- Email: [email protected]
Influence of cell cycle phases on the electrical activity and hormone release in a transformed line of anterior pituitary cells
Life Sciences, Vol. 40, pp. 2377-2384Printed in the U.S.A.
Pergamon Journals
INFLUENCE OF CELL CYCLE PHASES ON THE ELECTRICAL ACTIVITY AND HORMONE RELEASE IN A TRANSFORMED LINE OF ANTERIOR PITUITARY CELLS Mireille Denechaud1 , Jean-Marc Israel 1 , Zohalr Mishal 2, Jean-Dtdter Vincent 1 11NSERM Unit6 176, Rue Cam=lie Salnt-Sa~ns 33077 Bordeaux Cedex France. 2Service de cytofluonm6trle 7 rue G. Moquet 94800 ~TiIlejutf.
(Received in final form April I, 1987) SummaP~ Electrophysiological experiments have shown that about 50% of cultured GH3 cells (tumoral cell hne, from the anterior pituitary gland) are inexcltable l e. they do not display action potentials either spontaneously or when depolarized by a current pulse We report here this inexcltabthty may be related to cellular kmetms Thus we have studied the relationship between the various phases of the cell cycle, the electrophyslological properties of GH3/B6 cells and spontaneous or induced Prolactm and Growth Hormone (GH) release rates. Asynchronous populations of viable cells were stained wsth Hoechst 33 342 DNA fluorescent dye, and sorted using a flow cytometer into G1 and S phases After selection intracellular potentials were recorded using a single glass micro-electrode ; the basal or TRH stimulated rates of PRL and GH secretLons were determined by RIA Electrical properties of the cells l e resting potentials, input membrane resistance and excitability, reached a maximum for cells in G2+M phases Only cells m G2+M displayed action potentials and TRH increased their secret=on by 5 times for GH and by 6 times for PRL In G1 and S phases the cells were electrically inactive and secretion rates remained at their basal levels. These findings demonstrate that the mechanism of stimulus secretion couphng is dependent upon the phases of the cell cycle. The GH3/B6 is a subclone of the GH3 cell line =solated from a rat anterior pituitary tumor which has been shown to secrete growth hormone (GH) and prolectln (PRL), (1-2). The GH3/B6 cells have also been shown to be electrically excitable and able to generate spontaneous or evoked actson potent=als which have a prominent calcium component (3-4) Since a factor, such as thyreohberlne (TRH), that influences hormone release from GH3/B6 cells, also has an effect on the calcium dependent electrical activity of these cells, it has been suggested that the electrical activity may provide a control of calcium entry and act at least as a hormone release modulator. Paradomcally, not all of the clonal cell populahon is excitable . less than half of the total clonal cell population is excitable It has been previously suggested that th=s could be due to membrane damage by the electrode or to the use of non appropriate culture conditions (4) We now suggest that another parameter l e. cells kinetics, may be implicated =n this discrepancy We have studied the possible relationship between the different phases of the cell cycle, the spontaneous and evoked PRL and GH release and the electrophyslology properties of the GH3/B6 cells Matenals and Methods 1) Cell line and culture conditions GH3/B6 cells were currently grown as a monolayer in plastic tissue culture flasks contammg HAM F10 medium supplemented with 12.5% inactivated horse serum and 2 5% foetal calf serum and maintained at 37 ° C. Cells were selected at their gth or 11th passage
0024-3205/87 $3.00 + .00 Copyright (c) 1987 Pergamon Journals Ltd.
2378
Cell Cycles and Hormone Release in Pituitary
Vol. 40, No. 24, 1987
2) Cell sortrno Six days after the last passage, the cells were trypsinised (30 sec - 0 05% trypsin at room temperature), and gently mechanically dispersed with a roden Pasteur pipette. Cell suspensmons were incubated for 60rain (37°C) in Phosphate Buffer Saline (PBS) med=um containing t p.g/ml of the DNA fluorescent dye (bisbenzlmldazole Hoechst 33342) This dye brads both specifically and quantitatively to DNA and once bound becomes strongly fluorescent (excitation wavelength, 350 nm). It is taken up by living cells and is non toxic (5). Viable cells were sorted according to their DNA content using a flow cytometer (EPIVS V, Coulter Electronics Hlaleah, FIonde) equipped with a 300mW Argon Laser (Spectra physics Mountain Vtew, Cahfornia) Cells were selected m Gf post-mrtotic phase and Jn DNA synthesm phase (Fig 1) ; 106 cells of each subpopulation were grown m culture medrum at 37°C for at least 12 hrs, before the next experiment. The doubling time was 16 hrs and after 4 doubling ttmes, we considered that the cells had become asynchronous 3) Electroohvstoloqrcal exoeriments The dish containing the cells was placed Jn a chamber fixed on the stage of an inverted microscope (Nikon TM2). Just before the recording experiments, the culture medium was replaced by a recording medium of the following compositron in mM/I : KCI, 5.6 ; NaCI, 156 ; CaCI 2, 2 , Hepes buffer, 5 ; glucose, 10 ; pH, 7 4. A single mlcroelectrode bndge amplifier (MENTOR) was used to record the potential of the cell and to inject transmembrane current. Input membrane resmtance was evaluated by passing rather depolarizing or hyperpolarizing currents and correlated wuth the reduced potentral deflections, according the Ohm law. Microelectrodes were pulled from cahbrated tubes (clark) with an horizontal pulled (Narishige, PN3) and have a tip diameter less than 0.2 ~m They were filled with CH3COOK (4M) and had a reststance ranging from 80 to 120 M.Q. During the expenment, the medium temperature was kept at 36 ° C by a warm atr system. 4) Rad~mmmunoassav of PRL and GH Before the experiments the cells were washed wtth HAM F10 at 37°C (without sera)and then covered with 2ml of medium with or without thyreoltberme (TRH) at 10"6M, for 30 mm at 37°C. The amounts of PRL and GH secreted in the medium were determined by radioimmunoassay using reagents provided by NIAMDD. Results are expressed m ng/ml/106 cells, basal and stimulated release are compared in the same experimental condttions Results 1) Kmetms analvsm of the cell ooDufatJon GH3/B6 cells were studied during their exponential growing period The number of cells in the various phases of the cell cycle was estimated by analysing their d=strrbutron in G1 (post-mitotic state) , S (DNA synthesis) ; G2 (pre-mltotic state) and M (mitosis). It was thus possible to calculate the duration of each one of these phases. The cycle was estimated to be 16h in duration, with G1 11h00, S 3h30, G2 + M : lh30. Cells were sorted into G1 and S phases (Fig. 1) The intensity of cellular fluorescence is proportional to the DNA content of the cell, and the intensity of scattered light is proportional to cell size. As a cell goes through different phases of the cell cycle, its DNA content increases as well as its size (Fig 2) This correlates the values obtained by measuring the apparent diameter (E)) of the cells under microscopic visuahsation (E~ G1- 3 5 + 0 4 Fro, n = 58 , £) S~ 7 t ± 0 3 p.m, n = 4 3 , ~ G 2 + M - 11 4 ± 16p.m, n = 30) 2) Electroohvstoloalcal
orooertleS
The electrophysrologmal parameters studied were resting potential, input membrane resistance and excitability, i.e. the ability of the cells to generate actron potentmls. - M e a n Resting p o t e n t i a l (Table I , Fig 3) varied from -16mV (cells in the G1 phase) to -30mV (S phase) and reached -50mV (G2+M phase). For each experiment, the resting potential was measured 10 to 20 seconds after impalement so as to ensure a stable recording - M e a n I n p u t m e m b r a n e resistances (Table I ; Fig. 4) also increased from G1 (30MQ) to S (75M~) and G2+M (152M~)
Vol. 40, No. 24, 1987
Cell Cycles and Hormone Release in Pituitary
3OOO
°.
2500
2000
•
°
•
•
•
°
~6 15OO
•I
i
"1 •1
!
II
t
• IG L • I I" I" I"
~OOO
500
I''TI S fI G'2\+ I i ~ .
;
I I
. . . . . . . "" 50
,I
t .
I00 ~
sorhng
.......... "
I 150
~"-
200
w~ndows
,
ii-
250
Fluorescence
channel
number
FIG. 1 DNA dlstrlbuhon for an asynchronous exponentially growmg GH3/B6 cells population A cell enters the G1 phase upon division Phase durations are deduced from the number of cells in each phase Post-mitotic phase G1 - 11h00, DNA synthesis S : 3h30, pre-mltobc phase G2+M . 1h30
300 "~ 2 5 0 E
~
200
~
150
....'....-
E
o
I00
2
5O
L
%
i
i
i
50
I00
150
i
200
i
250
S c a t t e r Ilgth channel number
FIG. 2 Relation between cellular DNA content (measured and cell size (measured as scattered light)
as
dye
fluorescence)
2379
2380
Cell Cycles and Hormone Release in Pituitary
Vol. 40, No. 24, 1987
TABLE I Electrophyslologmal parameters mean values (:t: S D) related to the cell cycle phases. Cellular cycle (number of cells recorded) Resting Potentials (mY)
G1 (42)
S (37)
G2+M (64)
-16 (+2)
-30 (:1:2)
-50 (:1:6)
30 (+7)
75 (+15)
152 (±30)
Input membrane Resistance ( M ~ )
-60
-o o
-50
i..
_-F
2O0.
.Q
E v
-40.
u
0 ~
x
LU
i-
,uv'=~,
,/
c
2 _~
-30_
Ioo
o I.-
,f
E o
~
-20 _
5c
I
GI 5
~,--..,,-_S,,..~_2_~M I0
FIG 3
15
....... Hours
_GI. . . . . . . .
S ~2.~ M
[
i
i
5
I0
15
FIG 4
Resting potentJals O(FJg 3) Input membrane resistance i ( F I g measured in vanous phases of the cell cycle
4) mean values
Hours
Vol. 40, No. 24, 1987
Cell Cycles and Hormone Release in Pituitary
2381
- Rechhcation properties of the membrane (Fig. 5) : The input resistance of the membrane measured at different potentials, was quite linear for G1 and S phase cells. However, a decrease of the input membrane resistance correspondmg to the delayed rectificatton was observed for the G2+M phase cell
+1
jS
~
-100 mV
-70
I
~
l
-30
+ 0.5 (% E
÷ 20
;o
A
'
~
'
/
'
4
7
Membrane
potenha~ ]>
-C~5
/-
-1
FIG. 5 Polarizing current intensity versus membrane potential values for cells in G1 (&), S ( i ) , G2+M ( 0 ) phases
-Acttve properties (Ftg 6) varied with the cell cycle phases . no electrical acttvlty was recorded from cells in G1 and S phases, i.e it was impossible to trigger action potentials by depolarizing current pulses, even after the cells were hyperpolarlzed by current Only cells in G2+M phase were able to trigger action potentials under depolarizing current or after the cessation of hyperpolarizmg current pulses
20mV S
b) . . . .
, L__l
I
!
r
L__Jo.InA 2sec
, -120mY G2*M b)
,
,~_L_J.~
I-'~.
2sec 150msec
r - -
1--
I O.InA
2sec
FIG. 6 Electrical activities show that cells in the G2+M phase are able to trigger action potentials (arrow heads)
2382
Cell Cycles and Hormone Release in Pituitary
Vol. 40, No. 24, 1987
3) Prolactln and GH release The basal or TRH stimulated rates of PRL and GH release are expressed m Table II (in ng/ml/106 cells). Basal rates for PRL (F=g 7~. and GH (Fig. 8) are not slgmficantly different m the various phases of the cell cycle (~ 100 ng/ml/10b cells)
O
o
500
G H
PRL
C
Asynchr
.6"
30C G1
S
G2
Asy~
40( 30O
G1
i i
S
20O 10C
10£ Ct
TRH
ct TRH
Ct
TRH
ct
mH
rhn Hn ct TRH
FIG 7
ct TRH
0iTR.
ct
FIG 8
Basal r a t e [ ] and TRH sbmulated[-] release of prolactm for synchronous cells (m G1, S and G2+M) and for asynchronous cells
Basal ratel~]and TRH st=mulatedl--] release of growth hormone for synchronous cells On G1, S and G2+M) and for asynchronous cells
Cells do not respond srgmflcantly to TRH dunng G1 and S phases However, considenng the cells ~n G2+M phase the release was increased by 6 times for PRL and by 5 times for GH In the asynchronous populabon, stimulated secretion rates were about twine the basal release for GH and reached about 30% from the basal release for prolactm. DlSOUSSfon In agreement with the previous results obtained with GH3, our results have shown that the responsiveness to TRH may be closely related to the different cell cycle phases such that TRH stimulated secretion reached 5 to 6 times the basal release in G2+M phase In his rev=ew, Pawhkowsky (6) indicated that secretory function and cell proliferation m the endocrine glands are closely related and more precisely, this author indicated that release of hormone is a Ca-dependent process, preceeded by cell membrane depolansation and Ca-reflux , mftosms has also been shown to be a Ca-dependent process mitotic actwity m the bone marrow and thymus was induced by the mjectron of calcium chloride (7). So, this suggests that the link between secreUon and mrtosJs in the endocrine gland cells may be mediated by the Ca 2+ influx but the interrelations between the mechanisms of these observed phenomena are at present poorly understood.
TRH
Vol. 40, No. 24, 1987
Cell Cycles and Hormone Release in Pituitary
2383
Among the first studies deahng with the relatson between hormone hberatlon and the different phases of the cell cycle, Varga (8) observed that m melanoma cells the biological response to MSH was only seen m the G2+M phase. Clausen (9) studying the vanatlons in hormone production m GH3 cells observed that the presence of TRH m the culture medium for 3 days, induce an increase in PRL production the percentage of cells in each phase of the cell cycle was unchanged yet. The author suggested that cells m various phases of the cell cycle had differing ablhtsss to synthemze and secrete prolactm Several authors have studied the variation of mtracellular PRL concentration in relation to the phases of the cell cycle. Using the techmque of flow microfluorometry, Kiino (10) showed that prolactm storage increased as cells passed from G1 to M and he confirmed Clausen's results by observing that estradlol increased intracellular PRL three to sevenfold, but did not affect the d~strLbution of cells m each phase of the cell cycle The above results d~ffer from those of Fmvre Bauman et al (11) who showed that TRH-induced release reached a maximum in the G1 state. This dmcrepancy may be due to the fact that Falvre Bauman et at. used serum deprivation to synchromse the cells and this may alter the normal functson of the cells Surks (12) described a slgmficant increase in cellular DNA and cellular as well as secreted GH, in S phase cultures of a rat pituitary cell hne that produces GH He concluded that GH productmn rate observed m asynchronous cultures may be an integrated value for all phases of the cell cycle Concerning the electrophyslologlcal properties of pstuitary cells, Duff/ et al (2) showed that sp=ke acttvity could not be triggered by current pulse rejection when the resting potential was less than -40 mV These authors proposed that the low values of membrane potential and the inexmtabdlty observed for more than 50 % of clonal cell population were not necessarily due to a cells damage occurmg after empalement Our results suggest that cell kmetms may influence the electrical properties of secretory cells in that only cells m phases G2 and M are exmtable In our expenments, cells m these two phases represented 15 % of the total populatmn The observation that Duff/ et al. (2) found a greater percentage of exmtable cells (about 50 %) may be due to an experimental b=as m that it m easier to record from large cells and we have shown that large cells are in the end phase of the cell cycle (= e G2 and M). However, the statmtlcal factor for evaluating the sub-population of the cell cycle may cause us to underestimate our percentage of exc=table cells by not taking into account those cells wNch are at the end of the S phase and, by their s0ze, may be included in the G2+M category Boonstra et al (13) observed that membrane potential values and K + activity from neuroblastoma clonal cells were high in mitosis, decreased in G1 phase and rose again during S phase Membrane potential value vaned from -20 mV (G1) to -50 mV (G2 - M) This is in agreement with our observations which showed the decrement of the relative value of resting potential from -16 to -50 mV. We have observed an increase of the membrane resmtance from 30 to 152 MQ from the G1 phase to the G2+M phase. However, if the resistance is reported to the cell surface unit, we found a relative decrease of 1 8 between the specific resistance in Gt and G2+M phases Thin agree with Boonstra's observations (13) whmh reported a decrease of about 1 7. The hyperpolanzat~on of the resting membrane potential has to be related to a specific increase of the membrane conductance for potassium =ons, because it's value depends on the extracellular potassium concentration (results not shown). This is necessary for tnggenng action potentials it has been shown that reward currents (sodium and calcium) for GH3 cells are reactivated at membrane potentials less than -40mV (14) However, the rectification observed for depolanzmg potentials w=th cells in G2+M phase =s due to a specific increase of the potasmum currents (15). In conclusion, our results have shown that GH3/B6 cells were excitable only m the G2+M state, and that TRH was able to s~gmficantly increase growth hormone and protactln hberatlons only in the G2+M phase. These finding are in accordance with the"stimulus-secretion" couphng proposed by Douglas (1968) but it implies that this couphng is dependent of the phases of the cell cycle.
2384
Cell Cycles and Hormone Release in Pitultary
Vol. 40, No. 24, 1987
p.cknowledoments The authors gratefully acknowledge the sktlful asststance of Mrs D Verrier and Mrs J Arsaut m maintaining the cell hne and for the RIA and Miss A. Ferret-Bouin. We would like to thank Professor Mawas (Facult6 de Marseille) and Professor Rosenfeld (CNRS, Villejulf) for allowmg us to use their cell sorters The dye Hoechst 33342 was a gtft from the Riedel de Hahn laboratory The RIA ktt was a gift from Drs Parlow and Raitl of the NIAMDD We are grateful to Mr S. Vltelho who labelled the growth hormone w0th 1251
References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Y. KIDOKORO, Nature 258. 741-743 (1975) B DUFY, J D. VINCENT, H FLEURY, P. DU PASQUIER, D. GOURDJI, A TlXlER-VlDAL, Sctence 204. 509-511 (1979) WW DOUGLAS, Br J Pharmacol 34, 451-474 (1968) B DUFY and J D. VINCENT, Serono Svmooslum N°49. p.89-99, Academtc Press, London (1982) D J ARNDT-JOUIN and T.M. JOUIN, Htstochem-Cytochem 25, 585-589 (1977). M PAWLIKOWSKI, Life Sctences 30, 315-320, 1982 G.R SMITH, M.L. GURSON, AJ KIDDEL, A.D PERRIS, J, Endocrinology 65. 45-53 (1975) J M VARGA, A DIPASQUALE, J. PAWELECK, J.S MC GUIRE, A.B LERNER, Proc Nat Acad Sct. USA, .L1.., 1590-1593 (1974) O P F CLAUSEN, K M GAUTVlK, E HAUG, J. Cell Phystol. 94, 205-214 (1978). D R. KIINO, D E BURGER, P S DANNIES, J. of Cell Btol. 93, 459-462 (1982). A FAIVRE-BAUMAN, D GOURDJI, D GROUSELLE, A TlXlER-VlDAL, Biochem. and BIophys Res Comm. . ~ 50-67, 1975. I SURKS, Endocrinology 114, 873-87g (1984) J BOONSTRA, C.L MUMMERY, L.G.J. TERTOOLEN, P Van Der SAAG, S.W De LAAT, J of Cell Physiol 107, 75-83 (1981). D R MATTESON and C M ARMSTRONG, J. Gen PhysJol 83. 371-394 (1984) J M DUBINSKY and G.S. OXFORD, J. Gen. Physiol. 83. 309-339 (1984).
Pergamon Journals
INFLUENCE OF CELL CYCLE PHASES ON THE ELECTRICAL ACTIVITY AND HORMONE RELEASE IN A TRANSFORMED LINE OF ANTERIOR PITUITARY CELLS Mireille Denechaud1 , Jean-Marc Israel 1 , Zohalr Mishal 2, Jean-Dtdter Vincent 1 11NSERM Unit6 176, Rue Cam=lie Salnt-Sa~ns 33077 Bordeaux Cedex France. 2Service de cytofluonm6trle 7 rue G. Moquet 94800 ~TiIlejutf.
(Received in final form April I, 1987) SummaP~ Electrophysiological experiments have shown that about 50% of cultured GH3 cells (tumoral cell hne, from the anterior pituitary gland) are inexcltable l e. they do not display action potentials either spontaneously or when depolarized by a current pulse We report here this inexcltabthty may be related to cellular kmetms Thus we have studied the relationship between the various phases of the cell cycle, the electrophyslological properties of GH3/B6 cells and spontaneous or induced Prolactm and Growth Hormone (GH) release rates. Asynchronous populations of viable cells were stained wsth Hoechst 33 342 DNA fluorescent dye, and sorted using a flow cytometer into G1 and S phases After selection intracellular potentials were recorded using a single glass micro-electrode ; the basal or TRH stimulated rates of PRL and GH secretLons were determined by RIA Electrical properties of the cells l e resting potentials, input membrane resistance and excitability, reached a maximum for cells in G2+M phases Only cells m G2+M displayed action potentials and TRH increased their secret=on by 5 times for GH and by 6 times for PRL In G1 and S phases the cells were electrically inactive and secretion rates remained at their basal levels. These findings demonstrate that the mechanism of stimulus secretion couphng is dependent upon the phases of the cell cycle. The GH3/B6 is a subclone of the GH3 cell line =solated from a rat anterior pituitary tumor which has been shown to secrete growth hormone (GH) and prolectln (PRL), (1-2). The GH3/B6 cells have also been shown to be electrically excitable and able to generate spontaneous or evoked actson potent=als which have a prominent calcium component (3-4) Since a factor, such as thyreohberlne (TRH), that influences hormone release from GH3/B6 cells, also has an effect on the calcium dependent electrical activity of these cells, it has been suggested that the electrical activity may provide a control of calcium entry and act at least as a hormone release modulator. Paradomcally, not all of the clonal cell populahon is excitable . less than half of the total clonal cell population is excitable It has been previously suggested that th=s could be due to membrane damage by the electrode or to the use of non appropriate culture conditions (4) We now suggest that another parameter l e. cells kinetics, may be implicated =n this discrepancy We have studied the possible relationship between the different phases of the cell cycle, the spontaneous and evoked PRL and GH release and the electrophyslology properties of the GH3/B6 cells Matenals and Methods 1) Cell line and culture conditions GH3/B6 cells were currently grown as a monolayer in plastic tissue culture flasks contammg HAM F10 medium supplemented with 12.5% inactivated horse serum and 2 5% foetal calf serum and maintained at 37 ° C. Cells were selected at their gth or 11th passage
0024-3205/87 $3.00 + .00 Copyright (c) 1987 Pergamon Journals Ltd.
2378
Cell Cycles and Hormone Release in Pituitary
Vol. 40, No. 24, 1987
2) Cell sortrno Six days after the last passage, the cells were trypsinised (30 sec - 0 05% trypsin at room temperature), and gently mechanically dispersed with a roden Pasteur pipette. Cell suspensmons were incubated for 60rain (37°C) in Phosphate Buffer Saline (PBS) med=um containing t p.g/ml of the DNA fluorescent dye (bisbenzlmldazole Hoechst 33342) This dye brads both specifically and quantitatively to DNA and once bound becomes strongly fluorescent (excitation wavelength, 350 nm). It is taken up by living cells and is non toxic (5). Viable cells were sorted according to their DNA content using a flow cytometer (EPIVS V, Coulter Electronics Hlaleah, FIonde) equipped with a 300mW Argon Laser (Spectra physics Mountain Vtew, Cahfornia) Cells were selected m Gf post-mrtotic phase and Jn DNA synthesm phase (Fig 1) ; 106 cells of each subpopulation were grown m culture medrum at 37°C for at least 12 hrs, before the next experiment. The doubling time was 16 hrs and after 4 doubling ttmes, we considered that the cells had become asynchronous 3) Electroohvstoloqrcal exoeriments The dish containing the cells was placed Jn a chamber fixed on the stage of an inverted microscope (Nikon TM2). Just before the recording experiments, the culture medium was replaced by a recording medium of the following compositron in mM/I : KCI, 5.6 ; NaCI, 156 ; CaCI 2, 2 , Hepes buffer, 5 ; glucose, 10 ; pH, 7 4. A single mlcroelectrode bndge amplifier (MENTOR) was used to record the potential of the cell and to inject transmembrane current. Input membrane resmtance was evaluated by passing rather depolarizing or hyperpolarizing currents and correlated wuth the reduced potentral deflections, according the Ohm law. Microelectrodes were pulled from cahbrated tubes (clark) with an horizontal pulled (Narishige, PN3) and have a tip diameter less than 0.2 ~m They were filled with CH3COOK (4M) and had a reststance ranging from 80 to 120 M.Q. During the expenment, the medium temperature was kept at 36 ° C by a warm atr system. 4) Rad~mmmunoassav of PRL and GH Before the experiments the cells were washed wtth HAM F10 at 37°C (without sera)and then covered with 2ml of medium with or without thyreoltberme (TRH) at 10"6M, for 30 mm at 37°C. The amounts of PRL and GH secreted in the medium were determined by radioimmunoassay using reagents provided by NIAMDD. Results are expressed m ng/ml/106 cells, basal and stimulated release are compared in the same experimental condttions Results 1) Kmetms analvsm of the cell ooDufatJon GH3/B6 cells were studied during their exponential growing period The number of cells in the various phases of the cell cycle was estimated by analysing their d=strrbutron in G1 (post-mitotic state) , S (DNA synthesis) ; G2 (pre-mltotic state) and M (mitosis). It was thus possible to calculate the duration of each one of these phases. The cycle was estimated to be 16h in duration, with G1 11h00, S 3h30, G2 + M : lh30. Cells were sorted into G1 and S phases (Fig. 1) The intensity of cellular fluorescence is proportional to the DNA content of the cell, and the intensity of scattered light is proportional to cell size. As a cell goes through different phases of the cell cycle, its DNA content increases as well as its size (Fig 2) This correlates the values obtained by measuring the apparent diameter (E)) of the cells under microscopic visuahsation (E~ G1- 3 5 + 0 4 Fro, n = 58 , £) S~ 7 t ± 0 3 p.m, n = 4 3 , ~ G 2 + M - 11 4 ± 16p.m, n = 30) 2) Electroohvstoloalcal
orooertleS
The electrophysrologmal parameters studied were resting potential, input membrane resistance and excitability, i.e. the ability of the cells to generate actron potentmls. - M e a n Resting p o t e n t i a l (Table I , Fig 3) varied from -16mV (cells in the G1 phase) to -30mV (S phase) and reached -50mV (G2+M phase). For each experiment, the resting potential was measured 10 to 20 seconds after impalement so as to ensure a stable recording - M e a n I n p u t m e m b r a n e resistances (Table I ; Fig. 4) also increased from G1 (30MQ) to S (75M~) and G2+M (152M~)
Vol. 40, No. 24, 1987
Cell Cycles and Hormone Release in Pituitary
3OOO
°.
2500
2000
•
°
•
•
•
°
~6 15OO
•I
i
"1 •1
!
II
t
• IG L • I I" I" I"
~OOO
500
I''TI S fI G'2\+ I i ~ .
;
I I
. . . . . . . "" 50
,I
t .
I00 ~
sorhng
.......... "
I 150
~"-
200
w~ndows
,
ii-
250
Fluorescence
channel
number
FIG. 1 DNA dlstrlbuhon for an asynchronous exponentially growmg GH3/B6 cells population A cell enters the G1 phase upon division Phase durations are deduced from the number of cells in each phase Post-mitotic phase G1 - 11h00, DNA synthesis S : 3h30, pre-mltobc phase G2+M . 1h30
300 "~ 2 5 0 E
~
200
~
150
....'....-
E
o
I00
2
5O
L
%
i
i
i
50
I00
150
i
200
i
250
S c a t t e r Ilgth channel number
FIG. 2 Relation between cellular DNA content (measured and cell size (measured as scattered light)
as
dye
fluorescence)
2379
2380
Cell Cycles and Hormone Release in Pituitary
Vol. 40, No. 24, 1987
TABLE I Electrophyslologmal parameters mean values (:t: S D) related to the cell cycle phases. Cellular cycle (number of cells recorded) Resting Potentials (mY)
G1 (42)
S (37)
G2+M (64)
-16 (+2)
-30 (:1:2)
-50 (:1:6)
30 (+7)
75 (+15)
152 (±30)
Input membrane Resistance ( M ~ )
-60
-o o
-50
i..
_-F
2O0.
.Q
E v
-40.
u
0 ~
x
LU
i-
,uv'=~,
,/
c
2 _~
-30_
Ioo
o I.-
,f
E o
~
-20 _
5c
I
GI 5
~,--..,,-_S,,..~_2_~M I0
FIG 3
15
....... Hours
_GI. . . . . . . .
S ~2.~ M
[
i
i
5
I0
15
FIG 4
Resting potentJals O(FJg 3) Input membrane resistance i ( F I g measured in vanous phases of the cell cycle
4) mean values
Hours
Vol. 40, No. 24, 1987
Cell Cycles and Hormone Release in Pituitary
2381
- Rechhcation properties of the membrane (Fig. 5) : The input resistance of the membrane measured at different potentials, was quite linear for G1 and S phase cells. However, a decrease of the input membrane resistance correspondmg to the delayed rectificatton was observed for the G2+M phase cell
+1
jS
~
-100 mV
-70
I
~
l
-30
+ 0.5 (% E
÷ 20
;o
A
'
~
'
/
'
4
7
Membrane
potenha~ ]>
-C~5
/-
-1
FIG. 5 Polarizing current intensity versus membrane potential values for cells in G1 (&), S ( i ) , G2+M ( 0 ) phases
-Acttve properties (Ftg 6) varied with the cell cycle phases . no electrical acttvlty was recorded from cells in G1 and S phases, i.e it was impossible to trigger action potentials by depolarizing current pulses, even after the cells were hyperpolarlzed by current Only cells in G2+M phase were able to trigger action potentials under depolarizing current or after the cessation of hyperpolarizmg current pulses
20mV S
b) . . . .
, L__l
I
!
r
L__Jo.InA 2sec
, -120mY G2*M b)
,
,~_L_J.~
I-'~.
2sec 150msec
r - -
1--
I O.InA
2sec
FIG. 6 Electrical activities show that cells in the G2+M phase are able to trigger action potentials (arrow heads)
2382
Cell Cycles and Hormone Release in Pituitary
Vol. 40, No. 24, 1987
3) Prolactln and GH release The basal or TRH stimulated rates of PRL and GH release are expressed m Table II (in ng/ml/106 cells). Basal rates for PRL (F=g 7~. and GH (Fig. 8) are not slgmficantly different m the various phases of the cell cycle (~ 100 ng/ml/10b cells)
O
o
500
G H
PRL
C
Asynchr
.6"
30C G1
S
G2
Asy~
40( 30O
G1
i i
S
20O 10C
10£ Ct
TRH
ct TRH
Ct
TRH
ct
mH
rhn Hn ct TRH
FIG 7
ct TRH
0iTR.
ct
FIG 8
Basal r a t e [ ] and TRH sbmulated[-] release of prolactm for synchronous cells (m G1, S and G2+M) and for asynchronous cells
Basal ratel~]and TRH st=mulatedl--] release of growth hormone for synchronous cells On G1, S and G2+M) and for asynchronous cells
Cells do not respond srgmflcantly to TRH dunng G1 and S phases However, considenng the cells ~n G2+M phase the release was increased by 6 times for PRL and by 5 times for GH In the asynchronous populabon, stimulated secretion rates were about twine the basal release for GH and reached about 30% from the basal release for prolactm. DlSOUSSfon In agreement with the previous results obtained with GH3, our results have shown that the responsiveness to TRH may be closely related to the different cell cycle phases such that TRH stimulated secretion reached 5 to 6 times the basal release in G2+M phase In his rev=ew, Pawhkowsky (6) indicated that secretory function and cell proliferation m the endocrine glands are closely related and more precisely, this author indicated that release of hormone is a Ca-dependent process, preceeded by cell membrane depolansation and Ca-reflux , mftosms has also been shown to be a Ca-dependent process mitotic actwity m the bone marrow and thymus was induced by the mjectron of calcium chloride (7). So, this suggests that the link between secreUon and mrtosJs in the endocrine gland cells may be mediated by the Ca 2+ influx but the interrelations between the mechanisms of these observed phenomena are at present poorly understood.
TRH
Vol. 40, No. 24, 1987
Cell Cycles and Hormone Release in Pituitary
2383
Among the first studies deahng with the relatson between hormone hberatlon and the different phases of the cell cycle, Varga (8) observed that m melanoma cells the biological response to MSH was only seen m the G2+M phase. Clausen (9) studying the vanatlons in hormone production m GH3 cells observed that the presence of TRH m the culture medium for 3 days, induce an increase in PRL production the percentage of cells in each phase of the cell cycle was unchanged yet. The author suggested that cells m various phases of the cell cycle had differing ablhtsss to synthemze and secrete prolactm Several authors have studied the variation of mtracellular PRL concentration in relation to the phases of the cell cycle. Using the techmque of flow microfluorometry, Kiino (10) showed that prolactm storage increased as cells passed from G1 to M and he confirmed Clausen's results by observing that estradlol increased intracellular PRL three to sevenfold, but did not affect the d~strLbution of cells m each phase of the cell cycle The above results d~ffer from those of Fmvre Bauman et al (11) who showed that TRH-induced release reached a maximum in the G1 state. This dmcrepancy may be due to the fact that Falvre Bauman et at. used serum deprivation to synchromse the cells and this may alter the normal functson of the cells Surks (12) described a slgmficant increase in cellular DNA and cellular as well as secreted GH, in S phase cultures of a rat pituitary cell hne that produces GH He concluded that GH productmn rate observed m asynchronous cultures may be an integrated value for all phases of the cell cycle Concerning the electrophyslologlcal properties of pstuitary cells, Duff/ et al (2) showed that sp=ke acttvity could not be triggered by current pulse rejection when the resting potential was less than -40 mV These authors proposed that the low values of membrane potential and the inexmtabdlty observed for more than 50 % of clonal cell population were not necessarily due to a cells damage occurmg after empalement Our results suggest that cell kmetms may influence the electrical properties of secretory cells in that only cells m phases G2 and M are exmtable In our expenments, cells m these two phases represented 15 % of the total populatmn The observation that Duff/ et al. (2) found a greater percentage of exmtable cells (about 50 %) may be due to an experimental b=as m that it m easier to record from large cells and we have shown that large cells are in the end phase of the cell cycle (= e G2 and M). However, the statmtlcal factor for evaluating the sub-population of the cell cycle may cause us to underestimate our percentage of exc=table cells by not taking into account those cells wNch are at the end of the S phase and, by their s0ze, may be included in the G2+M category Boonstra et al (13) observed that membrane potential values and K + activity from neuroblastoma clonal cells were high in mitosis, decreased in G1 phase and rose again during S phase Membrane potential value vaned from -20 mV (G1) to -50 mV (G2 - M) This is in agreement with our observations which showed the decrement of the relative value of resting potential from -16 to -50 mV. We have observed an increase of the membrane resmtance from 30 to 152 MQ from the G1 phase to the G2+M phase. However, if the resistance is reported to the cell surface unit, we found a relative decrease of 1 8 between the specific resistance in Gt and G2+M phases Thin agree with Boonstra's observations (13) whmh reported a decrease of about 1 7. The hyperpolanzat~on of the resting membrane potential has to be related to a specific increase of the membrane conductance for potassium =ons, because it's value depends on the extracellular potassium concentration (results not shown). This is necessary for tnggenng action potentials it has been shown that reward currents (sodium and calcium) for GH3 cells are reactivated at membrane potentials less than -40mV (14) However, the rectification observed for depolanzmg potentials w=th cells in G2+M phase =s due to a specific increase of the potasmum currents (15). In conclusion, our results have shown that GH3/B6 cells were excitable only m the G2+M state, and that TRH was able to s~gmficantly increase growth hormone and protactln hberatlons only in the G2+M phase. These finding are in accordance with the"stimulus-secretion" couphng proposed by Douglas (1968) but it implies that this couphng is dependent of the phases of the cell cycle.
2384
Cell Cycles and Hormone Release in Pitultary
Vol. 40, No. 24, 1987
p.cknowledoments The authors gratefully acknowledge the sktlful asststance of Mrs D Verrier and Mrs J Arsaut m maintaining the cell hne and for the RIA and Miss A. Ferret-Bouin. We would like to thank Professor Mawas (Facult6 de Marseille) and Professor Rosenfeld (CNRS, Villejulf) for allowmg us to use their cell sorters The dye Hoechst 33342 was a gtft from the Riedel de Hahn laboratory The RIA ktt was a gift from Drs Parlow and Raitl of the NIAMDD We are grateful to Mr S. Vltelho who labelled the growth hormone w0th 1251
References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Y. KIDOKORO, Nature 258. 741-743 (1975) B DUFY, J D. VINCENT, H FLEURY, P. DU PASQUIER, D. GOURDJI, A TlXlER-VlDAL, Sctence 204. 509-511 (1979) WW DOUGLAS, Br J Pharmacol 34, 451-474 (1968) B DUFY and J D. VINCENT, Serono Svmooslum N°49. p.89-99, Academtc Press, London (1982) D J ARNDT-JOUIN and T.M. JOUIN, Htstochem-Cytochem 25, 585-589 (1977). M PAWLIKOWSKI, Life Sctences 30, 315-320, 1982 G.R SMITH, M.L. GURSON, AJ KIDDEL, A.D PERRIS, J, Endocrinology 65. 45-53 (1975) J M VARGA, A DIPASQUALE, J. PAWELECK, J.S MC GUIRE, A.B LERNER, Proc Nat Acad Sct. USA, .L1.., 1590-1593 (1974) O P F CLAUSEN, K M GAUTVlK, E HAUG, J. Cell Phystol. 94, 205-214 (1978). D R. KIINO, D E BURGER, P S DANNIES, J. of Cell Btol. 93, 459-462 (1982). A FAIVRE-BAUMAN, D GOURDJI, D GROUSELLE, A TlXlER-VlDAL, Biochem. and BIophys Res Comm. . ~ 50-67, 1975. I SURKS, Endocrinology 114, 873-87g (1984) J BOONSTRA, C.L MUMMERY, L.G.J. TERTOOLEN, P Van Der SAAG, S.W De LAAT, J of Cell Physiol 107, 75-83 (1981). D R MATTESON and C M ARMSTRONG, J. Gen PhysJol 83. 371-394 (1984) J M DUBINSKY and G.S. OXFORD, J. Gen. Physiol. 83. 309-339 (1984).