- Email: [email protected]
Arachidonic acid stimulates 45calcium efflux and hPL release in isolated trophoblast cells
Life Sciences, Vol. 38, pp. 99-i07 Printed i n ~ e U.S.A.
Pergamon Press
ARACHIDONIC ACID STIMULATES 45CALCIUM EFFLUX AND HPL RELEASE IN ISOLATED TROPHOBLAST CELLS Philip Zeitler I, Elizabeth Murphy*, and Stuart Handwerger 2 Departments of Pediatrics and Physiology, Duke University Medical Center, Durham, NC and the *National Institute of Environmental Health Sciences, Research Triangle Park, NC (Received in final form October 24, 1985) Summary Previous investigations from this laboratory have indicated that arachidonic acid stimulates a rapid, dose-dependent and reversible increase in hPL release which is not dependent on cyclooxygenase or lipoxygenase metabolism. To investigate further the mechanism by which arachidonic acid stimulates the release of hPL, the effect of arachidonic acid on the release of 45Ca from perifused cells prelabelled with 45CACI was examined in an enriched cell culture population of term human syncytiotrophoblast. Arachidonic acid (i0-I00 uM) stimulated a dose-dependent, rapid, and reversible increase in the release of both 4JCa and hPL from the perifused placental cells. On the other hand, palmitic acid had little effect on either hPL release or 45Ca release even at concentrations as high as i00 uM. lonophore A23187 (i-I0 uM) also stimulated a dose-dependent and reversible increase in hPL release. Since arachidonic acid increases the mobilization of cellular calcium, as reflected by the increased 45calcium efflux, and since an increase in the intracellular calcium concentration appears to stimulate an increase in hPL release, these results suggest that the stimulation of hPL release by arachidonic acid may be due, at lease in part, to the effects of the fatty acid on cellular calcium mobilization. Human placental lactogen (hPL) is a gestational polypeptide secreted by the placenta which has striking chemical similarity to both human growth hormone and prolactin. HPL secretion, however, is not responsive to factors known to regulate these pituitary hormones (for review, see i). Although the secretogogues responsible for the physiologic regulation of hPL secretion remain unclear, a role for arachidonic acid in the regulation of hPL secretion is suggested by recent studies from this laboratory (2,3). In these studies, arachidonic acid was found to stimulate a rapid dose-dependent increase in human placental lactogen (hPL) release in vitro (2,3). The mechanisms by which arachidonic acid stimulate hPL release are unknown, but the stimulation does not appear to be due to cyclooxygenase or lipoxygenase products of arachidonic acid since inhibitors of these pathways did not prevent the action of arachidonic acid.
ISupported by grant from the National Institutes of Health (GM0771) 2Supported by a grant from the National Institutes of Health (HD-07447), to whom reprint requests should be addressed. 0024-3205/86 $3.00 + .00 Copyright (c) 1986 Pergamon Press Ltd.
i00
Arachidonic Acid, Calcium Flux and hPL
Vol. 37, No, 2, 1986
Arachidonic acid stimulates the flux of calcium across isolated sarcoplasmic mitochondrial, and plasma membranes (4-7) and stimulates the hydrolysis of phosphoinositides in trophoblast cells (submitted for publication). Since the hydrolysis of phosphoinositides has been associated with cellular calcium mobilization and hormone release in a wide variety of cell types (8,9), we have examined whether the effect of arachidonic acid on hPL release is associated with an increase in calcium efflux. Since the placenta is a heterogeneous tissue containing numerous cell types, the studies were performed utilizing an enriched fraction of trophoblast cells which synthesize and release hPL° Methods PLACENTAL CELL PREPARATION Placental tissue was obtained immediately after delivery from women with normal pregnancies of 37 to 40 weeks gestation. Permission to obtain the tissue was granted by the Human Investigation Committee of Duke University. Individual cotyledons were dissected from the placenta and placed in chilled RPMI-1640 (Flow Laboratories) buffered with 12.5 mM HEPES (Sigma) (pH 7.5) for transport to the laboratory. An enriched fraction of syncytiotrophoblast cells was prepared as previously described (3). Briefly, after removal of decidual tissue from the maternal aspect, placenta was washed and coarsely minced. Minced tissue was dissociated at 37°C for 60 minutes with 0.1% collagenase (Type III; Worthington Enzymes) and 0.1% hyaluronidase (Sigma) in RPMI-1640 buffered with bicarbonate (2mM) and HEPES (12.5 mM), pH 7.5, containing I% fetal calf serum, 0.01% soybean trypsin inhibitor, and 0.001% deoxyribonuclease I (Worthington Enzymes). The resulting cell suspension was filtered through nitex cloth (150 uM mesh), and the cells were collected by centrifugation at i00 x g for 6 min. After washing twice in RPMI-1640, cell number and viability were assessed by trypan blue exclusion The dissociated cells were separated by isopycnic centrifugation on linear gradients of 40% Percoll (Sigma) (0.8-1.0 x 108 cells/gradient). Previous studies (3) indicate that greater than 90% of the cells which synthesize and release hPL are recovered with a density of 1.02-1.01 gm/ml and account for approximately 15% of the total placental DNA. This procedure provides a 10-15 fold enrichment of the population in hPL-producing syncytiotrophoblast. MEASUREMENT OF 45CALCIUM EFFLUX Freshly prepared cells were resuspended in 20 mls of media containing 8 uCi/ml 45CACI 2 (Amersham; 2 uCi/umol) and incubated for two hours. After incubation, cells were sedimented at 100 x g for 6 mino Two parallel columns (0.15 ml bed volume) of acetylated Sephadex G-IO (Pharmacia) (I0) were formed in 1 ml syringes cut to a volume of 0.4 ml and plugged at the tip with acetylated cotton wool. The columns were washed and packed by perifusion with distilled H20 at a rate of 1 ml/min with a dual-channel peristaltic pump. After extensive washing, the medium was changed to RPMI-1640, pH 7.6 continuously bubbled with 5%CO2-95%O 2 and maintained at 37°C in a water bath. Perifusion with RPMI-1640 was continued for 30 column volumes to ensure equilibration before addition of cells. After removal of duplicate 0.05 ml aliquots of the cell suspension for protein determination by the method of Bradford (ii), 0.2 ml of the cell suspension was layered over each of the two columns and the columns were perifused with RPMI-1640 at a rate of 0.5 ml/min. The release of 45-calcium by trophoblast large initial component due to extracellular
cells decreased with time, with a unincorporated 45Ca° However,
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
i01
after approximately thirty minutes, the release of radioactivity from the cells remained nearly constant. When two parallel columns were perifused with control medium, efflux rates for the two columns after the initial thirty minute period were essentially identical for three to four hours of perifusion (fig IA). Therefore, all experiments involving the effects of variables on the rate of calcium efflux were performed after an initial 40 minutes perifusion period. In each experiment, one cell column was used to examine the effects of arachidonic or palmitic acid, while the other column was used to examine the effects of vehicle alone. Arachidonic acid (Calbiochem-Behring) and palmitic acid (Sigma) were dissolved in absolute ethanol with a final ethanol concentration less than 0.3%. After forty minutes of perifusion in control media, one column was perifused for 30 minutes with medium containing the substance to be examined, while the parallel control column was perifused with medium containing vehicle alone. After the thirty minute exposure period, both columns were perifused with control medium for an additional 40 minutes. The effluent was collected in two min fractions and an aliquot was counted by liquid scintillation spectroscopy in i0 ml Aquasol II (New England Nuclear). At the end of each perifusion, the radioactivity remaining in the cells and columns was determined. 5% SDS/5mM EDTA was added to each column and the contents of the column were counted by liquid scintillation spectroscopy. The desaturation curve relating the radioactivity to time for each column was integrated according to the method of Uchikawa and Borle (12). The sum of all the radioactivity in each effluent period plus the radioactivity left in the cells at the end of the efflux was taken as the total radioactivity present in the cells at the beginning of the desaturation period. The percentage of the total radioactivity remaining in the cells at each time point was then obtained by sequentially subtracting the radioactivity of each fraction. The Fractional Efflux Rate (FER) for each fraction was determined as: FER=cpme/[(Cpmc)meant]
x i00
where cpm e is the radioactivity appearing in the effluent during the time interval t and (cpmc)mean is the mean radioactivity remaining in the cells during the time interval. FER values for the experimental fractions were determined as a fraction of the corresponding values for the control column (FER ratio). The FER ratio for each experiment was then graphed as a function of time on the same scale and the area under each curve during the period of perifusion, representing the total isotope release, was determined by planimetry. Determination
of hPL release from perifused cells
hPL content of the effluent fractions was determined by homologous radioimmunoassay (13). Since the amount of hPL release into the media during each time interval was below the detectability of the hPL radioimmunoassay, effluent fractions were pooled and concentrated. Every seven consecutive tubes (15 min) during the perifusion were combined and the proteins precipitated by addition of tricholoracetic acid to a final concentration of 10%. After centrifugation at 3200 x g for 30 min, the supernatants were decanted and the pellets allowed to drain. The pellets were resuspended in 0.4 ml of 0.2 M phosphate-buffered saline, pH 7.4, before radioimmunoassay. Recovery of a known quantity of hPL by this method ranged from 86-92%. EFFECTS OF A23187 ON hPL RELEASE In order to examine the effects on hPL release of elevating the intracellular calcium concentration with the calcium ionophore A23187, cells (106/well)
102
Arachidonic Acid, Calcium Flux and hPL
I.stA
Vol. 38, No. 2, ]986
B
22o
1.5-
oUJ
1.2-
W ._J W n~
i. I _
d 8_ c-
t--nW u-
18 14; I0 6
1.0
2 I
20
I
40
I
1
60
1
I
80
I
I
I00
M I N UTES
30
50
70
90
MINUTES FIGURE 1
The effect of 50 uM arachidonic acid on FER and hPL release from perifused placental cells. After forty minutes of perifusion~ cells were exposed for 30 minutes (bar) to 50 uM arachidonic acid or to vehicle alone. A) the effect of arachidonic acid on FER. The stippled area represents the results of control experiments (N=8) in which two parallel columns were perifused with control media. The results are expressed as the ratio of FER values for the two columns, mean + S.E. 4--O represents the results of a typical experiment in which an experimentaT column was exposed for 30 minutes (bar) to 50 uM arachidonic acid and a parallel control column was exposed to vehicle alone. The results are expressed as the ratio of experimental to control FER values. B) the effect of arachidonic acid on hPL release. The height of the hatched bar represents the ratio of hPL release per 15 minutes by two control columns perifused with media alone, mean + S.E (N=4 experiments). The height of the open bar represents the ratio of hPL release per 15 minutes in an experimental column perifused with media containing 50 uM arachidonic acid compared to a parallel column perifused with media containing vehicle alone.
were seeded onto 24-well culture plates (Linbro plastics) pre-coated with collagen (Vitrogen i00, Flow laboratories) in i ml of RPMI-1640, pH 7.4 containing i0% FCS, 12.5 mM Hepes, 2mM NaHC03, 50 ug/ml gentamicin (GIBCO) and 5 ug/ml Amphotericin B (Squibb). Cell density was approximately 106 cells/well and cells were maintained in an atmosphere of 5%CO2-95%air. The medium was changed at 48 hours. After 72 hrs in culture, the culture medium was removed and the cells washed with I ml of either RPMI-1640 or calcium, magnesium free Hank's balanced salt solution containing i mM EGTA (HBSS). The medium was then replaced with 1 ml RPMI-1640 or HBSS and the cells were incubated for an additional two hours. After removal of a 0.3 ml sample, ionophore A23187 (Calbiochem-Behring) or vehicle alone (DMSO) were added to the cells and the total volume readjusted to i ml. At
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
103
PALMITIC ACID 100
"E
E ~'500 rr o
ARACHIDONIC ACID
n~ 4 0 0 hi a Z D 300 IAl n~ 200
I00
0
I0 50 I00 FATTY ACID, .I.LM
FIGURE 2 The dose-dependent effect of arachidonic acid on FER in perifused placental cells. After forty minutes of perifusion, cells were exposed for 30 minutes to the indicated doses of arachidonic acid, palmitic acid or to vehicle alone. For each experiment, the ratio of experimental to control FER values were graphed as shown in figure IA. The bars represent the relative areas under these curves during the 30 minute exposure to arachidonic acid (~), palmitic acid ( [ ] ) , or vehicle as determined by planimetry, mean + S.E. (N=6 for 0 uM, N=3 for all other doses). *, P < 0.05; **, P < 0.01
various times, 0.3 ml samples were removed and the media replaced with an equivalent amount of medium containing the agent in the appropriate final concentration, hPL concentrations in all samples were determined by homologous radioimmunoassay (13). All experiments were performed in triplicate, and statistical differences between sample means were tested by analysis of variance followed by Dunnett's test. Results
After a 40 min pre-incubation period, exposure of placental cells for 30 minutes to 50 uM arachidonic acid stimulated a rapid increase in the release of 45calcium (fig IA) which was accompanied by a 17.8-fold increase in hPL release (fig IB). The fractional efflux rate (FER) increased approximately 5 minutes after the beginning of the arachidonic acid perifusion and remained elevated for the entire period of exposure. FER returned toward baseline within 5 minutes after the arachidonic acid perifusion was discontinued. Likewise, hPL release increased within 15 minutes after the start of the arachidonic acid exposure, remeained elevated throughout the exposure period, and decreased rapidly when the arachidonic acid perifusion was discontinued. In contrast, exposure of cells to vehicle alone had no effect on either the rate of 45calcium efflux or hPL release. In other control experiments, arachidonic acid had no effect on
104
Arachidonic
Acid, Calcium Flux and hPL
Vol. 38, No. 2, 1986
45calcium release from columns containing heat-inactivated taining no cells (data not shown).
cells or columns con-
The effect of arachidonic acid on FER and hPL release was dose-dependent. As shown in fig 2 and 3, cells exposed for 30 minutes to I0 uM and 50 uM arachidonic acid released 242% and 453% more 45Ca and 354% and 1279% more hPL than control cells. In addition, the effect of arachidonic acid on FER was dependent on the duration of exposure to the fatty acid. Decreasing the duration of arachidonic acid exposure to i0 or 20 min resulted in a corresponding decrease in the period of increased 45Ca efflux (fig 4).
'I
PALMIT~ACI IC~ D ARACHIDONI ACICD
140
uJlO-
09 <1 DJ
FIGURE 3
UJ Or
6O_
c-
2 0 I0 50 FATTY ACI D,.I.I.M
The dose-dependent effect of arachidonic acid on hPL release from perifused placental cells. After forty minutes of perifusion, cells were exposed for 30 minutes to the indicated doses of arachidonic acid ( I ) , palmitic acid ( ~ ), or to vehicle alone. The bars represent the ratio of hPL release per 30 minutes in the experimental columns compared to control columns perifused with media containing vehicle alone, mean + S.E. (N=3). *, P < 0.05; **, P-< 0.01
In contrast, palmitic acid had no effect on hPL release and only a small effect on 45Ca release, even at a concentration of i00 mM (fig 2,3). Since stimulation of hPL release by arachidonic acid was associated with an increase in 45Ca efflux, the effect of the calcium ionophore A23187 on hPL release was examined. As shown in fig 5, A23187 stimulated an increase in hPL release, with 2 uM and i0 uM resulting in 37.8 % and 130.9 % increases respectively in media containing 1.6 mM Ca 2+. A23187, however, had no significant effect on hPL release from cells incubated in calcium-free medium°
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
105
1.2
0
FIGURE 4
I.I
l.-
The effect of varying the duration of arachidonic acid exposure on FER in perifused placental cells. After forty minutes of perifusion, cells were exposed for i0 minutes (bar) to I0 uM arachidonic acid or to vehicle alone. Both columns were then exposed to control medium for 30 minutes, followed by an additional 20 minute exposure to arachidonic acid or vehicle and a final 30 minute perfusion in control medium. The results are expressed as the ratio of experimental to control FER values.
rv hl h
1.0 i ~ i i
III i l I 20 40
I
I
1
60
I
I
i
I
I
80 I00 MINUTES
i20
I
I
140
30
T 20 r-
-
nn .E
T "r
I0
,
,r
!
,
i
0
I
2
I0
0
I
2
I0
A 2318T, jJ.M FIGURE 5 The effect of ionophore A23187 on hPL release from placental cells. After 72 hours in culture, placental cells (106/well) were washed and incubated for 2 hours in either 1 ml RPMI ( 4 ) or calcium, magnesium free Hank's Balanced Salt solution ( J ). After removal of a 0.3 ml sample, ionophore A23187 or vehicle alone (DMSO) were added to the cells and the volume readjusted to 1 ml. After 2 hours, media were removed and assayed for hPL as described. Bars represent accumulted hPL ng, mean + S.E.M (N=4) ** P < 0 01.
106
Arachidonic
Acid, Calcium Flux and hPL
Vol. 38, No. 2, 1986
Discussion The results of this study indicate that arachidonic acid stimulates dosedependent, rapid and reversible increases in 45calcium and hPL release from an enriched fraction of hPL-secreting placental cells. The stimulation of 45calcium efflux in response to arachidonic acid was due to mobilization of calcium and not due to displacement of extracellular or non-specificially bound 45Ca since arachidonic acid had no effect on 45calcium release from heatinactivated cells or columns containing no cells The magnitude of the increase in 45calcium efflux and hormone release in placental cells responding to arachidonic acid is similar to the effect of arachidonic acid on 45calcium efflux and prolactin release from GH 3 cells of pituitary origin (14). An increase in 45calcium efflux may reflect a direct increase in the rate of flux of calcium across the plasma membrane or, alternatively, it may reflect a secondary response to a primary increase in the intracellular calcium concentration. Previous investigations have suggested that arachidonic acid may act to promote the passive diffusion of calcium across phospholipid membranes (2,4-7). Furthermore, arachidonic acid has been reported to promote the release of calcium from mitochondria and endoplasmic reticulum in liver and kidney cells (5). Therefore, arachidonic acid may stimulate the release of 45calcium from placental cells by stimulating a primary influx of calcium into the cytoplasm. Alternatively, arachidonic acid may promote release of calcium from intracellular stores. This change in calcium mobilization would then lead to an elevation of the intracellular free calcium concentration. Further clarification of this response will require the direct measurement of the intracellular free calcium concentration during exposure to arachidonic acid. Attempts to measure the intracellular free calcium concentration utilizing the fluorescent dye Quin2 have been unsuccessful due to marked autofluorescence of the fatty acid. The experiments with A23187 suggest that elevation of the intracellular calcium concentration stimulates an increase in hPL release. The calcium dependence of the response to A23187 indicates that the effect is due to changes in calcium distribution rather than an effect of A23187 itself. The source of the calcium mobilized by A23187 is not clear and may be either extracellular or intracellular. The stimulation of hPL release by arachidonic acid was similarly calcium-dependent (data not shown). Since the increase in calcium efflux induced by arachidonic acid most likely reflects an increase in intracellular calcium mobilization, the stimulation of hPL release by arachidonic acid may be due to the effect of the fatty acid on cellular calcium mobilization. Although the effect of arachidonic acid on hPL release is markedly diminished in calcium-free medium, the basal rate of hPL release is increased in calcium-free media (15 16). The reasons for this difference in the calcium dependence of basal and stimulated hPL release is not clear, but the observation suggests that hormone release under these two conditions may represent different cellular processes. In support of this suggestion, the increase in hPL release seen in calcium-free medium is transient and is followed by a period of decreased hormone release, while the release of hPL in response to arachidonic acid persists for many hours (2,3). We have recently reported that arachidonic acid stimulates the hydrolysis of phosphoinositides in an enriched fraction of hPL-secreting placental cells (submitted for publication). In these studies, phosphoinositide hydrolysis w a s demonstrated by an increase in the release of [3H]-myoinositol from perifused ' cells exposed to arachidonic acid, as well as the speciflc loss of 3 2 P r a d ioactivity from phosphatidylinositol and phosphatidylserine. Since phosphoinositide hydrolysis has been associated with calcium mobilization in a variety of cell
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
107
types (8,9), the effects of arachidonic acid on calcium metabolism described here may be due, at least in part, to the stimulation of phosphoinositide hydrolysis. Arachidonic acid stimulates the release of a variety of hormones, including prolactin~ luteinizing hormone, and histamine (14,17,18). In addition, arachidonic acid is released in response to receptor stimulation by a variety of secretagogues (19-22). These observations, together with the results presented here, suggest that the release of free arachidonic acid in response to receptoroccupancy may play an important role in the stimulation of hormone release by influencing cellular calcium metabolism. Furthermore, arachidonic acid has been reported to stimulate guanylate cyclase (23-25) and protein kinase C (26) suggesting that arachidonic acid release may link receptor occupancy to the stimulation of a number of intracellular processes associated with hormone secretion.
References I. S. HANDWERGER, T. HURLEY and A. GOLANDER, In Iffy, L. and H. Kaminetzky (eds.), P_rrin_ciplesandPracttice_of_Obstetrics an_d_Gynecol_ogy, Wiley and Sons, New York, pp 243-261 (1981). 2. S. HANDWERGER, J. BARRETT, S. BARRY, E. MARKOFF, P. ZEITLER, B. CWIKEL and M. SIEGEL, Mol. Pharmacol. 20 609-613 (1981). 3. P. ZEITLER, E. MARKOFF and S. HANDWERGER, J. Clin. Endocrinol. Metab. 57 812-818 (1983) 4. D. SELLER and W. HASSELBACH, Eur. J. Biochem. 21 385-387 (1971). 5. I. ROMAN, P. GMAJ, C. NORVIELA and S. ANGIELSKI, Eur. J. Biochem. 102 615-623 (1979). 6. M. VOLPI, P. NACCACHE and R. SHA'AFI, Biochem. Biophys. Res. Commun. 92 1231-1237 (1980). 7. A. CHEAH, Biochim. Biophys. Acta 648 113-119 (1981). 8. R. MICHELL, Biochim. Biophys. Acta 415 81-147 (1977). 9. R. FARESE, Metabolism 32 628-644 (1983). I0. K. REED, Biochem. Biophys, Res. Commun. 50 1136-1142 (1973). ii. M. BRADFORD, Anal. Biochem. 72 248-254 (1976) 12. T. UCHIKAWA and A. BORLE, Am. J. Physiol. 234 1234-1238 (1978). 13. S. HANDWERGER and L. SHERWOOD, In Jaffe, B. and H. Behrman (eds.), Methods 9fHormone_Radioi_mmun0assa~, Aca--demic Press, New York, pp. 417-426 (~74)~-14. R. KOLESNICK, I. MUSACCHIO, C. THAW and M. GERSHENGORN, Am. J. Physiol. 246 E458-E462 (1984). 15. S. HANDWERGER, P.M. CONN, J. BARRETT, S. BARRY and A. GOLANDER, Am. J. Physiol. 240 E550-E555 (1981). 16. V. CHOY and W. WATKINS, J. Endocrinol. 69 347-358 (1976). 17. Z. NAOR and K. CATT, J. Biol. Chem. 256 2226-2229 (1981). 18. A. NAKAO, A. BUCHANAN and P. POTOBAR, Int. Archs. Allergy. Appl. l~unol. 63 3 0 4 3 (1981). 1 9 R. BELL, N. BAENZIGER and P. MAJERUS, Prostaglandins 20 269-274 (1980). 20. B. HAYE, S. CHAMPION and C. JACQUEMIN, Febs. Letts. 30 253-260 (1973). 21. M. SCHREY and R. RUBIN, J. Biol. Chem. 254 11063-11066 (1980). 22. S. LAYCHOCK, Diabetes 32 6-13 (1983). 2 3 D. LEIBER and S. HARBON~ Mol. Pharmacol. 21 654-663 (1981). 24. D. WALLACH and I. PASTAN, J. Biol. Chem. 251 5802-5809 (1976). 25. R. BRIGGS and F. DERUBERTIS, Biochem. Pharmacol. 29 717-722 (1980). 26. L. MCPHAIL, C. CLAYTON and R. SNYDERMAN, Science 224 622-625 (1984).
Pergamon Press
ARACHIDONIC ACID STIMULATES 45CALCIUM EFFLUX AND HPL RELEASE IN ISOLATED TROPHOBLAST CELLS Philip Zeitler I, Elizabeth Murphy*, and Stuart Handwerger 2 Departments of Pediatrics and Physiology, Duke University Medical Center, Durham, NC and the *National Institute of Environmental Health Sciences, Research Triangle Park, NC (Received in final form October 24, 1985) Summary Previous investigations from this laboratory have indicated that arachidonic acid stimulates a rapid, dose-dependent and reversible increase in hPL release which is not dependent on cyclooxygenase or lipoxygenase metabolism. To investigate further the mechanism by which arachidonic acid stimulates the release of hPL, the effect of arachidonic acid on the release of 45Ca from perifused cells prelabelled with 45CACI was examined in an enriched cell culture population of term human syncytiotrophoblast. Arachidonic acid (i0-I00 uM) stimulated a dose-dependent, rapid, and reversible increase in the release of both 4JCa and hPL from the perifused placental cells. On the other hand, palmitic acid had little effect on either hPL release or 45Ca release even at concentrations as high as i00 uM. lonophore A23187 (i-I0 uM) also stimulated a dose-dependent and reversible increase in hPL release. Since arachidonic acid increases the mobilization of cellular calcium, as reflected by the increased 45calcium efflux, and since an increase in the intracellular calcium concentration appears to stimulate an increase in hPL release, these results suggest that the stimulation of hPL release by arachidonic acid may be due, at lease in part, to the effects of the fatty acid on cellular calcium mobilization. Human placental lactogen (hPL) is a gestational polypeptide secreted by the placenta which has striking chemical similarity to both human growth hormone and prolactin. HPL secretion, however, is not responsive to factors known to regulate these pituitary hormones (for review, see i). Although the secretogogues responsible for the physiologic regulation of hPL secretion remain unclear, a role for arachidonic acid in the regulation of hPL secretion is suggested by recent studies from this laboratory (2,3). In these studies, arachidonic acid was found to stimulate a rapid dose-dependent increase in human placental lactogen (hPL) release in vitro (2,3). The mechanisms by which arachidonic acid stimulate hPL release are unknown, but the stimulation does not appear to be due to cyclooxygenase or lipoxygenase products of arachidonic acid since inhibitors of these pathways did not prevent the action of arachidonic acid.
ISupported by grant from the National Institutes of Health (GM0771) 2Supported by a grant from the National Institutes of Health (HD-07447), to whom reprint requests should be addressed. 0024-3205/86 $3.00 + .00 Copyright (c) 1986 Pergamon Press Ltd.
i00
Arachidonic Acid, Calcium Flux and hPL
Vol. 37, No, 2, 1986
Arachidonic acid stimulates the flux of calcium across isolated sarcoplasmic mitochondrial, and plasma membranes (4-7) and stimulates the hydrolysis of phosphoinositides in trophoblast cells (submitted for publication). Since the hydrolysis of phosphoinositides has been associated with cellular calcium mobilization and hormone release in a wide variety of cell types (8,9), we have examined whether the effect of arachidonic acid on hPL release is associated with an increase in calcium efflux. Since the placenta is a heterogeneous tissue containing numerous cell types, the studies were performed utilizing an enriched fraction of trophoblast cells which synthesize and release hPL° Methods PLACENTAL CELL PREPARATION Placental tissue was obtained immediately after delivery from women with normal pregnancies of 37 to 40 weeks gestation. Permission to obtain the tissue was granted by the Human Investigation Committee of Duke University. Individual cotyledons were dissected from the placenta and placed in chilled RPMI-1640 (Flow Laboratories) buffered with 12.5 mM HEPES (Sigma) (pH 7.5) for transport to the laboratory. An enriched fraction of syncytiotrophoblast cells was prepared as previously described (3). Briefly, after removal of decidual tissue from the maternal aspect, placenta was washed and coarsely minced. Minced tissue was dissociated at 37°C for 60 minutes with 0.1% collagenase (Type III; Worthington Enzymes) and 0.1% hyaluronidase (Sigma) in RPMI-1640 buffered with bicarbonate (2mM) and HEPES (12.5 mM), pH 7.5, containing I% fetal calf serum, 0.01% soybean trypsin inhibitor, and 0.001% deoxyribonuclease I (Worthington Enzymes). The resulting cell suspension was filtered through nitex cloth (150 uM mesh), and the cells were collected by centrifugation at i00 x g for 6 min. After washing twice in RPMI-1640, cell number and viability were assessed by trypan blue exclusion The dissociated cells were separated by isopycnic centrifugation on linear gradients of 40% Percoll (Sigma) (0.8-1.0 x 108 cells/gradient). Previous studies (3) indicate that greater than 90% of the cells which synthesize and release hPL are recovered with a density of 1.02-1.01 gm/ml and account for approximately 15% of the total placental DNA. This procedure provides a 10-15 fold enrichment of the population in hPL-producing syncytiotrophoblast. MEASUREMENT OF 45CALCIUM EFFLUX Freshly prepared cells were resuspended in 20 mls of media containing 8 uCi/ml 45CACI 2 (Amersham; 2 uCi/umol) and incubated for two hours. After incubation, cells were sedimented at 100 x g for 6 mino Two parallel columns (0.15 ml bed volume) of acetylated Sephadex G-IO (Pharmacia) (I0) were formed in 1 ml syringes cut to a volume of 0.4 ml and plugged at the tip with acetylated cotton wool. The columns were washed and packed by perifusion with distilled H20 at a rate of 1 ml/min with a dual-channel peristaltic pump. After extensive washing, the medium was changed to RPMI-1640, pH 7.6 continuously bubbled with 5%CO2-95%O 2 and maintained at 37°C in a water bath. Perifusion with RPMI-1640 was continued for 30 column volumes to ensure equilibration before addition of cells. After removal of duplicate 0.05 ml aliquots of the cell suspension for protein determination by the method of Bradford (ii), 0.2 ml of the cell suspension was layered over each of the two columns and the columns were perifused with RPMI-1640 at a rate of 0.5 ml/min. The release of 45-calcium by trophoblast large initial component due to extracellular
cells decreased with time, with a unincorporated 45Ca° However,
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
i01
after approximately thirty minutes, the release of radioactivity from the cells remained nearly constant. When two parallel columns were perifused with control medium, efflux rates for the two columns after the initial thirty minute period were essentially identical for three to four hours of perifusion (fig IA). Therefore, all experiments involving the effects of variables on the rate of calcium efflux were performed after an initial 40 minutes perifusion period. In each experiment, one cell column was used to examine the effects of arachidonic or palmitic acid, while the other column was used to examine the effects of vehicle alone. Arachidonic acid (Calbiochem-Behring) and palmitic acid (Sigma) were dissolved in absolute ethanol with a final ethanol concentration less than 0.3%. After forty minutes of perifusion in control media, one column was perifused for 30 minutes with medium containing the substance to be examined, while the parallel control column was perifused with medium containing vehicle alone. After the thirty minute exposure period, both columns were perifused with control medium for an additional 40 minutes. The effluent was collected in two min fractions and an aliquot was counted by liquid scintillation spectroscopy in i0 ml Aquasol II (New England Nuclear). At the end of each perifusion, the radioactivity remaining in the cells and columns was determined. 5% SDS/5mM EDTA was added to each column and the contents of the column were counted by liquid scintillation spectroscopy. The desaturation curve relating the radioactivity to time for each column was integrated according to the method of Uchikawa and Borle (12). The sum of all the radioactivity in each effluent period plus the radioactivity left in the cells at the end of the efflux was taken as the total radioactivity present in the cells at the beginning of the desaturation period. The percentage of the total radioactivity remaining in the cells at each time point was then obtained by sequentially subtracting the radioactivity of each fraction. The Fractional Efflux Rate (FER) for each fraction was determined as: FER=cpme/[(Cpmc)meant]
x i00
where cpm e is the radioactivity appearing in the effluent during the time interval t and (cpmc)mean is the mean radioactivity remaining in the cells during the time interval. FER values for the experimental fractions were determined as a fraction of the corresponding values for the control column (FER ratio). The FER ratio for each experiment was then graphed as a function of time on the same scale and the area under each curve during the period of perifusion, representing the total isotope release, was determined by planimetry. Determination
of hPL release from perifused cells
hPL content of the effluent fractions was determined by homologous radioimmunoassay (13). Since the amount of hPL release into the media during each time interval was below the detectability of the hPL radioimmunoassay, effluent fractions were pooled and concentrated. Every seven consecutive tubes (15 min) during the perifusion were combined and the proteins precipitated by addition of tricholoracetic acid to a final concentration of 10%. After centrifugation at 3200 x g for 30 min, the supernatants were decanted and the pellets allowed to drain. The pellets were resuspended in 0.4 ml of 0.2 M phosphate-buffered saline, pH 7.4, before radioimmunoassay. Recovery of a known quantity of hPL by this method ranged from 86-92%. EFFECTS OF A23187 ON hPL RELEASE In order to examine the effects on hPL release of elevating the intracellular calcium concentration with the calcium ionophore A23187, cells (106/well)
102
Arachidonic Acid, Calcium Flux and hPL
I.stA
Vol. 38, No. 2, ]986
B
22o
1.5-
oUJ
1.2-
W ._J W n~
i. I _
d 8_ c-
t--nW u-
18 14; I0 6
1.0
2 I
20
I
40
I
1
60
1
I
80
I
I
I00
M I N UTES
30
50
70
90
MINUTES FIGURE 1
The effect of 50 uM arachidonic acid on FER and hPL release from perifused placental cells. After forty minutes of perifusion~ cells were exposed for 30 minutes (bar) to 50 uM arachidonic acid or to vehicle alone. A) the effect of arachidonic acid on FER. The stippled area represents the results of control experiments (N=8) in which two parallel columns were perifused with control media. The results are expressed as the ratio of FER values for the two columns, mean + S.E. 4--O represents the results of a typical experiment in which an experimentaT column was exposed for 30 minutes (bar) to 50 uM arachidonic acid and a parallel control column was exposed to vehicle alone. The results are expressed as the ratio of experimental to control FER values. B) the effect of arachidonic acid on hPL release. The height of the hatched bar represents the ratio of hPL release per 15 minutes by two control columns perifused with media alone, mean + S.E (N=4 experiments). The height of the open bar represents the ratio of hPL release per 15 minutes in an experimental column perifused with media containing 50 uM arachidonic acid compared to a parallel column perifused with media containing vehicle alone.
were seeded onto 24-well culture plates (Linbro plastics) pre-coated with collagen (Vitrogen i00, Flow laboratories) in i ml of RPMI-1640, pH 7.4 containing i0% FCS, 12.5 mM Hepes, 2mM NaHC03, 50 ug/ml gentamicin (GIBCO) and 5 ug/ml Amphotericin B (Squibb). Cell density was approximately 106 cells/well and cells were maintained in an atmosphere of 5%CO2-95%air. The medium was changed at 48 hours. After 72 hrs in culture, the culture medium was removed and the cells washed with I ml of either RPMI-1640 or calcium, magnesium free Hank's balanced salt solution containing i mM EGTA (HBSS). The medium was then replaced with 1 ml RPMI-1640 or HBSS and the cells were incubated for an additional two hours. After removal of a 0.3 ml sample, ionophore A23187 (Calbiochem-Behring) or vehicle alone (DMSO) were added to the cells and the total volume readjusted to i ml. At
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
103
PALMITIC ACID 100
"E
E ~'500 rr o
ARACHIDONIC ACID
n~ 4 0 0 hi a Z D 300 IAl n~ 200
I00
0
I0 50 I00 FATTY ACID, .I.LM
FIGURE 2 The dose-dependent effect of arachidonic acid on FER in perifused placental cells. After forty minutes of perifusion, cells were exposed for 30 minutes to the indicated doses of arachidonic acid, palmitic acid or to vehicle alone. For each experiment, the ratio of experimental to control FER values were graphed as shown in figure IA. The bars represent the relative areas under these curves during the 30 minute exposure to arachidonic acid (~), palmitic acid ( [ ] ) , or vehicle as determined by planimetry, mean + S.E. (N=6 for 0 uM, N=3 for all other doses). *, P < 0.05; **, P < 0.01
various times, 0.3 ml samples were removed and the media replaced with an equivalent amount of medium containing the agent in the appropriate final concentration, hPL concentrations in all samples were determined by homologous radioimmunoassay (13). All experiments were performed in triplicate, and statistical differences between sample means were tested by analysis of variance followed by Dunnett's test. Results
After a 40 min pre-incubation period, exposure of placental cells for 30 minutes to 50 uM arachidonic acid stimulated a rapid increase in the release of 45calcium (fig IA) which was accompanied by a 17.8-fold increase in hPL release (fig IB). The fractional efflux rate (FER) increased approximately 5 minutes after the beginning of the arachidonic acid perifusion and remained elevated for the entire period of exposure. FER returned toward baseline within 5 minutes after the arachidonic acid perifusion was discontinued. Likewise, hPL release increased within 15 minutes after the start of the arachidonic acid exposure, remeained elevated throughout the exposure period, and decreased rapidly when the arachidonic acid perifusion was discontinued. In contrast, exposure of cells to vehicle alone had no effect on either the rate of 45calcium efflux or hPL release. In other control experiments, arachidonic acid had no effect on
104
Arachidonic
Acid, Calcium Flux and hPL
Vol. 38, No. 2, 1986
45calcium release from columns containing heat-inactivated taining no cells (data not shown).
cells or columns con-
The effect of arachidonic acid on FER and hPL release was dose-dependent. As shown in fig 2 and 3, cells exposed for 30 minutes to I0 uM and 50 uM arachidonic acid released 242% and 453% more 45Ca and 354% and 1279% more hPL than control cells. In addition, the effect of arachidonic acid on FER was dependent on the duration of exposure to the fatty acid. Decreasing the duration of arachidonic acid exposure to i0 or 20 min resulted in a corresponding decrease in the period of increased 45Ca efflux (fig 4).
'I
PALMIT~ACI IC~ D ARACHIDONI ACICD
140
uJlO-
09 <1 DJ
FIGURE 3
UJ Or
6O_
c-
2 0 I0 50 FATTY ACI D,.I.I.M
The dose-dependent effect of arachidonic acid on hPL release from perifused placental cells. After forty minutes of perifusion, cells were exposed for 30 minutes to the indicated doses of arachidonic acid ( I ) , palmitic acid ( ~ ), or to vehicle alone. The bars represent the ratio of hPL release per 30 minutes in the experimental columns compared to control columns perifused with media containing vehicle alone, mean + S.E. (N=3). *, P < 0.05; **, P-< 0.01
In contrast, palmitic acid had no effect on hPL release and only a small effect on 45Ca release, even at a concentration of i00 mM (fig 2,3). Since stimulation of hPL release by arachidonic acid was associated with an increase in 45Ca efflux, the effect of the calcium ionophore A23187 on hPL release was examined. As shown in fig 5, A23187 stimulated an increase in hPL release, with 2 uM and i0 uM resulting in 37.8 % and 130.9 % increases respectively in media containing 1.6 mM Ca 2+. A23187, however, had no significant effect on hPL release from cells incubated in calcium-free medium°
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
105
1.2
0
FIGURE 4
I.I
l.-
The effect of varying the duration of arachidonic acid exposure on FER in perifused placental cells. After forty minutes of perifusion, cells were exposed for i0 minutes (bar) to I0 uM arachidonic acid or to vehicle alone. Both columns were then exposed to control medium for 30 minutes, followed by an additional 20 minute exposure to arachidonic acid or vehicle and a final 30 minute perfusion in control medium. The results are expressed as the ratio of experimental to control FER values.
rv hl h
1.0 i ~ i i
III i l I 20 40
I
I
1
60
I
I
i
I
I
80 I00 MINUTES
i20
I
I
140
30
T 20 r-
-
nn .E
T "r
I0
,
,r
!
,
i
0
I
2
I0
0
I
2
I0
A 2318T, jJ.M FIGURE 5 The effect of ionophore A23187 on hPL release from placental cells. After 72 hours in culture, placental cells (106/well) were washed and incubated for 2 hours in either 1 ml RPMI ( 4 ) or calcium, magnesium free Hank's Balanced Salt solution ( J ). After removal of a 0.3 ml sample, ionophore A23187 or vehicle alone (DMSO) were added to the cells and the volume readjusted to 1 ml. After 2 hours, media were removed and assayed for hPL as described. Bars represent accumulted hPL ng, mean + S.E.M (N=4) ** P < 0 01.
106
Arachidonic
Acid, Calcium Flux and hPL
Vol. 38, No. 2, 1986
Discussion The results of this study indicate that arachidonic acid stimulates dosedependent, rapid and reversible increases in 45calcium and hPL release from an enriched fraction of hPL-secreting placental cells. The stimulation of 45calcium efflux in response to arachidonic acid was due to mobilization of calcium and not due to displacement of extracellular or non-specificially bound 45Ca since arachidonic acid had no effect on 45calcium release from heatinactivated cells or columns containing no cells The magnitude of the increase in 45calcium efflux and hormone release in placental cells responding to arachidonic acid is similar to the effect of arachidonic acid on 45calcium efflux and prolactin release from GH 3 cells of pituitary origin (14). An increase in 45calcium efflux may reflect a direct increase in the rate of flux of calcium across the plasma membrane or, alternatively, it may reflect a secondary response to a primary increase in the intracellular calcium concentration. Previous investigations have suggested that arachidonic acid may act to promote the passive diffusion of calcium across phospholipid membranes (2,4-7). Furthermore, arachidonic acid has been reported to promote the release of calcium from mitochondria and endoplasmic reticulum in liver and kidney cells (5). Therefore, arachidonic acid may stimulate the release of 45calcium from placental cells by stimulating a primary influx of calcium into the cytoplasm. Alternatively, arachidonic acid may promote release of calcium from intracellular stores. This change in calcium mobilization would then lead to an elevation of the intracellular free calcium concentration. Further clarification of this response will require the direct measurement of the intracellular free calcium concentration during exposure to arachidonic acid. Attempts to measure the intracellular free calcium concentration utilizing the fluorescent dye Quin2 have been unsuccessful due to marked autofluorescence of the fatty acid. The experiments with A23187 suggest that elevation of the intracellular calcium concentration stimulates an increase in hPL release. The calcium dependence of the response to A23187 indicates that the effect is due to changes in calcium distribution rather than an effect of A23187 itself. The source of the calcium mobilized by A23187 is not clear and may be either extracellular or intracellular. The stimulation of hPL release by arachidonic acid was similarly calcium-dependent (data not shown). Since the increase in calcium efflux induced by arachidonic acid most likely reflects an increase in intracellular calcium mobilization, the stimulation of hPL release by arachidonic acid may be due to the effect of the fatty acid on cellular calcium mobilization. Although the effect of arachidonic acid on hPL release is markedly diminished in calcium-free medium, the basal rate of hPL release is increased in calcium-free media (15 16). The reasons for this difference in the calcium dependence of basal and stimulated hPL release is not clear, but the observation suggests that hormone release under these two conditions may represent different cellular processes. In support of this suggestion, the increase in hPL release seen in calcium-free medium is transient and is followed by a period of decreased hormone release, while the release of hPL in response to arachidonic acid persists for many hours (2,3). We have recently reported that arachidonic acid stimulates the hydrolysis of phosphoinositides in an enriched fraction of hPL-secreting placental cells (submitted for publication). In these studies, phosphoinositide hydrolysis w a s demonstrated by an increase in the release of [3H]-myoinositol from perifused ' cells exposed to arachidonic acid, as well as the speciflc loss of 3 2 P r a d ioactivity from phosphatidylinositol and phosphatidylserine. Since phosphoinositide hydrolysis has been associated with calcium mobilization in a variety of cell
Vol. 38, No. 2, 1986
Arachidonic Acid, Calcium Flux and hPL
107
types (8,9), the effects of arachidonic acid on calcium metabolism described here may be due, at least in part, to the stimulation of phosphoinositide hydrolysis. Arachidonic acid stimulates the release of a variety of hormones, including prolactin~ luteinizing hormone, and histamine (14,17,18). In addition, arachidonic acid is released in response to receptor stimulation by a variety of secretagogues (19-22). These observations, together with the results presented here, suggest that the release of free arachidonic acid in response to receptoroccupancy may play an important role in the stimulation of hormone release by influencing cellular calcium metabolism. Furthermore, arachidonic acid has been reported to stimulate guanylate cyclase (23-25) and protein kinase C (26) suggesting that arachidonic acid release may link receptor occupancy to the stimulation of a number of intracellular processes associated with hormone secretion.
References I. S. HANDWERGER, T. HURLEY and A. GOLANDER, In Iffy, L. and H. Kaminetzky (eds.), P_rrin_ciplesandPracttice_of_Obstetrics an_d_Gynecol_ogy, Wiley and Sons, New York, pp 243-261 (1981). 2. S. HANDWERGER, J. BARRETT, S. BARRY, E. MARKOFF, P. ZEITLER, B. CWIKEL and M. SIEGEL, Mol. Pharmacol. 20 609-613 (1981). 3. P. ZEITLER, E. MARKOFF and S. HANDWERGER, J. Clin. Endocrinol. Metab. 57 812-818 (1983) 4. D. SELLER and W. HASSELBACH, Eur. J. Biochem. 21 385-387 (1971). 5. I. ROMAN, P. GMAJ, C. NORVIELA and S. ANGIELSKI, Eur. J. Biochem. 102 615-623 (1979). 6. M. VOLPI, P. NACCACHE and R. SHA'AFI, Biochem. Biophys. Res. Commun. 92 1231-1237 (1980). 7. A. CHEAH, Biochim. Biophys. Acta 648 113-119 (1981). 8. R. MICHELL, Biochim. Biophys. Acta 415 81-147 (1977). 9. R. FARESE, Metabolism 32 628-644 (1983). I0. K. REED, Biochem. Biophys, Res. Commun. 50 1136-1142 (1973). ii. M. BRADFORD, Anal. Biochem. 72 248-254 (1976) 12. T. UCHIKAWA and A. BORLE, Am. J. Physiol. 234 1234-1238 (1978). 13. S. HANDWERGER and L. SHERWOOD, In Jaffe, B. and H. Behrman (eds.), Methods 9fHormone_Radioi_mmun0assa~, Aca--demic Press, New York, pp. 417-426 (~74)~-14. R. KOLESNICK, I. MUSACCHIO, C. THAW and M. GERSHENGORN, Am. J. Physiol. 246 E458-E462 (1984). 15. S. HANDWERGER, P.M. CONN, J. BARRETT, S. BARRY and A. GOLANDER, Am. J. Physiol. 240 E550-E555 (1981). 16. V. CHOY and W. WATKINS, J. Endocrinol. 69 347-358 (1976). 17. Z. NAOR and K. CATT, J. Biol. Chem. 256 2226-2229 (1981). 18. A. NAKAO, A. BUCHANAN and P. POTOBAR, Int. Archs. Allergy. Appl. l~unol. 63 3 0 4 3 (1981). 1 9 R. BELL, N. BAENZIGER and P. MAJERUS, Prostaglandins 20 269-274 (1980). 20. B. HAYE, S. CHAMPION and C. JACQUEMIN, Febs. Letts. 30 253-260 (1973). 21. M. SCHREY and R. RUBIN, J. Biol. Chem. 254 11063-11066 (1980). 22. S. LAYCHOCK, Diabetes 32 6-13 (1983). 2 3 D. LEIBER and S. HARBON~ Mol. Pharmacol. 21 654-663 (1981). 24. D. WALLACH and I. PASTAN, J. Biol. Chem. 251 5802-5809 (1976). 25. R. BRIGGS and F. DERUBERTIS, Biochem. Pharmacol. 29 717-722 (1980). 26. L. MCPHAIL, C. CLAYTON and R. SNYDERMAN, Science 224 622-625 (1984).