Embed Size (px)
Citation preview
Gynecol. Endocrinoi . 2 (1988) 1-10
Pulsatile (3-endorphin release f rom the human pituitary in vitro
W. G. Rossmanith’, M . Gambacciani’, J. H. Liu3, W. H. Swartz, V. S. Tueros, S. S. C. Yen4 and 11. D . Rasmussenj
D e p a r t m m t qf Reproductivt, Medicine, School of Medicine (T-002), University qf Cal i forn ia , S a n Diego, L a J o l l a , Cal i fovnia 92093, U S A
Abstract
An in vitro perifusion system was used to characterize spontaneous immunoreactive 6-endorphin ($-END) release from 10 human fetal (20-23 weeks gestation) and 2 human adult anterior pituitaries.
Spontaneous $-END release from fetal anterior pituitaries was pulsatile, with a mean ( A SE) pulse interval of9.1 f. 0.5 minutes, pulse amplitude of 120.8 k 16.1 pg with nadir to peak increment of 106.0 f 32.9%, and overall release rate of209.7 t 65.0 pgi2 minutes. Blockade of calcium activity with 10 p M verapamil and 4 m M EGTA suppressed the frequency and amplitude of the spontaneous pulsatile iP-endorphin release (n = 2). Adrninistration of 2 n M human CRF for 20 minutes at the end of 2 p e r i h i o n s induced 205 and 883% increases of $-END release over the preceding basal levels. Administration of 2 nM CRF for 50 minutes at the end of another perifusion led to a greater and prolonged increase (maximum 4620% relative to the immediately preceding basal level) in $-END release. Addition of 56 m M KC1 during the last 20 minutes of this prolonged CRF stimulation further increased $-END release (to 7680% relative to the baseline preceding the CRF stimulation). Each of 4 quarters of adult anterior pituitaries (2 quarters each from 1 male and 1 female) also released $-END in a pulsatile fashion, with a pulse interval of 11.8 iz 2.0 minutes, pulse amplitude o f7 .4 k 0.8 ng with nadir to peak increment of 51.4 k 15.3%, and overall release rate of21.7 f 2.9 ng/2 minutes.
1 Research Fcllow in Reproductive Endocrinology. Present address: Ikpartnient of Obstetrics-Gynecology, University of Ulm, Prittwitzstrasse 43. 11-7900 U l m i l l . , West Germany
2 Research Fellow in Reproductive Endocrinology. Present address: Ilepartmcnr of Obstetrics and Gynecology, University of Pisa, Via Roma 67, 56100 I’isa, Italy
3 Clinical Associate Physician, Clinical Research Center 4 Clayton Foundation Investigator 5 To whom all correspondence should be addressed
1
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
2 Rossmanith, Gambacciani, Liu, Swavtz, Tuevos, Yen and Rasmusserr
These studies demonstrate that $-END release from the isolated human anterior pituitary in vitvo is characterized by high-frequency pulses, independent of hypothalamic stimulation. This spontaneous calcium-dependent pulsatile $-END release apparently reflects the activity of an intrapituitary pulse-generating mechanism.
Introduction
Previous studies by our laboratory',' and other^^,^ have suggested that a variety of adenohypophyseal hormones are released in an intrinsically episodic fashion, independent of hypothalamic regulation and possibly reflecting common basal secretory mechanisms. Although hypothalamic regulation of pituitary 6-endorphin (@-END) secretion has been extensively studied5-', this basal release pattern of pituitary P-END in the absence of hypothalamic modulation has not been elucidated. Accordingly, a perifusion system was utilized to analyze spontaneous P-END release from human anterior pituitaries irz vitvo.
Methods Permission to use human pituitaries was obtained from the University of Cali- fornia, San Diego Human Subjects Committee. Pituitaries from 5 male and 5 female fetuses were obtained at autopsy immediately following urea and PGF2, induced termination and delivery at 20-23 weeks of gestation. Gestational age was estimated from body weight and crown-rump length'". The time between induction and delivery was less than 6 hours in all cases. After the intermediate and posterior lobes had been removed with the aid of a dissecting microscope, each anterior pituitary was briefly rinsed in 10 ml Medium 199 (Grand Island Biological Co., New York) and then placed in a flow-through perifusion chamber (50 p1 volume) maintained at 37 "C and perfused by oxygenated Medium 199 (pH 7.3) containing 0.05% bovine serum albumin (Pentex, Miles Laboratories). After a 150 minute stabilization period, the perifusion effluent was collected in 2 minute fractions for 160 minutes into chilled tubes containing 0.4 ml phosphate buffer with 0.1% triton and 50 mM EDTA, and then frozen (-70 "C).
In the first study, 5 fetal anterior pituitaries (2 male and 3 female) were each perfused with medium alone for the entire 160-minute sample collection period. In order to ascertain whether fluctuations in immunoreactive $-END levels during these perifusions represented true pulsatile release and not merely variability due to experimental and assay procedures, an additional control perifusion was performed with synthetic human p-END 1-31 (Sigma, St Louis, Missouri) added to the medium perfusing an empty chamber.
In a second study, 2 fetal anterior pituitaries (1 male and 1 female) were peri- fused with medium alone for the first 60 minutes of the sampling period and then with medium containing 4 mM of the calcium chelator EGTA and 10 pM of the membrane calcium channel blocker verapamil (Sigma, St Louis) for the remaining 100 minutes. Two additional fetal anterior pituitaries (male) were each perifused for 140 minutes with medium alone and then for 20 minutes with
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
Pulsatilr [ h n d o v p h i n Yelease-fiom the human pituitavy in vitro 3
medium containing 2 n M human CRF (Peninsula Lab., Belmont, California). One fetal anterior pituitary (female) was perifused with medium alone for the initial 110 minutes, and then with 2 n M CRF added to the medium for the remaining 50 minutes of the sample collection period, 56 m M KCI also being added during the last 20 minutes of this period.
In a final study, 1 anterior hemipituitary from a 39-year-old woman and 1 anterior hemipituitary from a 65-year-old man were obtained at autopsy within 6 hours post mortem. The other hemipituitary from each w a s utilized in other studies. The hemipituitaries were cut in half and each resulting quarter pituitary was placed in a separate 0.25 nil chamber and perifused at a flow rate of 0.6 ml/minute. After a 150-minute equilibration period the perifusate effluent was collected in 2-minute fractions for 160 minutes.
Perifusates from 2 additional fetal anterior pituitaries were pooled and ex- tracted on reverse phase C18 columns (Bond-Elut, Analytichem Int., Harbor City, California) following the protocol of Cahill rt a/.". Recovery of known amounts of (J-END added to the perifusion medium was 85595%. The dried extracts were reconstituted in 0.5 ml phosphate buffer with 0.lo/o Triton-X (PBT), applied directly to a Sephadex G-50 superfine column (0.9 X 55 cm) at 4 "C, eluted with PBT, and collected in 1 ml fractions.
$-END concentrations in perifusate and eluant fractions were determined in triplicate using a previously described radioimmunoassay ", with all samples from a single perifusion included in a single assay. The sensitivity of the assay was 2.7 pg/tube (corresponding to 32 pg released/2 minutes). The intra-assay coefficient of variation (CV) was 5.6%. Neither KCI, EDTA, verapa.mil nor CRF interfered with the assay at the concentrations used in these studies.
Pulsatile increases of $-END concentration in the sequential perifusate fractions were identified utilizing the PULSAR pulse detection algorithm devel- oped by Merriam and Wachter13. This program identifies pulses by evaluating the amplitude and duration of individual fluctuations in terms of multiples of the standard deviation (SII) of the assay. We modified this procedure by utilizing the SD of the results of the control (i.e., no tissue, but with human P-END added to the medium) perifusion instead of the assay SD. I n this way the SD reflected overall experimental variation (including the intra-assay variation). To be detected as a significant pulse, the $-END concentration of a single fraction had to be more than 2.0 SD greater than the smoothed baseline, calculated from the corresponding locally weighted moving average of 20 sequential fractions. The $-END concentration for 2, 3, 4 or 5 consecutive fractions had to be 1.5, 1 .O, 0.75 and 0.5 S D greater than the smoothed baseline, respectively, in order to be designated as a pulse.
Results
Each of the 10 human fetal anterior pituitaries released $-END in a pulsatile fashion. The mean I(* SE) pulse interval was 9.1 f 0.5 minutes, with a pulse amplitude of 120.8 :+_ 16.1 pg, nadir to peak increment of 106.0 f 32.9%, and overall release rate of209.7 k 65 pg/2 minutes. In contrast, no significant pulses were detected in the control perifusion (Figure l), confirming empirically that
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
4 Rossmanitk, Gambacciani, Liu, Swartz, Tueros, Yen and Rasmussen
- ,I. , I. i Q 22 wk d 22 wk N
.- m $I 20 wk
300 I d 22 wk
I
Figure 1 $-END concentrations during perifusion ofan empty chamber with P-END added to the medium (top left panel) and during perifusions of human fetal anterior pituitariek (2 male, 3 female) with medium alone. Asterisks indicate significant pulses.
our criteria for characterizing pulses were appropriately conservative (i. e., less than 1/80 or 1.25% false-positive detections). When the fractions from one perifusion were assayed twice in separate assays, there was a good concordance in the $-END profiles (Figure 2), with a significant (p < 0.01, r = 0.75) corre- lation between individual values in the 2 assays.
In 2 perifusions, addition of the calcium chelator EGTA and membrane calcium channel blocker verapamil reduced the $-END release rate (from 742.3 f 61.6 to 220.6 f 23.5 pg/2 minutes and from 357.7 k 36.0 to 7.8 k 0.1 mg/ 2 minutes), $-END pulse frequency (from 1 pulse/7.5 minutes to 1 pulse/ 11.1 minutes and from 1 pulse/7.5 minutes to 1 pulse/100 minutes), and pulse amplitude (from 458.7 k 105.3 to 106.0 f 58.1 pg and from 353.7 f 77.2 to 6.2 pg) (Figure 3). In 2 perifusions 20-minute stimulation by 2 nM CRF at the end of the perifusion induced 205 and 883% increases over the immediately preceding $-END release. In another experiment, administration of 2 nM human CRF for 50 minutes at the end of the perifusion led to a prolonged increase (maximum 4620% relative to the immediately preceding basal level)
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
Pulsatilr P-endorphin rrleasr f r o m the human pituitary in vitro
200 - z 2 N
8 P
. Y
w 2 1 0 0 . W J w a n
m z W I
- 20 wk
* * * I .
I
80 160 01 '
0
MINUTES
Figure 2 $-END concentrations in the perifusate samples of 1 h u m a n fetal anterior pituitary perifusion assayed twice in separate assays. Results of the separate assays are represented as closed 11s open circles. Two asterisks denote pulses detected in the results of both assays.
in $-END release. The addition of56 m M KCl during the last 20 minutes ofthis CRF stimulation further increased $-END release (to 7680% relative to the baseline preceding the CRF stimulation) (Figure 4). There were no consistent age- or sex-related differences in $-END release rate or pulsatility.
Each of the 4 quarters of the adult anterior pituitaries also released $-END in a pulsatile pattern (Figure S), with pulse intervals of 11.8 f 2.0 minutes, pulse amplitude of7 .4 k 0.8 ng, nadir to peak increments of51.4 k 15.3% and overall $-END release rate of21.7 iz 2.9 ng/2 minutes.
The @-END detected in the perifusates represented primarily authentic p-END, as demonstrated by gel exclusion chromatography of the extract of pooled perifusate collected during 2 fetal anterior pituitary perifusions (Figure 6). The most prominent immunoreactive peak coeluted with synthetic human f3-END 1-31, whereas less than 20% coeluted with the related endogenous opiate peptide human 6-lipotropin (P-LPH) or other unknown cross-reacting substances.
Discussion
This is the first demonstration that $-END is released from the human anterior pituitary in a high-frequency pulsatile pattern, independent of hypothalamic in- puts. The corresponding lack of significant pulsatile fluctuations in the P-END control perifusion and the concordance of $-END pulses in 2 separate assays
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
6
J 1000. w K
0 8 0 0 . Z LfJ .@ 600
400
200
0 -
Rossmanith,
-
-
.
Gambacciani, Liu, Swavtz, Tuevos, Yen and Rasmussen
d 22 2ooo r *
wk EGTA + VERAPAMIL
1600 -
1200 . n
z 800 -
cu \
g 4 0 0 -
Y
w v)
O - '
*
.A
EGTA + VERAPAMIL ' 1 : 21 wk I ,
I * 0 80 160
MINUTES
Figure 3 $-END release from 2 human fetal anterior pituitaries (both male) perifused with medium alone and then with 4 mM EGTA and 10 p M verapamil added to the medium, as indicated by the solid bars. Asterisks indicate significant pulses.
o f the same perifusate confirm that these fluctuations were not merely due to either procedural o r assay variability. The functional viability of the tissue was confirmed by the $-END secretory response to CRF and KCI stimulation at the end of the perifusion period. Furthermore, the ability of the calcium chelator EGTA and membrane channel blocker verapamil to suppress the amplitude and frequency o f these spontaneous $-END pulses confirmed that the pulsatile release was due to a calcium-dependent secretory process.
Spontaneous pulsatile release of prolactin from grafted pituitaries in rats3 and o f growth hormone and prolactin from perifused monkey hemipi tu i ta r ie~~ has been reported to occur with peak intervals of between 8 and 11 minutes. In
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
Pulsatile fi-endorphir? release-from the human pituitary in vitro
5 0 0
400
300
2 0 0 .
1 0 0 .
-
.
.
8 0 0 . cu \
0 0 0 . v
w cn 5 4 0 0 .
-I w Q:
0 z o o - z w I
0 - m
d 21 wk
(1000.
CRF -
CRF - * d 20 wk t
9 23 wk CRF ,Xd * t
MINUTES
7
Figure 4 $-END release from 3 human fetal anterior pituitaries (2 male and 1 female) perifused with medium alone and with 2 n M CRF and 56 m M KCI added to the medium, as indicated by the solid and open bars, respectively. Asterisks indicate significant pulses.
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
8
W u .
Rorsmanith, Gambacciani, Liu, Swartz, Tueror, Yen and Rasmussen
' . D 30
m 2 0 .
1 0 .
V, B-LPH B-END + + +
. F f i
0 . .
50
40
= 5 30 0 P Y
a
m z y 20
10
I
0 5 10 15 20 25 30 35
FRACTION (ml)
O L '
Figure 6 Gel exclusion (Sephadex G-50) column chromatography of an extract of pooled anterior pituitary perifusates. Peak heights are expressed as $-END equivalents. V, = void volume; P-LPH = human P-lipotropin retention volume; P-END = human P-endorphin 1-31 retention volume.
-
-
-
-
-
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
Pulsarile P-endorphin release-from the human pituitary in vitro 9
addition, w e have recently demonstrated pulsatile secretion of A C T H and LH from human pituitaries in vitro'.', with pulses occurring every 11-15 minutes. Together with the present demonstration of intrinsically pulsatile p-END re- lease at intervals of approximately 9 minutes, these similarities in the periodicity of spontaneous episodic hormone release suggest similar intrapituitary pulse- generating mechanisms.
As opiate binding sites have been demonstrated on cultured gonadotrophsI4, and the addition of either the opiate receptor antagonist naloxone or 6-END antiserum to pituitary cells in culture has been shown to stimulate gonadotropin secretion".", it is possible that intrinsically pulsatile p-END release may play a role irl the intrapituitary regulation of gonadotropin secretion. However, the true physiological significance of this spontaneous pulsatile pituitary hormone release remains to be determined.
Acknowledgements
This work was supported in part by NICHD Grant P50 HD12303 Unit 4 (D. D. Rasmussen), an Andrew Mellon Foundation Fellowship (M. Gambacciani), a German Research Foundation Grant Ro 657/1-2 (W. G. Rossmanith), and conducted in part by the Clayton Foundation, California Division (S. S. C. Yen).
We thank Carla LaPorte and Del Alsobrook for their excellent technical and secretarial assistance.
References
1. Gambacciani, M., Liu, J. H., Swartz, W. H., Tueros, V. S., Yen, S. S. C. and Rasmussen, D. D. (1987). Intrinsic pulsatility of luteinizing hormone release from the human pituitary in vitro. N-euroendocrinology, 45, 402
2. Gambacciani, M., Liu, J. H., Swartz, W. H., Tueros, V. S., Rasmussen, 11. D. and Yen, S. S. C. (1987). Intrinsic pulsatility of A C T H release from the human pituitary in vitro. Cl in . Endocrinol., 26, 557
3. Shin, S. H. and Reifel, C. W. (1981). Adenohypophysis has an inherent property for pulsatile prolactin secretion. Neuroendocrinology, 32, 139
4. Stewart, J. K. , Clifton, D. K. , Koerker, D. J . , Rogol, A. D., Jaffe, T. and Goodner, C. J. (1985). Pulsatile release of growth hormone and prolactin from the primate pituitary in vitro. Endocrinology, 116, 1
5. Haracz, J. L., Bloom, A. S., Wang, R. I. H. and Tseng, L. F. (1981). Effect of intraventricular P-endorphin and morphine on hypothalamic-pituitary- adrenal activity and the release of pituitary P-endorphin. Neuroendocrino- logy, 33, 170
6. Vermes, I., Mulder, G. H., Smelik, P. G. and Tilders, F. J. H . (1980). Differential control of P-endorphin/@-lipotropin secretion from anterior and intermediate lobes of the rat pituitary gland in vitro. Life Sci., 27, 1761
7. Abou Samra, A . B., Fevre-Montange, M. , Loras, B., Durand, A , , Tourni- aire, J. and Bertrand, J. (1985). Effect of indolamines on beta-endorphin release by rat anterior pituitary cells. Neuroendocrinology, 41, 490
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
10 Rossmanith, Gambacciani, Liu, Swartz, Tueros, Yen and Rasmussen
8. Chan, J. S. D., Lu, C. L., Seidah, N. G. and Chretien, M. (1982). Cortico- tropin releasing factor (CRF) : effects on the release of proopiomelanocortin (P0MC)-related peptides by human anterior pituitary cells in vitro. Endo- crinology, 111, 1388
9. Sapun-Malcolm, D., Farah, J. M. Jr and Mueller, G. P. (1986). Serotonin and dopamine independently regulate pituitary p-endorphin release in vivo. Neuroendocrinology, 42, 191
10. Patten, B. M. (1968). Human Embryology 3, p. 142. McGraw-Hill, New York
11. Cahill, C. A., Matthews, J. D. and Akil, H. (1983). Human plasma p- endorphin like peptides: a rapid, high recovery extraction technique and validation of radioimmunoassay. J . Clin. Endocrinol. Metab., 56, 992
12. Wilkes, M. M., Stewart, R. D., Bruni, J. F., Quigley, M. E., Yen, S. S. C. , Ling, N. and Chretien, M. (1980). A specific homologous radioimmuno- assay for human P-endorphin: direct measurement in biological fluids. J . Clin. Endocrinol. Metab., 50, 309
13. Merriam, G. R. and Wachter, K. W. (1982). Algorithms for the study of episodic hormone secretion. Am.]. Physiol., 243, E310
14. Simantov, R. and Snyder, S. H. (1977). Opiate receptor binding in the pituitary gland. Brain Res., 124, 178
15. Blank, M. S., Fabbri, A, , Catt, K. J. and Dufau, M. L. (1986). Inhibition of luteinizing hormone release by morphine and endogenous opiates in cultured pituitary cells. Endocrinology, 118, 2097
16. Cacicedo, L. and Sanchez Franco, F. (1986). Direct action ofopioid peptides and naloxone on gonadotropin secretion by cultured rat pituitary cells. Lfe Sci., 38, 617
Gyn
ecol
End
ocri
nol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Mel
bour
ne o
n 10
/26/
14Fo
r pe
rson
al u
se o
nly.
