Molecular

and Cellular Endocrinolow,

c 1991 Elsevier

MOLCEL

Scientific

78 (1991)

Publishers

Ireland,

33-44

33

Ltd. 0303-7207/91/$03.50

02504

G proteins in normal rat pituitaries and in prolactin-secreting pituitary tumors Claudia

Bouvier ’

‘, H&he

Research

Forget ‘, Ginette Lagack ‘, Roman Drews ‘, Daniel Sinnett Damian Labuda 2 and Robert Collu ’

Unit on Reproducrice

SteJustine

Pituitary

and Developmental

Biology,

Hospital and University of Montreal,

(Received

Key words: G protein;

tumor;

rat

Prolactin

7 November

’ Medical

Montreal,

1990; accepted

Genetrcs, Research

Quebec H3T

30 January

2,

Center,

I C5, Canada

1991)

secretion

Summary It is still undetermined which GTP-binding (G) protein is involved in the regulation of prolactin (PRL) and through which effector. This study shows that, when compared to normal pituitary tissue, the levels of cyo protein were very low in dopamine (DA)-resistant, PRL-secreting pituitary tumors 7315a and MtTWl5, while (YOmRNA was present in the two tumors. In the MtTW15 tumor ail. oli2 and ai levels were decreased while those of as42 and as47 were increased, and in the 7315a tumor ai2, c*i3 and p levels were decreased and those of (rs47 increased. In an estrone-induced, DA-sensitive prolactinoma the levels of a13 were greatly reduced. DA was unable to inhibit basal PRL release by 7315a and MtTW15 and basal CAMP accumulation by adenomatous and MtTW15 cells. Vasoactive intestinal peptide (VIP) increased both CAMP accumulation and PRL release by all cell preparations which could be suppressed by DA with adenomatous and 7315a but not with MtTW15 cells. These and previously published results provide circumstantial evidence that (~0, ail and a(i3 are all involved in the transduction of the DA inhibitory message while (~~47 transduces CAMP activating messages and as42 is responsible for the constitutive activation of L-type Ca*+ channels, adenylate cyclase and baseline PRL release. release

Introduction The inhibition of pituitary prolactin (PRL) release by dopamine (DA) is now well documented (MacLeod, 1976; Collu, 1981). This inhibitory effect is exerted through D,-dopaminergic receptors (Cronin et al., 1978; Stefanini et al., 1980) which

Address for correspondence: Dr. Robert Collu. Research Center, Ste-Justine Hospital, 3175 C6te Ste-Catherine, Montreal, QuCbec H3T lC5, Canada.

are negatively linked to adenylate cyclase (Kebabian and Calne, 1979; Onali et al.. 1981; Cronin et al., 1983) via guanine nucleotide (G) binding proteins (Kuno et al., 1983; Bouvier et al., 1986, 1987; Senogles et al., 1987; Elazar et al.. 1989) as documented for other receptor complexes (Birnbaumer et al., 1985; Lefkowitz et al., 1988). Recently, D, receptors have been purified (Williamson et al., 1988) and cloned (Bunzow et al.. 1988) and their glycoprotein nature has been established (Grigoriadis et al., 1988). However, the intracellular mechanisms implicated in the inhibi-

34

tion of PRL secretion are still under investigation. Several studies have recently indicated that the dopaminergic inhibition is probably exerted via several second messenger pathways (Vallar and Meldolesi. 1989), such as adenylate cyclase (Enjalbert and Bockaert. 1983) phosphoinositide turnover (Login et al., 1988; Pizzi et al., 1988) and intracellular calcium mobilization (Login et al., 1988). However, these effecters have not yet been related to any specific G protein transducer. G proteins are heterotrimers composed of three distinct subunits, (Y, p and y (Bockaert et al., 1987). The (Ysubunit of each G protein is distinct and seems to determine specificity in terms of receptor and effector coupling (Spiegel, 1987). Recent studies have shown the existence of multiple types of G protein, one of which (Gi) transduces inhibitory (Gilman, 1984) while another (Gs) transduces stimulatory messages to adenylate cyclase (Robishaw et al., 1986). Recently, several forms of Gsa have been cloned resulting from alternative splicing of a single mRNA (Robishaw et al.. 1986). These forms have been implicated in both stimulation of adenylate cyclase and activation of Cal+ channels (Mattera et al.. 1989). Multiple forms of Gicu resulting from separate genes have also been detected (Beals et al.. 1987; Murphy et al., 1987) and isolated from cDNA libraries: cuil, a(i2 and ai (Jones and Reed, 1987; Itoh et al., 1988). These Gi proteins appear to be linked to various second messenger systems in several tissues (Murphy et al., 1987). According to Attali and Vogel (1989) cyil is implicated in the inhibition of adenylate cyclase by kappa receptors, and also probably to inhibition of Ca2+ currents induced by kappa opiate agonists in rat spinal-dorsal root ganglion cultures (Attali et al., 1989). ai has been reported to mediate a2-adrenergic inhibition of adenylate cyclase in platelets (Simonds et al., 1989) and ai to be linked to cation channels (Mattera et al., 1989) in the renal medulla collecting duct (Light et al.. 1989). However, it is still undetermined whether ail, oli2 or Lyi3 or even all three are actually implicated in the dopaminergic inhibition of pituitary PRL release. Moreover. the by complex may also play a role since it can activate K+ channels in the heart (Kim et al., 1989). Another G protein called Go has also been identified (Van Meurs et al., 1987),

which function is still under investigation. Go has been found to be linked to Kf channels in the brain (Van Dongen et al., 1988) where there is a high concentration of this protein indicating that it may be linked to neuronal activity (Rouot et al., 1987). Interestingly, DA-induced decrease in Cal+ current in snail neurons is mediated by Go (Harris-Warwick et al., 1988). Recently we have studied the mechanisms of the regulation of PRL release using a pituitary DA-sensitive, PRL-secreting adenoma, and two transplantable, DA-resistant PRL-secreting tumors. 7315a and MtTWl5. The DA-resistant tumors harbor structural anomalies in their D, receptor complex, both at the D, receptor level, and in terms of G protein content. Indeed we have found that D, receptors present in 7315a and MtTWl5 tumors exist in various abnormal polymeric forms and are not associated with G proteins (Bouvier et al., 1990a). In addition, the content of (YO was found to be very low in both tumors while that of total ai and p subunits was normal (Collu et al., 1988). Our studies have also shown the existence of several differences in tumoral lactotroph response to inhibitory and stimulatory messages which may be due to anomalies present in receptor complexes (Lafond et al., 1989). In addition. Lafond et al. (1986) have demonstrated that the dopaminergic inhibition of PRL release by normal lactotrophs is mediated by two mechanisms: a Nat-dependent one operating under basal conditions, and a Nat-independent one which operates under stimulated conditions. With the present study we have looked for the presence of other anomalies in G protein content in 7315a and MtTWlS tumors. In addition, in order to try and determine in an indirect way whether and which G proteins are involved in CAMP-mediated modulation of PRL release, we have concomitantly evaluated, by immunoblot, the levels of various G proteins in several pituitary tissues as well as the CAMP accumulation and PRL release by cells obtained from the same tissues, under stimulatory and inhibiting conditions. Since (YOis almost completely absent in the two DA-resistant tumors (Collu et al., 1988) we have also verified, by Northern blot analysis, the presence of OLOmRNA in the various pituitary tissues, concomitantly with that of (YSmRNA.

35

Materials and methods Drugs and materials [“51]anti-IgG was purchased from DuPont Canada (Mississauga, Ontario, Canada) and Nitroplus 2000 from Micron Separations (Westborough, MA, U.S.A.). CAMP and ACTH radioimmunoassay (RIA) kits were purchased from Incstar Corporation (Stillwater, MI, U.S.A.), and estradiol RIA kit from Baxter Dade (Diidingen, F.R.G.). Vasoactive intestinal peptide (VIP) was obtained from Peninsula Laboratories (Belmont, CA, U.S.A.), tissue culture products from Gibco Canada (Burlington, Ontario, Canada) and culture dishes from Flow Laboratories (McLean, VA, U.S.4.). Methodology Induction of adenoma and tumors. Pituitary adenomas were induced by S.C. implantation of Silastic pellets filled with 20 mg of estrone in female inbred Fisher 344 (Charles River Canada, St-Constant, Quebec, Canada) ovariectomized rats, 85-105 g of body weight, as previously described (Bouvier et al., 1986). Buffalo and Wistar-Furth female rats (Harlan Sprague-Dawley, Indianapolis, IN, U.S.A.) 180 g of body weight were inoculated S.C. with a mince of 7315a and MtTWl5 tumor, respectively as previously described (Bouvier et al., 1987). Animals were housed under controlled conditions of light (12 h of light starting at 06:OO h) and temperature (22” C) and given water and rat pellets ad libitum. Fisher rats were killed by decapitation 4 months, and Buffalo and Wistar-Furth rats 3 and 6 weeks after implantation, respectively. Determination of G protein levels. The various pituitary tissues were removed after the animal sacrifice and used for membrane preparation. A pool of anterior pituitaries obtained from Sprague-Dawley female rats 180-200 g of body weight was arbitrarely chosen as control for all G protein assays except when indicated. Indeed, no significant differences in G protein pituitary levels were found between Sprague-Dawley, Fisher 344, Buffalo and Wistar-Furth strains of rats (data not shown). Tissues were homogenized in 25 mM TrisHCl, 0.25 M sucrose, 20 mM EDTA, pH 7.2 containing protease inhibitors (benzamidine 15

pg/ml, leupeptin 5 pg/ml, trypsin inhibitors 5 pg/ml and phenylmethylsulfonyl fluoride 1 mM) and centrifuged, as previously described (Bouvier et al., 1990a). The membrane pellets were kept for less than one day at - 80 o C. Membrane preparations (50 pg of protein) were resuspended in Tris-HCl 62.5 mM, glycerol lo%, sodium dodecyl sulfate (SDS) 2%, &mercaptoethanol 5% and bromophenol blue O.OOl%, boiled at 100” C for 5 min and then submitted to 11% SDS-polyacrylamide gel electrophoresis (SDSPAGE) according to the method of Laemmli (1970) followed by Western blot according to Gierschik et al. (1985) using G protein antisera raised against synthetic decapeptides analogous to various sequences of the corresponding G subunits and an [“‘I]anti-1gG as second antibody (lo5 cpm/ml). The antisera were provided to us by Dr. Allen M. Spiegel of NIH, and their specificity and cross-reactivity have been verified by him (Spiegel, 1990). In our experimental conditions, although some antisera reacted weakly with unidentified proteins, the strongest reactions were always obtained with the corresponding G protein subunits as identified by their electrophoretic mobility on SDS-PAGE using a cholate extract of brain tissue as standard. The autoradiographs (XAR Kodak film with lightning plus intensifying screens, DuPont) were exposed for 24 h at - 80 o C. Autoradiographs were then scanned with an LKB densitometer, in order to obtain a semi-quantitative evaluation of G protein levels. Optical density of the immunoblots was proportional to the amount of proteins applied to the gel within a range of 25-100 pg, as shown in Fig. 1 for (~0, as47 and as42. All values were expressed as percent of normal SpragueDawley pituitary levels. Interassay variability of the quantitation method varied depending on the G protein assayed but never exceeded 20% when evaluated by comparing densitometry data of the anterior pituitary pool with a cholate extract of brain tissue, routinely included in each assay. G protein amounts were also determined in anterior pituitaries of Buffalo and Wistar-Furth rats bearing the 7315a and MtTWl5 tumor respectively, using anterior pituitaries of normal Buffalo and Wistar-Furth rats respectively as controls.

36

Fig. 1. Linear correlation between increasing amounts of normal rat anterior pituitary membrane protein and the density of immunoblot bands of (YO(a), as47 (A) and as42 (m) expressed in arbitrary units. Correlation coefficients were > 0.98 in all cases.

Cell culture und hormone assuys. Normal anterior pituitaries, adenomas, 7315a and MtTW15 tumors were removed and the cells dispersed enzymatically with collagenase 0.35% (Wort~ngton, Mississauga, Ontario, Canada) and hyaluronidase 0.1% (Sigma Chemical Co., St Louis, MO, U.S.A.) during 3 h, and with pancreatin 0.25% (Gibco) for 20 min, according to Vale et al. (1972). Cells were then purified with a Percoll gradient O-90%, as previously described (Schumacher et al., 1978), and cultured in Dulbecco’s modified Eagle medium + 12.50/o serum (horse/calf, 10 : 2.5) at a density of 300,000 cells/ml for adenoma and MtTWI.5 tumors, and 600,000 cells/ml for 7315a tumors in multiple-well Flow dishes. After a 4-day culture period in a humidified atmosphere of 95% air and 5% CO,, cells were washed twice with sucrose 0.32 M and then incubated in the presence of various concentrations of either DA, VIP or DA + VIP during a 30 min period for evaluation of PRL secretion, and during a 15 min period for evaluation of cell CAMP accumulation in the presence of 3-isobutyl-l-methyl xanthine (MIX), an inhibitor of the enzyme phosphodiesterase (Aldrich Chemical Company, Milwaukee, WI, U.S.A.). At the end of the incubation period the supernatant was aspirated and stored at -20°C until assayed for PRL. For evaluation of CAMP accumulation, phosphate buffer was added to wells and the cells were sonicated, the supernatant was removed and kept at - 20°C until assayed. All

PRL and CAMP assays were done in duplicate on samples from sextuplicate wells. PRL was measured by RIA using reagents supplied by NIADDK and the results expressed in terms of rat PRL RP-2 standard. CAMP was measured by specific RIA using a commercial kit. DA was dissolved in ascorbic acid 0.1% and VIP was diluted in incubation medium. ACTH and estradiol were measured by RIA using commercial kits, while GH was kindly assayed by RIA for us by Dr. Gloria Tannenbaum of McGill University using reagents provided by NIADDK. ~s~~~~io~ of RNA and northern blot analysis. Total RNA was prepared by the guanidine thiocyanate cesium chloride method (Chirgwin et al., 1979) from the dissected tissues. RNA was quantified by measuring its absorbance at 260 nm. Denatured samples were electrophoresed through a formaldehyde containing 1.2% agarose gel and transferred to a nylon membrane (Hybond-N, Amersham, Oakville, Ontario, Canada). Membranes were prehybridized at 42°C for 1 h in hyb~dization solution (50% formamide, 10% dextran sulfate, 5 x SSC (1 x SSC = 150 mM NaCl, 15 mM sodium citrate), 1.0% SDS, and 250 pg/ml denatured salmon sperm DNA). Rat (YSand cyo cDNAs were kindly provided by Dr. R.R. Reed (Jones and Reed, 1987) and by Dr. H. Itoh (Itoh et al., 1988) respectively. Excised inserts, 1.1 kb EcoRI fragment for (YS and 1.6 kb SmaI/EcoRI fragment for (~0, were labelled with [cY-“P]~CTP (3000 Ci/mmol, ICN), using random primer labeling kit from Boehringer-Mannheim to a specific activity of about 2 x 10” dpm/pg. Hybridization was done at 42” C with a heat denatured probe at a concentration of 0.5-1.0 ng/ml for 18-24 h in hybridization solution. The membranes were then rinsed twice for 10 min at room temperature in 2 x SSC, 0.1% SDS, once for 15 min at 65’C in 1 x SSC, 0.1% SDS and for 15 min at 65OC in 0.1 x SSC, 0.1% SDS. Autoradiography was performed at - 70” C with an X-ray film (XAR Kodak) and an intensifying screen. Protein determinution. Protein was determined as described by Lowry et al. (1951), with bovine serum albumin as standard. ~t~ti~t~cs. Data were statistically evaluated by one-way analysis of variance and the Duncan multiple range test, or by Student t-test. Due to the

20% interassay variability of the semiquantitative G protein assay, only changes exceeding 20% were taken under consideration for statistical analyses. Results G proteins in various pituitaty tissues Fig. 2 shows representative immunoblots performed with membranes from four pituitary tissues and done with specific antisera raised against while the quantitative synthetic decapeptides, evaluation of all immunoblots is represented in Table 1. These results confirm the presence of very low amounts of cyo in the two DA-resistant tumors. They also indicate that the tumors harbor additional anomalies in G proteins, since in the MtTW15 the levels of ail, ai and ai were decreased and those of (~~42 and as47 were increased, while in the 7315a the levels of ai2, (~~42 and p were decreased. In the adenoma, only the levels of (ui3 were found to be significantly decreased. Table 2 shows the plasma levels of PRL, GH, ACTH and estradiol in various groups of rats either control or bearing the adenoma or either one of the two tumors. As expected. PRL levels were very high in rats with 7315a and MtTW15 tumors, and even more in rats carrying the adenoma. GH levels were greatly increased in rats with the MtTW15 tumor. Circulating levels of estradiol were very high in adenomatous rats since this tumor is induced by the implantation of estrone pellets. ACTH levels were not significantly modified in any of the experimental groups.

TABLE

Fig. 2. Representative immunoblots obtained from membranes (50 pg) of normal rat anterior pituitaries (AP), estrogen-induced pituitary adenoma (PA). 7315a (TA) and MtTWl5 (TW) transplantable tumors. lmmunoblots were done with antiserum Go/l for (~0. LD for ail, LE for ai2. SQ/2 for ai3, RM for us and SW for p.

1

QUANTITATION TUMORS

OF G PROTEIN

SUBUNITS

IN MEMBRANES

FROM

PITUITARY

ADENOMA

AND

DA-RESISTANT

Quantitation was done by densitometry of immunoblots obtained with antisera Go/l for ao. LD for ail. LE for ai2. SQ/2 RM for as and SW for fl. Values are mean * SEM obtained from 4-6 immunoblots performed for each G protein. Tissue

ao (%)

ail (%)

ai

Adenoma 7315a MtTW15

103+6 15*3 19*4

94*14 125+13 37* 9**

99*13 35*10 ** 41* 7**

**

p < 0.01 vs. normal

** ** anterior

pituitary

(S)

values arbitrarely

ai

(%)

31* 5 ** 73* 9 43k16 ** set at 1008,

cxs42 (S)

ns47 (%)

P (%)

77* 9 27+ 4** 239+33 **

7g* 3 115* 5 151*11**

80f4 64*5 114k6

by Duncan

multiple

range

test.

for ai3.

**

38 TABLE

2

PLASMA HORMONE VALUES OF RATS CARRYING ADENOMA OR DA-RESISTANT TUMORS Rats

Control

PRL

GH

ACTH

Estradiol

k/ml)

(Wml)

QwW

(pg/mU

92+12 96529 117*35 117+24

29+ 9 283&71 + 14+ 7 142 2

lo& 2 26& 9 2349rt 1 346 * 49 & 0.1 38F 16 203+23* 684* 779+144* 3856-+

Adenoma 7315a MtTWIS

* p < 0.01 vs. control

values by Student

performed with membrane preparations obtained from anterior pituitaries of rats bearing the transplantable tumors. The results of the quantitative evaluations are shown in Table 3. Contrary to results obtained with tumoral tissues, the levels of LYO,cxil and oi3 were increased while those of (~~42 and j? were practically unchanged in pituitaries of rats harboring the ‘7315a tumor. In pituitaries of rats transplanted with the MtTW15 tumor the levels of (us42 were significantly decreased, while those of all the other G proteins were not significantly modified.

AN

f-test.

In order to verify whether high levels of circulating hormones might be responsible for the extensive modifications in G protein amounts found in tumoral tissues, immunoblots were also TABLE

GOLXand Gscv mRNA Northern analysis of total rat pituitary, adenoma, the

RNA from normal two transplantable

3

G PROTEIN

AMOUNTS

IN ANTERIOR

PITUITARIES

OF RATS CARRYING

DA-RESISTANT

TUMORS

Quantitation was performed by densitometry of immunobiots obtained with antisera Go/l for do, LD for ail. LE for ni2, SQ/2 ai3, RM for o(s and SW for /?. Values are meankSEM obtained from 3-4 immunoblots performed for each G protein. Tissue

eo (8)

7315a MtTW15

13518 99+8

evil (%) 237+18 98+29

*

ai **

(%)

116125 831 7

ai

(%)

4435150 * 107i 3

* p r: 0.05, * * p < 0.01 vs. values in anterior pituitaries of Buffalo or Wistar-Furth MtTW15 tumor carrying rats respectively, by Student t-test.

0~42 (%)

as47 (%)

P (%)

89kll 36*

121*11 68+1J

137514 81&13

control

5**

rats arbitrarely

for

set at 100% for 7315a and

B

A

Go, mRNA

Gs, mRNA

428s

48s

i

AP

PA

TA

TW

AI=

?A

TA

TW

oligonucleotide probes directed to the a-subunits of Go (A) and Fig, 3. Northern analysis of RNA hybridized with 32P-radiolabelled Gs (E). Total RNA (10 pg) extracted from rat liver (L), normal anterior pituitary (AP), the adenoma (PA), the 7315a (TA) and the MtTW15 (TW) tumors, was resolved on denaturing agarose gels.

39

90 -

60 -

70 -

o-S.O. O---

0 -S.D. o--AUXW

MEWJMA A”“‘73W qutTN15

A **.* 7315s II----MtlnI5

60 -

60 -

T 04’

I -9

-6

-7

DA (log

T

-6

M)

DA (log

M)

Fig. 4. Representative dose-response curves of DA inhibition of PRL secretion and CAMP accumulation by cells various tissues. The experiments were repeated 3-6 times with similar results. SD = anterior pituitaries obtained from ley rats. Baseline values of PRL (ng/ml) were (mean+SEM): SD, 223+45; adenoma, 134+17 *; 7315a, 22k3 34+ 3 * *. Baseline values of CAMP (pmol/mg of protein) were (mean& SEM): SD, 30.0 + 2.5; adenoma, 17.5 + 7.0rt0.3**;MtTW15,17.5~2.0 **. *p

G proteins in normal rat pituitaries and in prolactin-secreting rat pituitary tumors.

It is still undetermined which GTP-binding (G) protein is involved in the regulation of prolactin (PRL) release and through which effector. This study...
1MB Sizes 0 Downloads 0 Views