Enhancement of ornithine decarboxylase activity in Ambystoma liver slices by ovine prolactin: An evaluation of possible mediators

GENERAL

AND

COMPARATIVE

ENDOCRINOLOGY

72, 9C96 (1988)

Enhancement of Ornithine Decarboxylase Activity in Ambystoma Liver Slices by Ovine Prolactin: An Evaluation of Possible Mediators RAGAA H. IBRAHIMANDJAMES Department

of Biological

E. PLATT

Sciences, University of Denver, Denver, Colorado 80208 Accepted May 11, 1988

Prolactin has been shown to increase the activity of omithine decarboxylase in a variety of mammalian tissues and in the pigeon crop sac. This study demonstrates a similar effect of ovine prolactin on omithine decarboxylase activity in liver slices taken from larval tiger salamanders (Ambystoma tigrinum). An evaluation of potential mediators of prolactin action in liver slices revealed that the effect of the hormone on enzyme activity was not blocked by ouabain, an inhibitor of the sodium pump reported to block other actions of prolactin. Oxytocin, which inhibits prolactin actions in A. tigrinum, blocked the increase in omithine decarboxylase activity induced by prolactin. Since previous results had implicated inositol phospholipid turnover in oxytocin action, the effects of the calcium ionophore, A 23 187, and of synthetic diacylglycerol were examined. Both agents blocked the increase in enzyme activity when they were combined with prolactin treatment. Verapamil, a calcium channel blocker, had a prolactin-like effect on the activity of omithine decarboxylase, and the combination of prolactin and verapamil produced a stimulation of the enzyme that was no greater than that observed with either the drug or prolactin alone, suggesting that both agents might be acting via a common cellular pathway. The tentative hypothesis that prolactin acts via a mechanism which lowers intracellular calcium is suggested. o 1988 Academic Press, Inc.

The diversity of the biological effects of prolactin (PRL) is a well-known and wellstudied phenomenon (e.g., see Nicoll and Bern, 1972; Nicoll, 1974; Clarke and Bern, 1980). Comparative endocrinologists have long pondered whether these many actions of PRL could be explained in terms of a single, universal mechanism of cellular action of PRL. Despite years of investigation by many workers using a variety of systems, the nature of the signal(s) which mediates cellular responses to PRL remains uncertain (see the recent volume edited by Rillema (1987a) for a review of these efforts). Although many signal molecules have been implicated in the mediation of PRL action (e.g., cyclic nucleotides, polyamines, prostaglandins, the sodium pump, a putative peptide produced by PRLinduced cleavage of the PRL receptor, and-most recently-the turnover of mem-

brane inositol phospholipids), none of these agents has been unequivocally established as a mediator of PRL action (Rillema, 1987a). In the search for a “common denominator” in PRL action, attention has recently been focused upon the induction of an increase in the activity of ornithine decarboxylase (ODC), the rate-limiting enzyme in cellular polyamine synthesis, by PRL. This response to the hormone occurs in a variety of target tissues within a few hours of exposure to PRL (Richards, 1975; Rillema, 1985; Russell et al., 1987). Although most of the studies have been done using mammalian tissues, a report by Horseman and Nolin (1985) has provided evidence that PRL can increase the activity of ODC in the pigeon crop sac. While the relationship between the rise in ODC activity and the subsequent effects of PRL re90

0016-6480/88 $1.50 Copyright 0 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.

PROLACTIN

AND

ORNITHINE

mains obscure, the effect on the enzyme is consistent and widespread. In the current study we have extended the demonstration of the occurrence of this response to Ambystoma tigrinum, a urodele amphibian. In addition, we have used the induction of ODC in liver slices of this species to investigate the nature of the cellular pathway which mediates this effect of PRL. Our results suggest that the rise in ODC activity induced in A. tigrinum liver by PRL may be mediated by a reduction in intracellular calcium, perhaps via a mechanism that reduces the turnover of membrane inositol phospholipid. MATERIALS

AND METHODS

Larval A. tigrinum used in these experiments were obtained from a local bait dealer in Denver, Colorado. The animals were kept in a constant temperature cold room (4 2 0.5”) prior to use. Larvae (10-12 animals per experiment) were removed from the cold room 4 days before each experiment. After 24 hr of acclimation to room temperature (22”), these animals were injected with ergocornine (Sandoz Pharmaceuticals) for 3 days (50 pg/animal/day) in order to suppress endogenous PRL levels (Piatt, 1976). On the day of each experiment, the animals to be used were anesthetized in 0.33 g/liter MS-222 (tricane methanesulfonate, Sigma Chemical Co., St. Louis, MO) and were sacrificed by decapitation. The larvae were then placed on ice and each animal’s liver was removed as rapidly as possible. Each liver was cut into equally sized portions (two, three, or four depending upon the experiment and the size of the animals used) and was minced with a razor blade and placed in amphibian Ringer’s buffer (ARB) (Frieden and Campbell, 1978). All tissue incubations were done in 35 x lo-mm plastic culture dishes containing 4 ml of ARB with treatments as described below. The dishes containing the liver slices were placed on a clinical rotator for 3 hr at 22”. The ODC assay was conducted using a modification of the procedure described by Endo (1983). This procedure utilizes ion exchange chromatography to separate polyamines from tissue homogenates on phosphocellulose columns, followed by elution of these products of ODC action and quantitation using spectrofluorometry. Tissues from the incubations are homogenized in 2.0 ml of an ice-cold phosphate buffer and the homogenates are treated with phosphocellulose powder to removed endogenous polyamines. The homogenates are then centrifuged at 15,OOOgfor 20-30

DECARBOXYLASE

IN

Ambystoma

91

min at 4”. The supernatant is reserved for assay of ODC and a 25-ul aliquot is removed for subsequent determination of protein content. The remaining supematant is combined with 1 .Oml of an assay mixture containing 0.2 M NaH,PO,, 0.2 M Na,HPO,, 0.5 mM dithiothreitol, 0.05 mMpyridoxal5-phosphate, and 1.O mM L-omithine, pH 6.7. The reaction is carried out at 20” (the temperature determined to be optimum for this species’ enzyme) for 3 hr and is terminated by the addition of perchloric acid. The reaction mixture is then neutralized with KOH and centrifuged at 75OOg for 10 min. The supematants are poured onto 3 X 0.6 cm phosphocellulose columns prepared for cationexchange chromatography. The columns are washed with borate buffer (0.2 M, pH 8.5) in 0.025 M NaCl to remove monoamines, followed by elution by polyamines with a high-salt wash (0.2 M borate buffer in 0.6 M NaCl). The polyamines in the eluate are quantified by the addition of fluorescamine (Sigma Chemical Co.) and the determination of relative fluorescence in a spectrofluorometer at an excitation wavelength of 390 nm and an emission wavelength of 475 nm. A standard curve using synthetic putrescine (Sigma Chemical Co.) is prepared for each experiment. The protein content of each supernatant is determined using the Hartree modification of the Lowry procedure (Hartree, 1972). The results are expressed as nmol polyamine/mg proteini30 min. Statistical comparisons of group means are made using the Student t test (two groups) or the one-way analysis of variance followed by the Student-Neuman-Keuls multiple range test for significance (more than two groups), with significance accepted at the 0.05 level of probability (Zar, 1974). In all experiments reported, ovine PRL (NIADDKoPRL-17) at a final concentration of 1.0 p&nl was used and PRL-treated liver slices were compared to slices incubated in ARB alone. Additional treatments were oPRL + ouabain (10 PM; Experiment 2); oPRL + oxytocin (1 mu/ml; Experiment 3); oPRL + diacylglycerol (1-oleoyl-2-acetyl-rat-glycerol, 25 pM; Experiment 4); oPRL + A 23 187 and A 23 187 alone (1 pM; Experiment 5); and oPRL + (?)-verapamil and verapamil alone (10 p,M; Experiment 6). Both diacylglycerol and A 23187 were dissolved in dimethyl sulfoxide (DMSO) and added in 20 pl of this solvent; consequently, 20 pl of DMSO alone was added to the other groups in Experiments 4 and 5. All chemicals were obtained from Sigma Chemical Co. In Experiments 2, 3, and 4, it was possible to obtain only three portions of liver per replicate due to the smaller size of the animals available for these experiments. Therefore, separate control experiments were subsequently done to test for any effects of ouabain, oxytocin, or diacylglycerol alone on ODC activity. Ten to 12 replicates were done per experiment for each treatment and ODC assays were performed on each portion of liver as described.

92

IBRAHIM

RESULTS

The effect of 1 yg/ml of oPRL on ODC activity in A. tigrinum liver slices is illustrated in Table 1. As can be seen, the specific activity of the enzyme is significantly increased after a 3-hr incubation with the hormone. This same stimulatory effect was observed in all subsequent experiments (see Table 2). Results obtained from experiments designed to examine possible mediators of the action of PRL on hepatic ODC are summarized together in Table 2. Ouabain, an inhibitor of the sodium pump (i.e., the Na+/K+-ATPase), did not alter the stimulatory effect of oPRL on ODC activity. In a separate control experiment, ouabain alone had no effect on enzyme activity (data not shown). Oxytocin-a neurohypophysial hormone which had previously been shown to inhibit other actions of PRL in larval A. tigrinum (Platt and LiCause, 1980; Platt et al., 1986)-blocked the increase in ODC activity when it was added to the culture medium along with oPRL (Experiment 3). A separate control experiment showed no effect of oxytocin alone on ODC activity (data not shown). Two drugs known to duplicate the effects of enhanced inositol phospholipid turnover were tested in Experiments 4 and 5. Synthetic diacylglycerol abolished the stimulatory effect of oPRL on ODC activity. This drug was without effect on the activity of TABLE THE

1

EFFECT OF OVINE PROLACTIN DECARBOXYLASE ACTIVITY IN LIVER SLICES

ON ORNITHINE

Ambystoma

Treatment

N

Mean specific activity of ornithine decarboxylase (nmol polyamine/mg protein/30 min 2 SEM)

None Prolactin

11 11

0.088 rtr 0.0131 0.341 + 0.0438”

a Significantly

different from control; P < 0.001.

AND

PLATT

the enzyme when tested by itself (data not shown). A 23187-a calcium ionophore which is assumed to act by elevating cytosolic free calcium, thus mimicking the action of inositol triphosphate-blocked PRL action. A 23187 alone appeared to lower ODC activity, but this difference was not statistically significant compared to control liver slices. Verapamil, a drug known to block the entry of calcium into cells through calciumspecific channel proteins, elevated liver ODC activity by itself to the same extent as oPRL. When verapamil and oPRL were combined, the increase in ODC activity was the same as that observed with either treatment alone. This may imply that the actions of these two molecules were not additive at the concentrations used. DISCUSSION

These results clearly show that oPRL at a concentration of 1 kg/ml elevates the activity of ODC in A. tigrinum liver slices. This effect is similar to the effect of PRL on ODC activity observed in a variety of mammalian tissues (Richards, 1975; Russell et al., 1984; Rillema, 1985) and in the pigeon crop sac (Horseman and Nolin, 1985). Initially, we evaluated the potential role of the cellular sodium pump in PRL action, as Falconer and Rowe (1975, 1977) proposed that some actions of PRL in mammary explants were mediated by an activation of this membrane pump. They had provided indirect evidence for this hypothesis by showing that ouabain-a specific inhibitor of the pump-could block some PRL actions in mammary tissue. However, other workers provided data suggesting that ouabain reduced the binding of iodinated PRL to mammary tissue (Houdebine and Djiane, 1980). Since we had observed that ouabain appeared to increase the binding of iodinated oPRL to A. tigrinum skin strips in vitro (Rao, 1986), we determined the effect this drug had on PRL

PROLACTIN

AND

ORNITHINE

DECARROXYLASE

TABLE EFFECTS

OF VARIOUS

DRUG AND HORMONE TREATMENTS DECAR~OXYLASE ACTIVITY IN

Treatment Experiment 2 None Prolactin Prolactinlouabain Experiment 3 None Prolactin Prolactin/oxytocin Experiment 4 None Prolactin Prolactinidiacylglycerol Experiment 5 None Prolactin A 23187 ProlactinlA 23 187 Experiment 6 None Prolactin Verapamil Prolactiniverapamil a Significantly b Significantly ’ Significantly ’ Significantly 0.001. e Significantly

different different different different

from from from from

control; control control control,

IN Ambystoma

93

2 ON THE PROLACTIN-INDUCED SLICES

RISE IN ORNITHINE

Ambystoma LIVER

N

Mean specific activity of ornithine decarboxylase (nmol polyamine/mg protein/30 min i. SEM)

10 10 10

0.233 +- 0.0304 0.510 * 0.0883” 0.515 +- 0.1216”

10 10 10

0.058 rt 0.0117 0.183 i 0.0241’ 0.075 2 0.0308

11 11 11

0.343 i 0.0207 0.491 * 0.0535” 0.290 +- 0.0328

12 12 12 12

0.079 0.195 0.036 0.056

i 0.0104 -+ 0.0432d -+ 0.0118 t 0.0187

10 10 10 10

0.062 0.208 0.221 0.188

k + I i

0.0141 0.0473’ 0.0442’ 0.0373’

F = 3.65, & = 2,9, P < 0.05. and from prolactin/oxytocin; F = 8.29, df = 2,9, P < 0.01. and from prolactinidiacylglycerol; F = 7.28, df = 2,10, P < 0.01. from A 23187, and from prolactin/A 23187; F = 8.21, df = 3,11, P <

different from control; F = 3.68, df = 3,9, P < 0.05.

acticln in the liver system. The results of Experiment 2 suggest that ouabain did not block PRL action, and a separate control experiment revealed no effect of ouabain alone on ODC activity. Thus, we conclude that the elevation of ODC activity by PRL in A. tigrinum liver does not involve the sodium pump. In earlier studies of PRL action in A. tigrinurn, we observed that many of its effects could be antagonized by oxytocin (Platt and LiCause, 1980, 1981; Platt et al., 1986). Further, we found that water loss induced in blocks of A. tigrinum tail fin tissue by oxytocin was likely to involve an elevation of cellular calcium mediated by the activation of membrane inositol phospholipid

turnover (Platt et al., 1987). Since PRL antagonized this action of oxytocin, we investigated the interaction of PRL and oxytocin in liver slices and the potential involvement of calcium and inositol phospholipid turnover. Experiment 3 demonstrated that oxytotin did antagonize the stimulation of ODC activity by PRL. Again, a separate control experiment showed no effect of oxytocin alone. These results were consistent with our earlier observations and suggested that-if oxytocin were acting via a calciumdependent pathway and were stimulating inositol phospholipid turnover as we had proposed (Platt et al., 1987)-then the phospholipid pathway might be involved in

94

IBRAHIM

the antagonism of PRL action by oxytocin in liver slices as well. The enhanced turnover of membrane inositol phospholipid has recently been implicated as a signal-generating mechanism in a number of cellular responses to ligand binding to cell surface receptors (e.g., see Nishizuka et al., 1984). Briefly, receptormediated cleavage of inositol phospholipids generates inositol phosphates (particularly inositol triphosphate) and diacylglycerol which both act as second messengers. Inositol triphosphate acts to release calcium ions from intracellular storage sites while diacylglycerol activates protein kinase C. The action of inositol triphosphate can be mimicked by agents like the calcium ionophore, A 23187. The action of endogenous diacylglycerols can be duplicated by various synthetic analogs of these compounds. In the previously cited study (Platt et al., 1987), we observed that both A 23187 and a synthetic diacylglycerol were able to mimic oxytocin action in blocks of tail fin tissue. We therefore examined their effects on PRL action in liver slices. As indicated by the results of Experiments 4 and 5, both diacylglycerol and A 23 187 blocked PRL elevation of ODC activity. A 23 187 was without effect by itself and a separate, unreported control experiment showed diacylglycerol alone to be without effect on ODC activity as well. Thus, it appears that raising intracellular calcium levels with A 23187 or activating protein kinase C with diacylglycerol results in the activation of some intracellular effector system(s) which blocks the activation of ODC by PRL. In the same report where we observed the effects of diacylglycerol and A 23187 on tail fin tissue (Platt et al., 1987), we also found that the calcium channel blocker, verapamil, inhibited oxytocin action. In view of this and of the above results, we examined the effects of verapamil on PRL action in liver tissue. Verapamil alone has a PRLlike effect on ODC activity. Furthermore, when both verapamil and oPRL were used

AND

PLATT

together, the stimulation of ODC levels was equivalent to that seen with either agent alone. Thus, it is possible that both are acting via the same cellular pathway. We assume that the effect of verapamil is to block the entry of calcium ions into liver cells and that it thereby lowers intracellular calcium, but we have no direct measurements of calcium levels to support this assumption. Such an interpretation is consistent with other results we have obtained, however. Both Rillema (1985, 1987b) and Russell et al. (1987) concluded that the activation of ODC by PRL in mammalian tissues was probably mediated by enhanced inositol phospholipid turnover, while our data suggest an inhibition of this pathway by PRL. However, a close examination of their results suggests that further evaluation of all these systems will be required before firm conclusions concerning the role of inositol phospholipids in PRL action can be made. While Rillema (1985) did provide data showing that both A 23187 and TPA (a phorbol ester which is thought to act as a diacylglycerol analogue) had PRL-like effects on ODC in mammary cells, it is relevant to note that neither drug alone nor a combination of the two had an effect that was comparable in magnitude to the effect of PRL alone. Further, although Rillema suggested that the nonadditive nature of the responses to A 23187, TPA, and PRL indicated that they were acting via a common pathway, examination of his data suggests that-when either the ionophore or TPA is combined with PRL treatment-the magnitude of the ODC activation is actually lower than that observed with PRL alone. This could be interpreted as an inhibitory effect of these two drugs on PRL action. The results obtained by Russell et al. (1987) are also equivocal. They examined the effects of PRL on ODC activity in Nb 2 lymphoma cells, a rat cell line that is absolutely dependent upon PRL for proliferation. In this system, it is clear that the effects of TPA and PRL on ODC activity were additive, as were the effects of A 23 187 and PRL. While

PROLACTIN

AND

ORNITHINE

TPA alone activated the enzyme, the effects of A 23187 alone on ODC were marginal and the ionophore actually inhibited cell proliferation which is regarded as the ultimate response to PRL in Nb 2 cells. Russell et al. (1987) also showed that inhibitors of protein kinase C could block the induction of ODC by PRL, but they did not evaluate the specificity of these inhibitors. In summary, our results show that PRL elevates ODC activity in A. tigrinum liver tissue as it does in the tissues of other vertebrate species examined. This action is not affected by inhibition of the sodium pump. It is blocked by oxytocin, A 23187, and diacylglycerol. The action of PRL is duplicated by verapamil. Collectively, these resuits suggest that PRL action in Ambystoma liver may be mediated by a mechanism that involves the lowering of intracellular calcium levels, perhaps via a reduction in the turnover of membrane inosito1 phospholipid. These results appear to differ from those obtained using mammalian tissues, where the data implicated an enhanced turnover of inositol phospholipid in the activation of ODC by PRL. Both direct and indirect effects of PRL on calcium levels and inositol phospholipid turnover are plausible; elucidation of the precise celMar pathways involved and explanation of species differences will require further investigation. ACKNOWLEDGMENTS The authors thank the National Institutes of Arthritis, Diabetes, Digestive, and Kidney Diseases for the gift of the two pituitary hormones and Sandoz Pharmaceuticals for the gift of the ergocomine.

REFERENCES Clarke, W. C., and Bern, H. A. (1980). Comparative endocrinology of prolactin. In “Hormones and Hormonal Peptides” (C. H. Li, Ed.), Vol. 8, pp. 105-197. Academic Press, New York. Endo, Y. (1983). A simple method for the determination of polyamines and histamine and its application to the assay of omithine and histidine decar-

DECARBOXYLASE

IN

Ambystoma

95

boxylase activities. In “Methods in Enzymology” (H. Tabor and C. W. Tabor, Eds.), Vol. 94, Polyamines, pp. 42-47. Academic Press, New York. Falconer, I. R., and Rowe, J. M. (1975). Possible mechanism for the action of prolactin on mammary cell sodium transport. Nature (London) 256, 327-328. Falconer, I. R., and Rowe, J. M. (1977). Effect of prolactin on sodium and potassium concentrations in mammary alveolar tissue. Endocrinoiogy 101, 181-186. Frieden, E., and Campbell, J. A. (1978). The effect of triiodothyronine on transport and increase of amino acids by bullfrog tadpole fin cells. Gen. Comp. Endocrinol. 36, 215-222. Hartree, E. F. (1972). Determination of protein: A modification of the Lowry method that gives a linear photometric response. Anal. Biochem. 48, 422427. Horseman, N. D., and Nolin, L. J. (1985). The mitogenie, but not differentiative, response of pigeon crop to prolactin is circadian phase-dependent. Endocrinology 116, 2085-2089. Houdebine, L. M., and Djiane, J. (1980). Effet de I’ouabaine sur l’action lactogbne de la prolactine et sur la niveau des recepteurs prolactiniques mammaries. Biochimie 62, 433-440. Nicoll, C. S. (1974). Physiological actions of prolactin. “Handbook of Physiology: Section 7: In Endocrinology” (E. Knobil and W. H. Sawyer, Eds.), Vol. 9, Part 2, pp. 253-292. Amer. Physiol. Sot., Washington, D. C. Nicoll, C. S., and Bern, H. A. (1972). On the actions of prolactin among the vertebrates: Is there a common denominator? In “Lactogenic Hormones” (G. E. W. Wolstenholme and J. Knight, Eds.), pp. 299-324. Churchill-Livingstone, Edinburgh/London. Nishizuka, Y., Takai, Y., Kishimoto, A., Kikkawa, U., and Kaibuchi, K. (1984). Phospholipid turnover in hormone action. Recent Prog. Hormone Res. 40, 301-345. Platt, J. E. (1976). The effects of ergocomine on tail fin height, spontaneous and T,-induced metamorphosis and thyroidal uptake of radioiodide in neotenic Ambystoma tigrinum. Gen. Camp. Endocrinol. 28, 71-81. Platt, J. E., Brown, G. B., Erwin, S. A., and McKinley, K. T. (1986). Antagonistic effects of prolactin and oxytocin on tail tin regression and acid phosphatase activity in metamorphosing Ambystoma tigrinum. Gen. Comp. Endocrinol. 61, 376-382. Platt, J. E., and LiCause, M. J. (1980). Effects of oxytocin in larval Ambystoma tigrinum: Acceleration of induced metamorphosis and inhibition of the antimetamorphic action of prolactin. Gen. Comp. Endocrinol. 41, 84-91.

96

IBUHIM

Plan, J. E., and LiCause, M. J. (1981). Oxytocininduced water loss in Ambystoma tissue in vitro. Amer.

Zool.

21, 912.

Platt, J. E., Rao, A. K., Angele, K., and LiCause, M. J. (1987). Oxytocin-induced water loss in Ambystoma tissue: An evaluation of the mechanism of oxytocin action in larval amphibians. Life Sci. Adv. Part A, in press. Rao, A. K. (1986). “An Examination of Radiolabeled Prolactin Binding to Tail and Liver Tissues ofAmbystoma tigrinum.” Ph.D. thesis, University of Denver. Richards, J. F. (1975). Ornithine decarboxylase activity in tissues of prolactin-treated rats. Biochem. Biophys.

Res.

Commun.

63, 292-299.

Rillema, J. A. (1985). Prolactin stimulation of ornithine decarboxylase activity in mammary gland may involve an activation of protein kinase C. Proc.

Sot.

Exp.

Biol.

Med.

178, 490-494.

Rillema, J. A. (1987a). The activation

of molecular

AND

PLATT

events by prolactin. In “Actions of Proiactin on Molecular Processes” (J. A. Rillema, Ed.), pp. l-4. CRC Press, Boca Raton, FL. Rillema, J. A. (1987b). Role of calcium ions and phospholipids in the prolactin regulation of its target cells. In “Actions of Prolactin on Molecular Processes” (J. A. Rillema, Ed.), pp. 107-119. CRC Press, Boca Raton, FL. Russell, D. H., Buckley, A. R., Montgomery, D. W., Larson, N. A., Gout, P. W., Beer, C. P., Putnum, C. W., Zukoski, C. F., and Kibler, P. (1987). Prolactin-dependent mitogenesis in Nb 2 node lymphoma cells: Effects of immunosuppressive cyclopeptides. .I. Immunol. 138, 276-284. Russell, D. H., Larson, D. F., Cardon, S. B., and Copeland, J. G. (1984). Cyclosporine inhibits prolactin induction of ornithine decarboxylase in rat tissues. Mol. Cell. Endocrinol. 3.5, 159-166. Zar, J. H. (1974). “Biostatistical Analysis,” Prentice Hall, Englewood Cliffs, NJ.