw, S.L., Effect of androgens on hypothalamic proopiomelanocortin, Neuroendocrinology, 47 (1988) 164-168. [24] Wardlaw, S.L., Regulation of /3-endorphin, corticotropin-like intermediate lobe peptide, and o:-melanotropin-stimulating hot-
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31 mone in the hypothalamus by testosterone, Endocrinology, 119 (1986) 19-24. [25] Wildt, L. and Leyendecker, G., Induction of ovulation by the chronic administration of naltrexone in hypothalamic amenorrhea, J. Clin. Endocrinol. Metab., 64 (1987) 1334-1335. [26] Wilkinson, C.W. and Dorsa, D.M., The effects of aging on molecular forms of beta- and gamma-endorphins in rat hypothalamus, Neuroendocrinology, 43 (1986) 124-131. [27] Yen, S.S.C., Quigley, M.E., Reid, R.L., Ropert, J.F. and Cetel, N.S., Neuroendocrinology of opioid peptides and their role in
31
the control of gonadotropin and prolactin secretion, Am. J. Obstet. Gynecol., 152 (1985) 485-493. [28] Yoburn, B.C., Nunes, F.A., Adler, B., Pasternak, G.W. and Inturrisi, C.E., Pharmacodynamic supersensitivity and opioid receptor upregulation in the mouse, J. Pharmacol. Exp. Ther., 239 (1986) 132-135. [29] Zakarian, S. and Smyth, D.G., fl-Endorphin is processed differently in specific regions of rat pituitary and brain, Nature, 296 (1982) 250-252.
Brain Research 648 (1994) 24-31
Research Report
Effect of opioid antagonism on -endorphin processing and proopiomelanocortin-peptide release in the hypothalamus Sharon B. Jaffe, Sylwester Sobieszczyk, Sharon L. Wardlaw * Departments of Medicine and Obstetrics and Gynecology, Columbia Unicersity College of Physicians and Surgeons, New York, NY 10032, USA
Accepted 22 February 1994
Abstract Previous studies have shown that chronic opioid receptor blockade has significant effects on POMC gene expression and peptide levels in the hypothalamus. We have now examined the effects of the opioid antagonist naltrexone on fl-EP processing in the hypothalamus and on the release of 2 POMC-derived peptides,/3-EP and 73-MSH, from the perifused hypothalamus in vitro. The /3-EP immunoactivity in the medial basal hypothalamus (MBH) of 7 rats infused for 1 week with naltrexone by osmotic minipump, was individually analyzed by HPLC and compared to 7 control rats. The mean ratio of /3-EP~_3~ compared to fl-EPl_27 plus fl-EP 1_26 was 2.34 +_0.41 in the naltrexone treated rats, significantly higher than the ratio of 1.26 +_ 0.09 in the control rats (P < 0.02). Thus in the setting of chronic opioid antagonism although /3-EP content decreases, there is relatively more /3-EPl_31, the biologically active opioid form of the peptide, compared to the C-terminally cleaved forms of fl-EP which have reduced biological activity. To study the effects of naltrexone on/3-EP and y3-MSH release, hypothalami were perifused in vitro with 10 6M naltrexone. Basal release of y3-MSH was significantly higher from the naltrexone treated brains compared to the controls (221 _+ 20 pg/60 min vs. 161 _ 6.7 pg/60 min) (P < 0.0l); KCI stimulated y3-MSH was also significantly higher in the naltrexone group (951 _ 94 vs. 543 _+ 85 pg/60 min) (P < 0.005). Basal release of fl-EP was 136 _+ 45 pg/60 min in the naltrexone treated brains compared to 93 _+ 15 pg in the controls, but this difference was not significant; KC1 stimulated release of/3-EP, however was significantly higher in the naltrexone group (558 _+ 103 vs. 275 +_49 pg/60 min) (P < 0.02). To study the acute and chronic effects of naltrexone in vivo on /3-EP and y3-MSH release, rats were either injected with naltrexone and sacrificed 40-60 min later or were infused with naltrexone for 7 days. Baseline Y3-MSH release was significantly higher in rats treated with naltrexone 40-60 min prior to the perifusion (P < 0.01). Baseline /3-EP release was below the limit of assay detection. No differences were noted in the responses of y3-MSH or /3-EP to KCI in either group. In contrast after chronic treatment with naltrexone for 1 week, baseline peptide release was not different from the control animals despite a more than 50% fall in peptide content. The 73-MSH and /3-EP responses to KCI stimulation, however, were significantly less in the naltrexone treated animals. Thus there is an increase in POMC peptide release acutely after treatment with naltrexone in vitro and in vivo. After 1 week of naltrexone, baseline POMC peptide release continues unchanged despite the fall in peptide content, however, the response to KC1 is blunted possibly reflecting the decrease in peptide content after chronic stimulation with naltrexone. We conclude that naltrexone has significant effects on POMC peptide release and on fl-EP processing in the hypothalamus. These results further demonstrate that the brain POMC system can respond to opioid blockade at several levels and are consistent with inhibitory feedback mechanisms for the autoregulation of the POMC system by endogenous/3-EP. Key words." Proopiomelanocortin; fl-Endorphin; y3-Melanocyte-stimulating hormone; Opioid antagonism; Naltrexone; Hypothalamus
1. Introduction
* Corresponding author. Department of Medicine, Columbia University College of Physicians & Surgeons, 630 West 168th Street, New York, NY 10032, USA. Fax: (1) (212) 305-2131. 0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0006-8993(94)00259-F
/ 3 - E n d o r p h i n (/3-EP), o n e of the three distinct classes of e n d o g e n o u s opioids, is synthesized in the arcuate n u c l e u s of the h y p o t h a l a m u s a n d has n u m e r o u s biological effects. In clinical m e d i c i n e fl-EP has b e e n proposed to play a role in a variety of states r a n g i n g from
S.B. Jaffe et aL / Brain Research 648 (1994) 24-31
drug addiction to the neuroendocrine control of pituitary function [1,7,9,27]. Regulation of brain /3-EP and its precursor proopiomelanocortin (POMC) by exogenous opiates has been studied with respect to mechanisms of opiate tolerance and dependence [3,16]. In addition studies with the short-acting opioid antagonist naloxone have demonstrated on a short-term basis that there is feedback regulation of /3-EP by endogenous opioids [18]. Little is known, however, about how brain /3-EP and its precursor POMC adapt to chronic opioid blockade. This is of importance since the long-acting opiate receptor antagonist naltrexone is being used on a chronic basis to treat opiate addiction [5]. In addition chronic treatment with naltrexone has been suggested for a variety of pathological conditions associated with high endogenous opioid levels [2,25]. We have previously shown in the rat that chronic treatment with naltrexone had significant effects on POMC gene expression and peptide levels in the hypothalamus [12,20]. After 1-4 weeks of treatment there was a significant decline in the concentration of/3-EP in the hypothalamus; a parallel decline in the concentrations of several other POMC-derived peptides, 3'3MSH, a-MSH and CLIP, was also noted. After 3 weeks of naltrexone administration there was also a significant increase in the concentration of POMC mRNA in the medial basal hypothalamus (MBH). This increase in POMC gene expression associated with a fall in peptide levels was felt to be consistent with a compensatory increase in brain fl-EP synthesis and release in the setting of chronic opioid receptor blockade. This study, however, did not measure POMC peptide release and did not characterize the immunoreactive forms of /3-EP to determine if there were specific effects of naltrexone on /3-EP processing. In the hypothalamus /3-EP~_3~ can be further processed to 13-EP1_27 and fl-EPl_26. The C-terminally cleaved forms of/3-EP have reduced biological activity and may also function as /3-EP antagonists [17]. In the present study we have therefore examined the effects of treatment with naltrexone on /3-EP processing and POMC peptide release in the hypothalamus of the rat.
2. Materials and methods 2.1. Animals and treatment schedules Adult male Sprague-Dawley rats, weighing 200-225 g, were obtained from Charles River (Wilmington, MA) and used in all studies. Naltrexone HC1 was kindly donated by Du Pont Pharmaceuticals (Wilmington, DE). Methohexital sodium was used as the anesthetic for the subcutaneous placement of osmotic minipumps (Alza, Palo Alto, CA) for naltrexone infusion. Two sets of studies were performed. The first study examined the effect of naltrexone on /3-EP processing in vivo in the hypothalamus. In the second series of studies a hypothalamic perifusion system was set up and validated in
25
order to examine the acute and chronic effects of naltrexone on /3-EP and y3-MSH release from the hypothalamus in vitro.
2.1.1. Effect of naltrexone on [3-EP processing Seven rats were infused with naltrexone, 2.4 mg/day, via subcutaneously implanted osmotic minipumps for a total of 7 days. Naltrexone was dissolved in 8.5% lactic acid for infusion. Seven rats with empty Silastic implants served as controls. After 7 days the hypothalamus of each rat was individually extracted and the /3-EP immunoactivity analyzed by HPLC and RIA as described below.
2.1.2. Effects of naltrexone on fl-EP and T3-MSH release (a) In an initial experiment hypothalami from 12 rats were sliced with a Mcllwain mechanical tissue chopper as described below and perifused (2 hypothalami per chamber including the MBH and AH) using the Acusyst-S perifusion system to determine basal fl-EP and y3-MSH release and response to depolarization with 56 mM KC1 at the beginning and end of a 5 hr perifusion period. (b) To study the effect of naltrexone on fl-EP and y3-MSH release in vitro 12 chambers containing 3MBH each were perifused with control media and 11 chambers with 3MBH each were perifused with 10 6M naltrexone dissolved in control medium. Hypothalami were equilibrated and perifused as described below. Peptide release was measured at 20 min intervals over a 60 min period prior to depolarization with 56 mM KCI and for a 60 rain period following KCI. (c) To study the acute and chronic effects of naltrexone injection on /3-EP and y3-MSH release in vitro, rats were treated in vivo with naltrexone and sacrificed within 40-60 min after one injection or after 7 days of chronic treatment. For the acute study 24 rats were injected subcutaneously with naltrexone (4 mg/kg) and were sacrificed 40-60 min later, 22 rats injected with saline served as controls. To study the chronic effects of naltrexone on fl-EP and y3-MSH release, 12 rats were infused with naltrexone 2.4 rag/day by subcutaneously implanted osmotic minipump for 7 days; 12 rats with empty silastic implants served as controls. Animals were sacrificed and the hypothalami (2MBH per chamber) were perifused as described below.
2.2. Tissue dissection and extraction of peptides Rats were sacrificed by decapitation between 10.00 and 11.00 h and the brain was quickly removed and placed in ice cold Gibco 199 with Hank's salts, and L-glutamine. The brain was then placed in a lucite brain holder, ventral side up, with razor blades positioned at 2-mm intervals for cutting sections. The medial basal hypothalamus (MBH) was dissected from a 2-mm coronal section cut immediately caudal to the optic chiasm. The dissection was limited laterally by the hypothalamic sulci and dorsally by the mammillothalamic tract. In one experiment the anterior hypothalamus (AH) was dissected from the next rostral 2-mm coronal section and was bordered dorsally by the anterior commissure. For HPLC analysis, each MBH was homogenized immediately in 1 ml of 0.2 N HC1 and processed as described below. At the end of the perifusion period in some experiments, the hypothalami were removed from the perifusion chambers and homogenized in 2 ml of 0.2N HCI. The 4000 × g supernatant was used for peptide RIA.
2.3. In vitro perfusion of rat hypothalami To measure in vitro release of POMC-derived peptides hypothalami were perifused, essentially as described by Nikolarakis et al., [18] using the Acusyst-S perifusion system (Endotronics, Coon Rapids, MN). Hypothalami were sliced with a Mcllwain mechanical tissue chopper in the sagittal plane at 250/xm intervals. Two or three
26
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31
hypothalami were placed in each perifusate chamber which was preincubated at 37 C with preoxygenated Gibco 199 with Earle's salts, L-glutamine and NaHCO 3 2200 m g / l supplemented with an additional 0.8 g / l glucose and 1 g / l BSA. The system was continuously gassed with O2-CO 2 (95:5). A multiple channel peristaltic pump delivered the media to the tissues at a defined rate. Initially, the chambers were perifused at 1 m l / m i n x 20-30 min followed by a 20-30 min period at 0.1 m l / m i n to achieve a stable rate of peptide release. The perifusion continued at 0.1 m l / m i n during sample collection. Samples were collected every 20 min into tubes containing 0.4 ml of 0.2N HCI for acidification and inactivation of proteases. Samples were frozen and subsequently assayed for/3-EP and y3-MSH by RIA. In an initial study release of /3-EP and y3-MSH from the MBH and AH was determined over a 5 hr period and the response to 56 mM K C I x 3 0 rain was studied at the beginning and end of this perifusion period. Subsequently all experiments were performed using two or three MBH per chamber and peptide release was measured over a 60 rain period before depolarization with 56 mM KCI and for a 60 min period after KCI.
2.4. Radioimmunoassays The fl-EP concentrations in the MBH extracts and in the acidified perifusates were assayed with an antiserum to human /3-EP raised in this laboratory which crossreacts equally with camel /3-EP, N-acetyl /3-EP and fl-EP I 27 and 30% on a molar basis with human /3-1ipotropin [24]. Synthetic camel /3-EP (Peninsula Laboratories, Belmont, CA) was used as the standard. For the/3-EP measurements shown in Fig. 4, samples were evaporated in a Speed Vac Concentrator prior to RIA to increase assay sensitivity. The effective level of detection for fl-EP per 20 min fraction was 12 pg. We have previously shown that the /3-EP immunoactivity in hypothalamic homogenates measured with this RIA elutes upon gel filtration in the same position as synthetic fl-EP standard [24]. Further characterization by ion exchange chromatography or HPLC revealed that the fl-EP immunoactivity in the hypothalamus consists primarily of flEPt_31 and to a lesser extent fl-EPl_27 and fl-EPI_26 [23,24]. The 73-MSH concentrations in the MBH extracts and in the perifusates were measured with an antiserum provided by Drs. Meador-Woodruff and Huda Akil (University of Michigan) directed against the midportion of rat Y3-MSH; rat y3-MSH synthesized by Dr. Larry Taylor from the same group was used as the standard [14]. The y3-MSH immunoactivity in the hypothalamic homogenates and perifusates was characterized by gel filtration as described below.
Cls reverse phase column (Vydac, Hesperia, CA) with the Waters 510 and 590 solvent delivery system, the 680 automated gradient controller and the U6K Universal Injector. Samples were eluted with a gradient of 80% acetronitrile containing 0.1% TFA (solvent B) against 0.1% TFA in water (solvent A) at a flow rate of 1 ml/min. The concentration of solvent B rose in a linear fashion from 30 to 35% over the first 5 min and then from 35 to 45% over the next 45 rain and from 45 to 50% over the next 5 min. The column was then washed with 100% solvent B for 10 min and returned to the starting condition of 30% B. 0.5 ml fractions were collected, evaporated in a Speed Vac Concentrator and dissolved in buffer for/3-EP RIA. The column was calibrated each day with 5 ng each of synthetic camel 13-EPI_31, fl-EPl_27 and fl-EPl_26.
3. Results
3.1. Effect o f naltrexone on f l - E P processing When
the /3-EP immunoactivity
noted HPLC
compared
to 7 control
profile of a naltrexone
of 7
rats. An example treated
of the
rat compared
to
a c o n t r o l r a t is s h o w n in F i g . 1. I n t h e c o n t r o l a n i m a l ~
1-31
1-27
1-26
1~0
8OO
Control 6OO
& c~.
2OO
0
i
40
2.5. Gel filtration Perifusate samples from several control perifusions were pooled, dried in a Speed Vac Concentrator and then resuspended in 0.05 N HCI containing 0.1% BSA. For comparison the extract from a control hypothalamic homogenate was also used for gel filtration. Samples were chromatographed on a Sephadex-G-50 (superfine) column (0.9x50 cm) in 0.05 N HCI and 0.1% BSA with a flow rate of 0.2 ml/min. Fractions (1 ml) were collected and assayed in duplicate by RIA. Both y3-MSH and fl-EP were measured after gel filtration of the hypothalamic extract but the entire sample was used for y3-MSH RIA after gel filtration of the perifusates. The column was calibrated with blue dextran, fl-EP, and fl-LPH.
in the MBH
rats, infused for 1 week with naltrexone, was individually a n a l y z e d b y H P L C , s i g n i f i c a n t d i f f e r e n c e s w e r e
50
60
70
i 8O
8OO
o
Naltrexone
600
"~ ~oo & 2OO
0
2.6. High-performance liquid chromatography
40
50'
60'
? '0
80
traction
Individual aliquots (0.5 ml) of the 4000x g supernatant from each MBH were evaporated in a Speed Vac Concentrator and dissolved in 0.2 ml of 0.1% trifluoroacetic acid (TFA) containing 24% acetronitrile and subjected to high-performance liquid chromatography (HPLC). HPLC was carried out on a 5-mm, 300 A pore,
Fig. l. Reverse phase HPLC elution profiles of the/3-EP immunoactivity in extracts of the MBH obtained from a control rat (upper panel) and from a rat treated with naltrexone for 7 days (lower panel). Arrows indicate the elution positions of the synthetic peptides fl~-EPl_31, flc-EPi 27 and flc-EPi 26.
27
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31
although much of the fl-EP-immunoactivity eluted in the position of fl-EPt_3~, a considerable portion also eluted in the positions of fl-EPl_27 and fl-EPl_26. In contrast after naltrexone almost all the immunoactivity could be accounted for by fl-EPI_ar The mean ratio of fl-EPt_3] compared to fl-EPl_27 plus fl-EPt_26 was 2.34 _+0.41 in the naltrexone treated rats, significantly higher than the ratio of 1.26 _+0.09 in the control rats (P < 0.02). The mean ratio of ~-EPI_31 to /3-EP,_27 was 3.03 + 0.46 in the naltrexone treated animals; this was also significantly higher than the ratio of 1.95 + 0.13 in the control animals (P < 0.05).
i
1000
KCl
KCI i P,
!
i
..~ 800
E 600
400 m
200
3.2. [3-EPand y3-MSH release during hypothalamic perifusion
IL
0
i
i
6O
120
i
160
i
240
i
300
300
When 6 chambers were perifused containing slices of AH and MBH from 2 rats each, there was stable peptide release over a 5 h period. In addition there was a good response to depolarization with KC1 at the beginning and at the end of the 5 h period (Fig. 2). Subsequently only MBH slices were used as peptide release appeared to be more consistent between chambers for this dissection.
250 /
200
~I
E
150
"'b
j ~
100 e~
3.3. Characterization of peptide immunoactivity in the perifusates compared to hypothalamic homogenates
50 0
The y3-MSH immunoactivity in the hypothalamic perifusates eluted upon gel filtration in 2 distinct peaks (Fig. 3). The second peak eluted in the same position as the fl-EP standard which corresponds to what has been reported for glycosylated 73-MSH; [6] the larger molecular weight peak eluted just after the /3-LPH standard. A similar pattern of 73-MSH immunoactivity was seen when the hypothalamic homogenate was chromatographed. This corresponds with a previous report demonstrating 2 peaks of y3-MSH related immunoactivity in the rat hypothalamus with the first peak corresponding to a molecule of an apparent molecular weight of 7100 [6]. As reported previously all of the/3-EP immunoactivity in the homogenate eluted in the position of the synthetic /3-EP standard [24].
3.4. Effect of perifusion with naltrexone in uitro on y3-MSH and [3-EP release Baseline and KC1 stimulated released of 73-MSH was significantly higher from hypothalami perifused with naltrexone compared to the control hypothalami. Baseline y3-MSH release was 161 _+ 6.7 pg/60 min in the control group compared to 221 _+20 pg/60 min in the naltrexone group (P < 0.01). Following stimulation with KCI, y3-MSH release increased to 543 _+85 pg/60 min in the control group compared to 950 _+94 pg/60 min in the naltrexone group (P < 0.005) (Fig. 4). The
6O
i
i
i
i
120
180
240
300
time (mln) Fig. 2. Individual in vitro release profiles of y3-MSH (upper panel) and/3-EP (lower panel) when 3 chambers were perifused containing hypothalamic slices (MBH plus AH) from 2 rats each. Fractions were collected at 15 rain intervals over a 5 h period. Response to depolarization with KCI was determined after 1 h and after 4 h of perifusion.
baseline release of fl-EP from the naltrexone group of 136 _+45 pg/60 min was not significantly higher than the baseline release from the control group of 93 + 15 pg/60 min. However following KCI stimulation, fl-EP release from the naltrexone group was 558 _+ 103 pg/60 min, significantly higher than the release of 275 _+49 pg/60 min from the control group (P < 0.02) (Fig. 4).
3.5. Acute and chronic effects of naltrexone treatment in vivo on 73-MSH and fl-EP release Baseline T3-MSH release from the perifused hypothalamus was significantly higher in rats treated in vivo with naltrexone 40-60 min prior to the perifusion. Mean y3-MSH release was 194 + 29 pg/60 min in the saline-treated animals compared to 367 + 47 pg/60 min in the naltrexone-treated animals (P < 0.01) (Fig. 5). Response to KC1 stimulation was similar in both groups 651 + 87 vs. 656 _+84 pg/60 min. Baseline/3-EP
28
S.B. Jaffe et al. / Brain Research 648 (1994) 24-31 Vo
LPH
[~-EP
41-
41.
41,
I000
Vt
800
600
& 600
5OO 400
o °~
"~
[]
saline
•
naltrexone
400
~
200
;~"
<60 0 baseline stimulated baseline s t i m u l a t e d
2OO
Acute
Chronic
250
100
w r 0
2
•
I
I
30
40
•
I 50
200 ~D
fraction
150
Fig. 3. Sephadex G-50 chromatography of the -x3-MSH immunoactivity in pooled hypothalamic perifusates. The arrows indicate the void volume, the elution positions of /3-LPH and /3-EP and the total volume.
100 ~..
<60
CQ.
baseline
.E
1000 "
E
800"
[] Control • Nattrexone
o~ 4oo p--
stimulated b a s e l i n e
Acute
stimulated
Chronic
Fig. 5. Mean release per hour of y3-MSH_+ SEM (upper panel) and /3-EP +S.E.M. (lower panel) during hypothalamic perifusion (2 MBH per chamber) after either acute or chronic treatment, with naltrexone in vivo as compared to saline treatment. Samples were collected at 20 min intervals. Peptide release is depicted over a 60 rain period before and after stimulation with KCI. The dashed line indicates the cumulative level of peptide detection over a 60 min period. * P < 0.01 compared to saline treated animals.
200
Baseline
Stimulated
700' "~
600 ¸
-~
500-
T
400-
300"
200100
Baseline
Stimulated
Fig. 4. Mean release per hour of y3-MSH _+S.E.M. (upper panel) and /3-EP + S.E.M. (lower panel) during hypothalamic perifusion (3 MBH per chamber) with either 1 0 - 6 M naltrexone or control media. Samples were collected at 20 rain intervals. Peptide release is depicted over a 60 rain period before and after stimulation with KCI. * P < 0.02 vs. control.
release was below the limit of assay detection; no difference was noted in the response to KC1 in either group. In contrast after chronic treatment with naltrexone for 1 week baseline 73-MSH release was not different from the control animals; basal /3-EP release was again below the limit of assay detection. The -,/3-MSH and /3-EP responses to KC! stimulation, however, were significantly less in the naltrexone treated rats (Fig. 5). Mean ,/3-MSH release after KC1 was 860 + 117 p g / 6 0 min in the control rats compared to 234 + 38 p g / 6 0 min in the naltrexone-treated rats ( P < 0.001). Mean ~ - E P release after KCI was 170 + 45 p g / 6 0 min in the controls compared to 74.2 + 6.7 p g / 6 0 min after naltrexone ( P < 0.01). At the end of the perifusion the hypothalami from each chamber were homogenized and assayed for 73-MSH and /3-EP peptide content. As expected there was a significant decrease in peptide content in the naltrexone group. 73-MSH was reduced from 24,950 + 690 to 11,700 + 400 p g / c h a m b e r and /3-EP was reduced from 14,030 + 600 to 6420 +_ 190 p g / c h a m b e r ( P < 0.001). Thus there is
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31
an increase in POMC peptide release shortly after an initial dose of naltrexone. After 1 week of naltrexone, when the hypothalamic content of several POMC-derived peptides is decreased, baseline POMC-peptide release is unchanged from control animals, however, the response to KC1 is blunted.
4. Discussion
Evidence accumulated from several studies using opioid receptor antagonists supports the concept that there is feedback regulation of POMC by endogenous opioids. We have previously reported that treatment with naltrexone for 1 month caused a fall in the concentration of fl-EP in the hypothalamus where POMC is synthesized as well as in several other brain regions where POMC neurons project [20]. The fall in the concentration of /3-EP was paralleled by a fall in the concentrations of a-MSH, CLIP and y3-MSH, the other major POMC-derived peptide products in the hypothalamus [12]. A decline in peptide content was noted as early as 2 days of treatment. Naltrexone was also noted to have a significant effect on POMC gene expression in the MBH [4,12]. After 3 weeks of naltrexone, the concentration of POMC mRNA in the MBH was higher in naltrexone treated animals compared to the controls. This increase in the concentration of POMC mRNA was seen in both intact and in castrated animals and was thus unrelated to any potential effect naltrexone may have had on LH and testosterone levels. Although the increase in POMC gene expression associated with a fall in peptide levels was interpreted as being consistent with a compensatory increase in brain /3-EP synthesis and release in the setting of chronic opioid receptor blockade, it was unknown if in fact POMC peptide release was altered. In addition it was unclear if the immunoactive form of fl-EP that was being measured was in fact the biologically active opioid peptide ]3-EPl_3V /3-EP which results from the posttranslational processing of POMC can be further processed through N-terminal acetylation or C-terminal cleavage to yield fl-EPl_26 and fl-EPl_27 [6,21,29]. These modifications greatly reduce or abolish opioid activity. N-acetyl /3-EP which is inactive in opiate receptor binding assays, is not a major form of/3-EP in the hypothalamus. In contrast, fl-EPl_26 and 1-27 which are present in the hypothalamus, have opioid receptor binding activity but have little analgesic activity. In addition fl-EPl_27 has been shown to antagonize the analgesic effects of/3-EP1_31 [17]. We therefore determined the effect of naltrexone on the relative amounts of the major forms of /3-EP in the hypothalamus: ~-EPI_31 , fl-EPl_27 and fl-EPL_26. We have shown that in the setting of chronic opioid antagonism although all 3 forms of /3-EP decrease, there is relatively more
29
/~-EPI_31, the form of the peptide with opioid bioactivity, compared to fl-EPl_27 ' which has opioid antagonist activity, and /3-EP1_26. This is in agreement with the results of Bronstein and colleagues who have shown by gel filtration of pooled hypothalamic samples that naltrexone treatment reduces the peak coeluting with fl-EPl_27 and fl-EPI_26 relative to the /3-EPl_31 peak [4]. This change in peptide ratios could result from a specific effect on processing enzymes responsible for the C-terminal proteolysis of /3-EP or could be a reflection of increased/~-EP turnover and release with less time available for further processing. Additionally there may be relative changes in peptide degradation. Dopaminergic agents have been shown to influence /3-EP processing in the neurointermediate lobe of the pituitary [8,15]. For example in addition to causing changes in /3-EP synthesis and release dopamine antagonism has been shown to affect a-N-acetyltransferase activity in the intermediate lobe [15]. Changes in /3-EP processing have also been reported in the hypothalamus associated with sexual maturation and aging [13,26]. There is also some data to suggest that in morphine treated rats there is an increase in the opiate inactive forms of /3-EP in the hypothalamus [3]. We now show in the setting of chronic opioid blockade that there is a relative decrease in the same opiate inactive forms of/3-EP in the hypothalamus. In the present study, when naltrexone was administered either in vitro or in vivo, significant effects on POMC peptide release were noted from the perifused hypothalamus. In a previous study, Nikolarakis et al. have shown that naloxone (10-6M) produced a calcium-dependent increase in the non-stimulated release of fl-EP in the perifused hypothalamus in vitro [18]. In our study 10-6M naltrexone in vitro increased basal T3-MSH release and KCI stimulated y3-MSH and/3-EP release. There was an increase in the non-stimulated release of y3-MSH from the hypothalamus of animals treated with naltrexone 40-60 min prior to sacrifice. Since non-stimulated /3-EP release was below the level of assay detection in these studies no statement can be made about /3-EP release but in general the pattern of release of /3-EP and y3-MSH have paralleled each other in vitro. After treatment with naltrexone for 1 week baseline y3-MSH release, which had been elevated acutely, was not significantly different from control. KCl-stimulated y3-MSH, however, was significantly less than the controls; a similar effect was seen with /3-EP release. These responses occurred at a time when the hypothalamic concentrations of y3-MSH and /3-EP were shown to be decreased by 53% and 54% respectively. Non-stimulated peptide release however appeared to be normal despite the low peptide content. In fact, if normalized to the decreased MBH peptide content, basal y3-MSH release could be interpreted as relatively increased after chronic naltrexone
30
S.B. Jaffe et al. / Brain Research 648 (1994) 24-31
treatment. These findings are consistent with a decrease in peptide content associated with increased peptide turnover after chronic stimulation with naltrexone. Opioid agonists and antagonists have also been shown to have electrophysiological effects on POMC neurons and to affect opioid receptor binding. When intracellular recordings were made from arcuate neurons in the guinea pig, mu opioid agonists were found to induce membrane hyperpolarization and to decrease the spontaneous firing of POMC neurons [10]. This is consistent with the hypothesis that opioid receptors function as autoreceptors which respond to endogenous/3-EP release. Chronic exposure to opioid antagonists including naltrexone has also been shown to increase opioid receptor binding sites in brain [11,19,22, 28]. Opioid receptor upregulation is seen after 8 or more days of treatment. An increase in the analgesic potency of opiates was also noted after a period of pretreatment with naltrexone [22,28]. Functional supersensitivity therefore parallels the increase in receptor binding after chronic exposure to naltrexone. Thus the changes in POMC peptide release and gene expression that are measured after opioid receptor blockade [12,18] are accompanied by an enhanced ability to bind endogenous /3-EP. In summary the current studies demonstrate acute and chronic effects of naltrexone on POMC peptide release from the hypothalamus in vitro. There is also a change in peptide processing as shown by a shift in the forms of/3-EP found in the hypothalamus resulting in a relative increase in the biologically active opioid form of/3-EP. These results are consistent with an inhibitory feedback mechanism for the autoregulation of POMC by /3-EP. When taken together with previous studies our results provide further support for a compensatory increase in POMC peptide release and synthesis in the setting of opioid receptor blockade.
Acknowledgements The technical assistance of Ms. Marion Richman and Mr. Samuel Yum is greatly appreciated. This work has been supported by NIDA Grant DA 07732.
References [1] Akil, H., Watson, S.J., Young, E, Lewis, M.E., Khachaturian, H. and Walker, J.M., Endogenous opioids: biology and function, Annu. Ret:. Neurosci., 7 (1984) 223-255. [2] Atkinson, R.L., Endocrine and metabolic effects of opiate antagonists, Z Clin. Psychiatry, 45 (1984) 20 24. [3] Bronstein, D.M., Przewlocki, R. and Akil, H., Effects of morphine treatment on pro-opiomelanocortin systems in rat brain, Brain Res., 519 (1990) 102-11.
[4] Bronstein, D.M., Pre- and posttranslational regulation of/3-endorphin biosynthesis in the CNS: effects of chronic naltrexone treatment, J. Neurochem., 60 (1993) 40-49. [5] Crabtree, B.L., Review of naltrexone, a long-acting opiate antagonist, Clin. Pharm., 3 (1984) 273-280. [6] Emeson, R.B. and Eipper, B.A., Characterization of proACTH/endorphin-derived peptides in rat hypothalamus, J. Neurosci., 6 (1986) 837-849. [7] Ferin, M., Van Vugt, D. and Wardlaw, S., The hypothalamic control of the menstrual cycle and the role of endogenous opioid peptides, Recent Prog. Horm. Res., 40 (1984) 441-485. [8] Ham, J. and Smyth, D.G., Regulation of bioactive /3-endorphin processing in rat pars intermedia, FEBS Lett., 175 (1984) 407411. [9] Howlett, T.A. and Rees, L.H., Endogenous opioid peptides and hypothalamo-pituitary function, Annu. Reu. Physiol., 48 (1986) 527-536. [10] Kelly, M.J., Loose, M.D. and Ronnekleiv, O.K., Opioids hyperpolarize /3-endorphin neurons via 1-receptor activation of a potassium conductance, Neuroendocrinology, 52 (1990) 268-275. [11] Lahti, R.A. and Collins, R.J., Chronic naloxone results in prohinged increases in opiate binding sites in brain, Eur. J. Pharmacol., 51 (1978) 185-186. [12] Markowitz, C.E., Berkowitz, K.M., Jaffe, S.B. and Wardlaw, S.L., Effect of opioid receptor antagonism on proopiomelanocortin peptide levels and gene expression in the hypothalamus, Mol. Cell. Neurosci., 3 (1992) 184 190. [13] Martensz, N.D., Changes in the processing of /3-endorphin in the hypothalamus and pituitary gland of female rats during sexual maturation, Neuroscience, 16 (1985) 625-640. [14] Meador-Woodruff, J.H., Pellerito, B., Bronstein, D., Lin, H-L., Ling, N. and Akil, H., Differential effects of haloperidol on the rat pituitary: decreased biosynthesis, processing and release of anterior lobe pro-opiomelanocortin, Neuroendocrinology, 51 (199(/) 294-303. [15] Millington, W.R., O'Donohue, T.L., Chappell, M.C., Roberts, J.L. and Mueller, G.P., Coordinate regulation of petide acetyltransferase activity and proopiomelanocortin gene expression in the intermediate lobe of the rat pituitary, Endocrinology, 118 (1986) 2024-2033. [16] Mocchetti, 1, Ritter, A. and Costa, E., Down-regulation of proopiomelanocortin synthesis and beta-endorphin utilization in hypothalamus of morphine-tolerant rats, J. Mol. Neurosci., 1 (1989) 33 38. [17] Nicolas, P. and Li, C.tt., /3-Endorphin-(l-27) is a naturally occurring antagonist to etorphine-induced analgesia, Proc. NatZ Acad. Sci. USA, 82 (1985) 3178-3181. [18] Nikolarakis, K.E., Almeida O.F.X. and Herz, A., Feedback inhibition of opioid peptide release in the hypothalamus of the rat, Neuroscience, 23 (1987) 143-148. [19] Pert, C., B. and Snyder, S.H., Opiate receptor binding enhancement by opiate administration in vivo, Biochem. Pharmacol., 25 (1975) 847-853. [20] Ragavan, V.V., Wardlaw, S.L., Kreek, M.J. and Frantz, A.G., Effect of chronic naltrexone and methadone administration on brain immunoactive ,6- endorphin in the rat, Neuroendocrinology, 37 (1983) 266-268. [21] Smith, A.I. and Funder, J.W., Proopiomelanocortin processing in the pituitary, central nervous system, and peripheral tissues, Endocr. Rel~., 9 (1988) 159-179. [22] Tempel, A., Gardner, E.L. and Zukin, R.S., Neurochemical and functional correlates of naltrexone-induced opiate receptor upregulation, J. Pharmacol. Exp. Ther., 232 (1984) 439-444. [23] Wardlaw, S.L., Effect of androgens on hypothalamic proopiomelanocortin, Neuroendocrinology, 47 (1988) 164-168. [24] Wardlaw, S.L., Regulation of /3-endorphin, corticotropin-like intermediate lobe peptide, and o:-melanotropin-stimulating hot-
S.B. Jaffe et al. /Brain Research 648 (1994) 24-31 mone in the hypothalamus by testosterone, Endocrinology, 119 (1986) 19-24. [25] Wildt, L. and Leyendecker, G., Induction of ovulation by the chronic administration of naltrexone in hypothalamic amenorrhea, J. Clin. Endocrinol. Metab., 64 (1987) 1334-1335. [26] Wilkinson, C.W. and Dorsa, D.M., The effects of aging on molecular forms of beta- and gamma-endorphins in rat hypothalamus, Neuroendocrinology, 43 (1986) 124-131. [27] Yen, S.S.C., Quigley, M.E., Reid, R.L., Ropert, J.F. and Cetel, N.S., Neuroendocrinology of opioid peptides and their role in
31
the control of gonadotropin and prolactin secretion, Am. J. Obstet. Gynecol., 152 (1985) 485-493. [28] Yoburn, B.C., Nunes, F.A., Adler, B., Pasternak, G.W. and Inturrisi, C.E., Pharmacodynamic supersensitivity and opioid receptor upregulation in the mouse, J. Pharmacol. Exp. Ther., 239 (1986) 132-135. [29] Zakarian, S. and Smyth, D.G., fl-Endorphin is processed differently in specific regions of rat pituitary and brain, Nature, 296 (1982) 250-252.