- Email: [email protected]
Production of prostaglandins by dispersed cells and fragments from bovine parathyroid glands
PROSTAGLANDINS
PRODUCTION
OF PROSTAGLANDINS BY DISPERSED CELLS AND FRAGMENTS FROM BOVINE PARATHYROID GLANDS
Edward M. Brown, M.D. and Stephen L. Swartz, M.D. Endocrine-Hypertension
Unit, Brigham and Women's Boston, MA 02115.
Hospital,
ABSTRACT We studied the production of prostaglandins by fragments and dispersed cells from bovine parathyroid glands. Fragments released 138±19 (SE), 132±21, 4.3±0.5, and 13+6.6 pg/mg/h of 6-keto-PGFl~ , PGF2~ , PGE2, and thromb~xane B~, respectively (n=7-26), while dispersed . cells releasea 414±110, 22~7.3, 27~3.8, and 29±11 pg/10 ° cells/h, respectively, of the same compounds (n=6-25). Indomethacin (i ~g/ml) inhibited the release of 6-keto-PGFl~ by 80-90% in fragments and cells, while mellitin stimulated release of this prostaglandin, suggesting de novQ synthesis of prostaglandins in these preparations. Calcium stimulated production of 6-keto-PGF 1 ~ by 1 3-fold in cells and 2.6-fold in fragments and also enhanced production of PGF2~ by 1.9-fold in fragments. Isoproterenol, on the other hand, had no effect on production of 6-keto-PGFl~ in either preparation. These results demonstrate that parathyroid tissue as well as parathyroid cells per se produce a variety of prostaglandins. We have previously shown that PGE 2 and PGF2~ modulate cAMP accumulation and PTH release in dispersed bovine parathyroid cells. The role of the endogenous production of prostaglandins by the parathyroid gland in the acute or chronic regulation of parathyroid function, however, remains to be determined. INTRODUCTION A number of hormonal factors modify cAMP accumulation and immunoreactive PTH (iPTH) release in dispersed bovine parathyroid cells (1-3). These include various catecholamines, secretin, and the prostaglandins PGE 2 and PGF2 ~. PGE 2 stimulates (4), while PGF2~ inhibits (5), both cAMP accumulation and PTH release. It is possible, therefore, that the production of prostaglandins locally within the parathyroid gland might play some physiological role in the regulation of parathyroid function. The production of prostaglandins by parathyroid tissue, however, has not been reported previously. For this reason, we have studied the production of 6-keto-PGFl~ (the stable, nonenzymatic degradation product of PGI2) , PGF 2 , PGE 2, and thromboxane B 2 by fragments and dispersed cells
JANUARY 1985 VOL. 29 NO. 1
35
PROSTAGLANDINS from bovine parathyroid glands. MATERIALS AND METHODS Dispersed bovine parathyroid cells were prepared by digestion of minced parathyroid tissue from one to three week old calves with collagenase and DNase, as described previously (6). After dispersal, the cellular pellet was sedimented for 5 minutes at 50 x g and resuspended several times in Minimal Essential Medium (Earle's Salts( without NaHCO 3, MgCI 2, or CaCI2), containing 20 mM HEPES ± pH 7.47 ("standard medium") as well as 0.5 mM MgS04, 1.5 mM CaCI2, and 2% bovine serum albumin (BSA). The cells were then maintained in the same medium at 37 ° until use. Cellular viability, as assessed by trypan blue exclusion, was routinely 95-100%. Cell counts were performed with a hemocytometer. 90-95% of the cells were parathyroid cells by light and electron microscopical criteria; most of the remaining cells were red blood cells. Incubations were carried out at 37 ° in disposable 5 cc polypropylene scintillation vials (Starstedt, Princeton, N.J.) using 0.5-2.0 cc of standard medium with 0.2% BSA and varying concentrations of MgSO 4 and CaC~2 as detailed below. Incubations contained 2-10 x 10 dispe~sed cells/cc or 10-40 mg wet weight/cc of fragments (1-2 mm ~ each) of parathyroid glands. Supernatant samples for subsequent determination of PTH or prostaglandins were stored frozen after allowing fragments to settle at unit gravity or sedimenting the cellular pellet at 250 x g for 2 minutes in a desk-top centrifuge (Sorvall GLC-2B). In experiments testing the effects of PGI 2 on PTH_release, the prostaglandin was dissolved at 10-3M in standard medium at pH 10 and was diluted 100-fold into incubation vials containing parathyroid cells just prior to the experiment. Radioimmunoassay for PTH was carried out as described previously (6), with a sheep antibovine parathyroid hormone antiserum (GW-I) recognizing the intact hormone as well as the C-terminal portion of the molecule. All samples from a given experiment were measured in the same assay. Calcium, magnesium, EGTA, and isoproterenol, at the concentrations used experimentally, had no effect on the radioimmunassay for PTH. i.
The following abbreviations have been employed: HEPES Hydroxyethylpiperazine ethanesulfonic acid; BSA Bovine serum albumin; EGTA - Ethyleneglycol-bis-(8-aminoethylether)-N, Nl-tetraacetic acid; 6-keto-PGFl~-6-ketoprostaglandin FI~; TxB 2 - Thromboxane B2; PGF2~- Prostaglandin F2~; PGE 2- Prostaglandin E 2, -
-
36
JANUARY 1985 V O L 29 NO. 1
PROSTAGLANDINS
6-keto P G F I % P G F ~ , TxB2, and PGE 2 were measured by the double antiSody raaioimmunoassay technique ~f Levine (7), as modified by Swartz (8). Briefly, rabbit anti-sera to 6-keto-PGF1~ , PGF2~, TxB 2 and PGE 2 were raised by immunizing raSbits with the prostaglandin coupled to human albumin. 200 ~i aliquots of supernatants from incubations of the parathyroid fragments or cells were added to 100 ~i ~f rabbit anti-PG antibody (in tris buffer) and 100 ~i of H-PG. The mixture was incubated at 37 ° C for 1 hour, after which 100 ~i of normal rabbit serum and 100 ~i of goat antl-rabbit globulin were added. After thorough mixing, the reaction mixture was incubated at 4 ° C overnight. The antibody-bound 3H-PG was precipitated by centrifugation and assayed for 3H by liquid scintillation spectrophotometry against a standard curve. In these radioimmunoassays, 6-keto-PGFle , PGF2~, TxB 2 and PGE 2 cross react with each other less than 1% (8). The lower limit of sensitivity for 6-keto-PGFl~ and TxB 2 is 10 pg/ml, and for varlationPGF~ and PGE 2 i s<10%20 pg/m(9)_ 1 To (8).evaluateThe intraassaYfurthertheC°efficientvalidityof and accuracy of the radioimmunoassay techniques, we added 44 pg of 6-keto-PGFl~ to eight different samples (i ml each); analysis of these specimens by radioimmunoassay showed a recovery of 101~7 percent (mean~SEM). Group means have been presented with the standard error of the mean as the index of dispersion. To determine whether significant changes in PG occurred with increasing amounts of extracellular calcium, the ANOVA computer program was used. This program computes the t value for the response at each calcium concentration and compares the appropriate control with the pooled variance. Homogeneity of variance was assessed by Bartlett's test (10). P values were obtained in Dunnett's tables for comparing multiple responses with a single control (ii). Other statistical analyses were carried out using the Fisher Exact Test. RESULTS Production of Drostaulandins bv fraqments ~nd dispersed cells from bovine parathyroid uiands. In preiiminary experiments, release of 6-ke£o-PGF]~ by fragments and dispersed parathyroid cells was linear for at least one hour and two hours, respectively. In subsequent experiments, therefore, production of prostaglandins was studied at 30 to 60 minutes with fragments and 30 to 120 minutes with cells. Fragments of bovine parathyroid glands released 138~19 (SE), 132~21, 4.3±0.5, and 13±6.6 pg/mg/h of 6-keto-PGFl~ , PGF2~, PGE2, and TxB2, respectively (n=7-26); dispersed cells, on the other hand, produced 4141110, 2217.3, 27~3.8, 29111 pg/106 cells/h, respectively, of the same compounds (n=6-25) (Figure i).
JANUARY 1985 VOL. 29 NO. 1
37
PROSTAGLANDINS
,ooFL
CELLS ~] FRAGMENTS
w£ u) 500f W
~
l:~ Z
o.
J
~
~ ~
~ _--
foo l30'
t
LI
I0
o
0
6- KETO PGFI~
Figure 1
38
PGF2(~
PGE2
Tx 82
Production of prostaglandins by fragments and dispersed cells from bovine parathyroid glands. Dispersed cells (open bars) or fragments (stipsled bars) were incubated for 30-120 minutes at 37 ° in standard medium with 0.2% BSA, 1.0 mM CaCI 2 and 0.5 mM MgSO 4. Results indicate the mean±SEM for 6-25 observations with cells and 7-26 observations with fragments, each measured in duplicate, expressed as release per hour.
JANUARY 1985 VOL. 29 NO. 1
PROSTAGLANDINS
Because 6-keto-PGFl~ was produced in greatest quantity by both cells and fragments, most subsequent studies were carried out using this prostaglandin to assess the effect of various agents on prostaglandin production. Effects of indomethacin or mellitin on Drostaulandin p _ ~ . Indomethacin, an inhibitor of cyclooxygenase, inhibited the production of 6-keto-PGFl~ by 91% and 78% in fragments of bovine parathyroid gland and dispersed bovine parathyroid cells, respectively (Figure 2A). In contrast, mellitin, a membrane-active peptide of bee venom which stimulates the biosynthesis of PG by interacting with phospholipid membrane bilayers, resulting in arachidonic acid release and PG production (12), stimulated the production of 6-keto-PGFl~ by 2.6- and 5.7-fold in fragments and cells, respectzvely (Figure 2B). These results suggest that both fragments of bovine parathyroid tissue and dispersed bovine parathyroid cells produce prostaglandins de novo. Effects of extracellular calcium on the production of 6-keto-PGFl~. Raising extracellular calcium from <10 -8 M 2 to 2 mM pr~-duced a 1.3-fold stimulation of the production of this prostaglandin in cells (Figure 3A) and a 2.6-fold stimulation of the production of 6-keto-PGF ~ by fragments of bovine parathyroid tissue (Figure 3B). ~he same change in extracellular calcium stimulated the production of PGF2~ by 1.9-fold in fragments (from 140~47 to 257~35 pg/mg/hr)7 All of these changes were statistically significant
(p<.01). Effect of isoproterenol on 6-keto-PGFl~" Droduction in cells ~ U I ~ . Isoproterenol had no ~-ffect on the production of 6-keto-PGFl~ by dispersed bovine parathyroid cells (Fig. 4A). In fragments of bovine parathyroid glands, this agonist caused a slight but statistically insignificant increase in the production of 6-keto-PGFl~ (Figure 4B). Effect of PGI2. on PTH release in dispersed bovine parathyroi4 c~lls We have not previously examined the effects of prostacyclin (PGI2) on PTH release. This agent_had no effect on PTH release at doses ranging from 10 -~ to 10 -5 M and at times from 5 to 30 minutes (not shown). Because of the lability of PGI2, we also added PGI 2 at 5 minute intervals in a 30 mznute experiment but likewise found no effect under these conditions. 2.
This is the approximate free calcium concentration 1 mM EGTA and no added CaCI 2 at pH 7.5.
JANUARY 1985 VOL. 29 NO. 1
with
39
PROSTAGLANDINS
~A. rn
mr*
5O0 ,~
-/-
~
- 9OO
200
~
- 600
oo
~
300
o
o°
200
250
Figure 2
40
o~--CONTROL
o
/ N
,oo ~
g
'
+ INDO
CONTROL
+ MELL
Effects of indomethacin or mellitin on production of 6-keto-PGFl~. Dispersed cells (Panel A) or fragments from bovine parathyroid glands (Panel B) were incubated for 60 minutes at 37 ° in standard medium with 0.2% BSA, 1.0 mM CaCI 2 and 0.5 mM MgS04 with no other additives (open bars) or with 1 ug/ml indomethacin (INDO; closed bars) or 1 ~ g/ml mellitin (MELL; stippled bars). Results indicate the mean±SEM for 6-8 observations in two experiments except for the effects of mellitin on cells where three observations are shown from one experiment. The inhibition of 6-keto-PGFl~ production was significant in cells (p<.001) as well as in fragments (p<.001). The stimulation of 6-keto-PGFle production by mellitin in cells and fragments was likewise significant (p<.05).
JANUARY 1985 VOL 29 NO. 1
PROSTAGLANDINS
A,
m em
E
v
200 m
u
I00
O
-
o~
z
I
I
I,
I
I
1
I
0.5
1.0
1.5
2.0
0
I
B.
0
20O O
150 u_ ~9
P w
I00
50 i
I
0
Ca++], Figure 3
mM
Effects of extracellular calcium on the production of 6-keto-PGFl~ by cells (A) and fragments (B) from bovine parathyroid glands. Cells or fragments were incubated for 60 minutes at 37 ° with 0.2% BSA, 0.5 mM MgSO4, and 1 mM EGTA and no added calcium (free calcium <10 -8 M) or the indicated concentrations of CaCI 2. Results show the mean±SEM for 12 observations in 4 experiments.
JANUARY 1985 VOL. 29 NO. 1
41
PROSTAGLANDINS
m
A.
E
300 Q.
200 i v
(/I ~p
u
4)
I00
o Q.
O,~
z"
o
t--
,7 .
300
b
t:l 0 n,"
LL Q. Ld I
~D I
I
I0 6
I 0 -5
ISOPROTERENOL Figure 4
42
, M
Effect of isoproterenol on 6-keto-PGFl~ production in cells (A) and fragments ~B). Cells or fragments were incubated for 30 or 60 minutes at 37 ° in standard medium with 0.2% BSA, 1.0 mM CaCl2, and 0.5 mM MgSO 4 with or without (-) isoproterenol. The results are the mean± SEM for 9 observations in 3 experiments (A) or 6 observations in 2 experiments (B).
JANUARY
1 9 8 5 V O L . 29 NO. 1
PROSTAGLANDINS DISCUSSION The present results demonstrate that both dispersed cells and fragments from bovine parathyroid glands produce a variety of prostaglandins. In both cells and fragments, the relative amounts of 6-keto-PGFle , PGE 2 and thromboxane B 2 produced were similar: 6-keto-PGFle, a stable non-enzymatic degradation product of prostacyclin (PGI2) , was produced in greatest quantity with production of lesser quantities of PGE 2 and thromboxane B 2 (the degradation product of thromboxane A2). The production of these prostaglandins by gland fragments, therefore, may reflect a predominant contribution by parathyroid cells themselves. Although the dispersed cell preparations used in these experiments were 90-95% pure, it is conceivable that the red blood cells which are the major contaminating cell type also produce prostaglandins. Other cell types were not routinely observed by electron microscopic evaluation of these preparations, including fat cells and platelets. The relative amount of PGF2~ produced by fragments, on the other hand, was more than that by cells, suggesting that this latter prostaglandin might be produced by a cell type(s) other than parathyroid cells per se. Because of the specificity of the antisera used to measure these compounds, it is unlikely that cross-reactivity could account for apparent production of any of these prostaglandins. The production of 6-keto-PGFle was stimulated by high calcium concentrations in both ceils and fragments, while the production of PGF2~ was also stimulated in cells. Phospholipase A 2 is known to be activated by calmodulin in a calcium-dependent manner in some cell types (14). Moreover, bovine parathyroid cells contain calmodulin (15) as well as calmodulin-dependent phosphodiesterase (15) and protein kinase(s) (16). It is possible, therefore, that the effects of calcium on prostaglandin production by dispersed cells and fragments from bovine parathyroid glands reflect the activation of this enzyme. It is not clear why there was less stimulation of 6-keto-PGFl~ production by calcium in dispersed cells than in fragments. A cell type other than parathyroid cells may be responsible for the bulk of the calcium-dependent accumulation of this metabolite. Alternatively, there may have been damage to this enzyme system during preparation of dispersed cells. Isoproterenol, on the other hand, which causes a 40-60-fold increase in cAMP accumulation in dispersed bovine parathyroid cells (17) had no effect on the production of 6-keto-PGFle by dispersed cells or fragments of parathyroid glands. In some systems cAMP has been found to inhibit (18) or stimulate (19) production of
JANUARY 1985 VOL. 29 NO. 1
43
PROSTAGLANDINS
prostaglandins. These mechanisms may not be present in parathyroid tissue or our experimental approach failed to detect them. It is of interest that bovine parathyroid cells respond in vitro to PGE 2 and PGF 2 , which stimulate (4) and inhibit (5), respectively, both cAMP accumulation and PTH release. It is possible, therefore, that local production of these prostaglandins might modulate parathyroid function. We previously found that indomethacin had no effect on calcium-regulated PTH release in dispersed bovine parathyroid cells (4,5). In these experiments, however, it is likely that there was sufficient dilution of prostaglandins produced by the dispersed cells to preclude biological effects at these cell densities. Kukreja, et al (2@), on the other hand, found no effect of indomethacin on serum calcium or iPTH in the rat, suggesting that endogenous production of prostaglandins did not acutely modify PTH release in this species. To our knowledge, similar studies are not available in the bovine species. Unlike PGE 2 and PGF2~, PGI2 had no acute effect on PTH release. Although we cannot ruIe out the possibility that PGI 2 was rapidly inactivated at the pH used in these experiments (pH 7.5), thereby precluding a biologic effect, a high concentration of the prostaglandin was employed (10-~M) and similar results were observed when it was added to the cells at 5 minutes intervals. Local production of this prostaglandin by the parathyroid gland, therefore, is of uncertain physiological significance. We have not tested the effects of thromboxanes on dispersed bovine parathyroid cells. The present results demonstrate that dispersed cells and fragments from bovine parathyroid glands produce large amounts of prostacyclin as well as readily measurable quantities of PGF2a , PGE2, and TxB 2. The production of 6-keto-PGFl~ and PGF2~ is stimulated by extracellular calcium but not by isoproterenol. Whether the endogenous production of prostaglandins within the parathyroid gland plays any role in the regulation of parathyroid function remains to be determined. ACKNOWLEDGEMENTS The authors gratefully acknowledge the expert secretarial work of Ms. Diane Rioux, Mrs. Nancy Orgill, and Mrs. Carmen Quintero as well as the excellent technical help of Joseph G. Thatcher, Edward J. Watson, and Robert Butters. This work was supported by PHS grants AM2591@ and AM3~028; Edward M. Brown is a recipient of Research Career Development Award AM~0627.
44
JANUARY 1985 VOL. 29 NO. 1
PROSTAGLANDINS
REFERENCES i. Peck, W.A., Klahr, S. Cyclic nucleotides in bone and mineral metabolism. In: Advances in Cyclic Nucleotide Research, Vol ii. (P. Greengard and G.A. Robison, eds.) Raven Press, New York, 1979. pp. 89-13~. 2. Heath, H., III. Biogenic amines and the secretion of parathyroid hormone and calcitonin. Endocr. Rev. 1:319-338. 198B. 3. Brown, E.M. PTH secretion in vivo and in vitro: regulation by calcium and other secretagogues. Mineral and Electrolyte Metabolism ~:130-15~. 1982. 4. Gardner, D.G., E.M. Brown, R.G. Windeck, and G.D. Aurbach. Prostaglandin E 2 stimulation of adenosine 3',5'-monophosphate accumulation and parathyroid hormone release in dispersed bovine parathyroid cells. Endocrinology I~3:577-582. 1978. 5. Gardner, D.G., E.M. Brown, R. Windeck, and G.D. Aurbach. Prostaglandin F2~ inhibits 3',5'-adenosine monophosphate accumulation and parathyroid hormone release from dispersed parathyroid cells. Endocrinology I~4:1-7. 1979. 6. Brown, E.M., J.G. Thatcher. Adenosine 3',5'-monophosphate (cAMP)-dependent protein kinase and the regulation of parathyroid hormone release by divalent cations and agents elevating cellular cAMP in dispersed bovine parathyroid cells. Endocrinology ~]~:1374-138B. 1982. 7. Levine, L. Levels of 13,14-dihydro-15-keto PGE 2 in some biological fluids as measured by radioimmunoassay. Prostaglandins ~4:1125-I130. 1977. 8. Swartz, S.L., G.H. Williams, N.K. Hollenberg, L. Levine, R.G.Dluhy, and T.J. Moore. Captopril-induced changes in prostaglandin production. J. Clin. Invest. 65:1257-1264. 198~. 9. Lackritz, R., D. Tulchinsky, K.J. Ryan, and L. Levine. Plasma prostaglandin metabolites in human labor. Am. J. Obstet. Gynecol. 131:484-489. 1978. i%. Snedecor, G.W., E.C. Cochran. Statistical Methods (ed 6). Ames Iowa State University Press, Iowa, 1967. p 296. ii. Dunnett, C.W. New tables for multiple comparisons with a control. Biometrics 2~:482-491. 1964. 12. Hassad, A., L. Levine. Stimulation of phospholipase and prostaglandin biosynthesis by mellitin in cell culture and in vivo. Res. Commum. Chem. Path. Pharmac. 18:5~7-157. 1977. 13. Brown, E.M., S. Hurwitz, G.D. Aurbach. Preparation of viable isolated bovine parathyroid cells. Endocrinology 99:1582-1588. 1976. 14. Wong, P.Y.-K., W.Y. Cheung. Calmodulin stimulates
JANUARY 1985 VOL. 29 NO. 1
45
PROSTAGLANDINS
15. 16. 17.
18.
19.
20.
human platelet phospholipase A2. Biochem. Biophys. Res. Commun. 9~:473-480. 1979. Brown, E.M. Calcium-regulated phosphodiesterase in bovine parathyroid cells. Endocrinology Ie7:1998-2003. 1980. Brown, E.M., J.G. Thatcher. Effects of calcium on the phosphorylation of endogenous proteins in sonicates of bovine parathyroid cells. Clin. Res. 31:381A. 1983. Brown, E.M., G.D. Gardner, R.A. Windeck, et al. Relationship of intracellular 3',5'-adenosine monophosphate accumulation to parathyroid hormone release from dispersed bovine parathyroid cells. Endocrinology 103:2323-2333. 1978. Malmsten, C., E. Granstrom, B. Samuelsson. cAMP inhibits synthesis of prostaglandin endoperoxide (PGG 2 in human platelets. Biochem. Biophys. Res. Commun. 68:569-576. 1976. Hamprecht, B., B.M. Jaffe, G.W. Philpott. Prostaglandin production by neuroblastoma, glioma, and fibroblast cell lines; stimulation by N6,O2-dibutyryl adenosine 3':5'-cyclic monophosphate. FEBS Letters 36:1193-1198. 1973. Shenerdiak, W.P., S.C. Kukreja, P.A. Johnson. Effect of prostaglandin El, E2 or indomethacin on serum parathyroid hormon~ ana'calcitonin in the rat. Prostaglandins Med. ~:641-645. 1981. Editor: P. Robertson
46
Received: 6-8-84
Accepted: 10-11-84
JANUARY 1985 VOL. 29 NO. 1
PRODUCTION
OF PROSTAGLANDINS BY DISPERSED CELLS AND FRAGMENTS FROM BOVINE PARATHYROID GLANDS
Edward M. Brown, M.D. and Stephen L. Swartz, M.D. Endocrine-Hypertension
Unit, Brigham and Women's Boston, MA 02115.
Hospital,
ABSTRACT We studied the production of prostaglandins by fragments and dispersed cells from bovine parathyroid glands. Fragments released 138±19 (SE), 132±21, 4.3±0.5, and 13+6.6 pg/mg/h of 6-keto-PGFl~ , PGF2~ , PGE2, and thromb~xane B~, respectively (n=7-26), while dispersed . cells releasea 414±110, 22~7.3, 27~3.8, and 29±11 pg/10 ° cells/h, respectively, of the same compounds (n=6-25). Indomethacin (i ~g/ml) inhibited the release of 6-keto-PGFl~ by 80-90% in fragments and cells, while mellitin stimulated release of this prostaglandin, suggesting de novQ synthesis of prostaglandins in these preparations. Calcium stimulated production of 6-keto-PGF 1 ~ by 1 3-fold in cells and 2.6-fold in fragments and also enhanced production of PGF2~ by 1.9-fold in fragments. Isoproterenol, on the other hand, had no effect on production of 6-keto-PGFl~ in either preparation. These results demonstrate that parathyroid tissue as well as parathyroid cells per se produce a variety of prostaglandins. We have previously shown that PGE 2 and PGF2~ modulate cAMP accumulation and PTH release in dispersed bovine parathyroid cells. The role of the endogenous production of prostaglandins by the parathyroid gland in the acute or chronic regulation of parathyroid function, however, remains to be determined. INTRODUCTION A number of hormonal factors modify cAMP accumulation and immunoreactive PTH (iPTH) release in dispersed bovine parathyroid cells (1-3). These include various catecholamines, secretin, and the prostaglandins PGE 2 and PGF2 ~. PGE 2 stimulates (4), while PGF2~ inhibits (5), both cAMP accumulation and PTH release. It is possible, therefore, that the production of prostaglandins locally within the parathyroid gland might play some physiological role in the regulation of parathyroid function. The production of prostaglandins by parathyroid tissue, however, has not been reported previously. For this reason, we have studied the production of 6-keto-PGFl~ (the stable, nonenzymatic degradation product of PGI2) , PGF 2 , PGE 2, and thromboxane B 2 by fragments and dispersed cells
JANUARY 1985 VOL. 29 NO. 1
35
PROSTAGLANDINS from bovine parathyroid glands. MATERIALS AND METHODS Dispersed bovine parathyroid cells were prepared by digestion of minced parathyroid tissue from one to three week old calves with collagenase and DNase, as described previously (6). After dispersal, the cellular pellet was sedimented for 5 minutes at 50 x g and resuspended several times in Minimal Essential Medium (Earle's Salts( without NaHCO 3, MgCI 2, or CaCI2), containing 20 mM HEPES ± pH 7.47 ("standard medium") as well as 0.5 mM MgS04, 1.5 mM CaCI2, and 2% bovine serum albumin (BSA). The cells were then maintained in the same medium at 37 ° until use. Cellular viability, as assessed by trypan blue exclusion, was routinely 95-100%. Cell counts were performed with a hemocytometer. 90-95% of the cells were parathyroid cells by light and electron microscopical criteria; most of the remaining cells were red blood cells. Incubations were carried out at 37 ° in disposable 5 cc polypropylene scintillation vials (Starstedt, Princeton, N.J.) using 0.5-2.0 cc of standard medium with 0.2% BSA and varying concentrations of MgSO 4 and CaC~2 as detailed below. Incubations contained 2-10 x 10 dispe~sed cells/cc or 10-40 mg wet weight/cc of fragments (1-2 mm ~ each) of parathyroid glands. Supernatant samples for subsequent determination of PTH or prostaglandins were stored frozen after allowing fragments to settle at unit gravity or sedimenting the cellular pellet at 250 x g for 2 minutes in a desk-top centrifuge (Sorvall GLC-2B). In experiments testing the effects of PGI 2 on PTH_release, the prostaglandin was dissolved at 10-3M in standard medium at pH 10 and was diluted 100-fold into incubation vials containing parathyroid cells just prior to the experiment. Radioimmunoassay for PTH was carried out as described previously (6), with a sheep antibovine parathyroid hormone antiserum (GW-I) recognizing the intact hormone as well as the C-terminal portion of the molecule. All samples from a given experiment were measured in the same assay. Calcium, magnesium, EGTA, and isoproterenol, at the concentrations used experimentally, had no effect on the radioimmunassay for PTH. i.
The following abbreviations have been employed: HEPES Hydroxyethylpiperazine ethanesulfonic acid; BSA Bovine serum albumin; EGTA - Ethyleneglycol-bis-(8-aminoethylether)-N, Nl-tetraacetic acid; 6-keto-PGFl~-6-ketoprostaglandin FI~; TxB 2 - Thromboxane B2; PGF2~- Prostaglandin F2~; PGE 2- Prostaglandin E 2, -
-
36
JANUARY 1985 V O L 29 NO. 1
PROSTAGLANDINS
6-keto P G F I % P G F ~ , TxB2, and PGE 2 were measured by the double antiSody raaioimmunoassay technique ~f Levine (7), as modified by Swartz (8). Briefly, rabbit anti-sera to 6-keto-PGF1~ , PGF2~, TxB 2 and PGE 2 were raised by immunizing raSbits with the prostaglandin coupled to human albumin. 200 ~i aliquots of supernatants from incubations of the parathyroid fragments or cells were added to 100 ~i ~f rabbit anti-PG antibody (in tris buffer) and 100 ~i of H-PG. The mixture was incubated at 37 ° C for 1 hour, after which 100 ~i of normal rabbit serum and 100 ~i of goat antl-rabbit globulin were added. After thorough mixing, the reaction mixture was incubated at 4 ° C overnight. The antibody-bound 3H-PG was precipitated by centrifugation and assayed for 3H by liquid scintillation spectrophotometry against a standard curve. In these radioimmunoassays, 6-keto-PGFle , PGF2~, TxB 2 and PGE 2 cross react with each other less than 1% (8). The lower limit of sensitivity for 6-keto-PGFl~ and TxB 2 is 10 pg/ml, and for varlationPGF~ and PGE 2 i s<10%20 pg/m(9)_ 1 To (8).evaluateThe intraassaYfurthertheC°efficientvalidityof and accuracy of the radioimmunoassay techniques, we added 44 pg of 6-keto-PGFl~ to eight different samples (i ml each); analysis of these specimens by radioimmunoassay showed a recovery of 101~7 percent (mean~SEM). Group means have been presented with the standard error of the mean as the index of dispersion. To determine whether significant changes in PG occurred with increasing amounts of extracellular calcium, the ANOVA computer program was used. This program computes the t value for the response at each calcium concentration and compares the appropriate control with the pooled variance. Homogeneity of variance was assessed by Bartlett's test (10). P values were obtained in Dunnett's tables for comparing multiple responses with a single control (ii). Other statistical analyses were carried out using the Fisher Exact Test. RESULTS Production of Drostaulandins bv fraqments ~nd dispersed cells from bovine parathyroid uiands. In preiiminary experiments, release of 6-ke£o-PGF]~ by fragments and dispersed parathyroid cells was linear for at least one hour and two hours, respectively. In subsequent experiments, therefore, production of prostaglandins was studied at 30 to 60 minutes with fragments and 30 to 120 minutes with cells. Fragments of bovine parathyroid glands released 138~19 (SE), 132~21, 4.3±0.5, and 13±6.6 pg/mg/h of 6-keto-PGFl~ , PGF2~, PGE2, and TxB2, respectively (n=7-26); dispersed cells, on the other hand, produced 4141110, 2217.3, 27~3.8, 29111 pg/106 cells/h, respectively, of the same compounds (n=6-25) (Figure i).
JANUARY 1985 VOL. 29 NO. 1
37
PROSTAGLANDINS
,ooFL
CELLS ~] FRAGMENTS
w£ u) 500f W
~
l:~ Z
o.
J
~
~ ~
~ _--
foo l30'
t
LI
I0
o
0
6- KETO PGFI~
Figure 1
38
PGF2(~
PGE2
Tx 82
Production of prostaglandins by fragments and dispersed cells from bovine parathyroid glands. Dispersed cells (open bars) or fragments (stipsled bars) were incubated for 30-120 minutes at 37 ° in standard medium with 0.2% BSA, 1.0 mM CaCI 2 and 0.5 mM MgSO 4. Results indicate the mean±SEM for 6-25 observations with cells and 7-26 observations with fragments, each measured in duplicate, expressed as release per hour.
JANUARY 1985 VOL. 29 NO. 1
PROSTAGLANDINS
Because 6-keto-PGFl~ was produced in greatest quantity by both cells and fragments, most subsequent studies were carried out using this prostaglandin to assess the effect of various agents on prostaglandin production. Effects of indomethacin or mellitin on Drostaulandin p _ ~ . Indomethacin, an inhibitor of cyclooxygenase, inhibited the production of 6-keto-PGFl~ by 91% and 78% in fragments of bovine parathyroid gland and dispersed bovine parathyroid cells, respectively (Figure 2A). In contrast, mellitin, a membrane-active peptide of bee venom which stimulates the biosynthesis of PG by interacting with phospholipid membrane bilayers, resulting in arachidonic acid release and PG production (12), stimulated the production of 6-keto-PGFl~ by 2.6- and 5.7-fold in fragments and cells, respectzvely (Figure 2B). These results suggest that both fragments of bovine parathyroid tissue and dispersed bovine parathyroid cells produce prostaglandins de novo. Effects of extracellular calcium on the production of 6-keto-PGFl~. Raising extracellular calcium from <10 -8 M 2 to 2 mM pr~-duced a 1.3-fold stimulation of the production of this prostaglandin in cells (Figure 3A) and a 2.6-fold stimulation of the production of 6-keto-PGF ~ by fragments of bovine parathyroid tissue (Figure 3B). ~he same change in extracellular calcium stimulated the production of PGF2~ by 1.9-fold in fragments (from 140~47 to 257~35 pg/mg/hr)7 All of these changes were statistically significant
(p<.01). Effect of isoproterenol on 6-keto-PGFl~" Droduction in cells ~ U I ~ . Isoproterenol had no ~-ffect on the production of 6-keto-PGFl~ by dispersed bovine parathyroid cells (Fig. 4A). In fragments of bovine parathyroid glands, this agonist caused a slight but statistically insignificant increase in the production of 6-keto-PGFl~ (Figure 4B). Effect of PGI2. on PTH release in dispersed bovine parathyroi4 c~lls We have not previously examined the effects of prostacyclin (PGI2) on PTH release. This agent_had no effect on PTH release at doses ranging from 10 -~ to 10 -5 M and at times from 5 to 30 minutes (not shown). Because of the lability of PGI2, we also added PGI 2 at 5 minute intervals in a 30 mznute experiment but likewise found no effect under these conditions. 2.
This is the approximate free calcium concentration 1 mM EGTA and no added CaCI 2 at pH 7.5.
JANUARY 1985 VOL. 29 NO. 1
with
39
PROSTAGLANDINS
~A. rn
mr*
5O0 ,~
-/-
~
- 9OO
200
~
- 600
oo
~
300
o
o°
200
250
Figure 2
40
o~--CONTROL
o
/ N
,oo ~
g
'
+ INDO
CONTROL
+ MELL
Effects of indomethacin or mellitin on production of 6-keto-PGFl~. Dispersed cells (Panel A) or fragments from bovine parathyroid glands (Panel B) were incubated for 60 minutes at 37 ° in standard medium with 0.2% BSA, 1.0 mM CaCI 2 and 0.5 mM MgS04 with no other additives (open bars) or with 1 ug/ml indomethacin (INDO; closed bars) or 1 ~ g/ml mellitin (MELL; stippled bars). Results indicate the mean±SEM for 6-8 observations in two experiments except for the effects of mellitin on cells where three observations are shown from one experiment. The inhibition of 6-keto-PGFl~ production was significant in cells (p<.001) as well as in fragments (p<.001). The stimulation of 6-keto-PGFle production by mellitin in cells and fragments was likewise significant (p<.05).
JANUARY 1985 VOL 29 NO. 1
PROSTAGLANDINS
A,
m em
E
v
200 m
u
I00
O
-
o~
z
I
I
I,
I
I
1
I
0.5
1.0
1.5
2.0
0
I
B.
0
20O O
150 u_ ~9
P w
I00
50 i
I
0
Ca++], Figure 3
mM
Effects of extracellular calcium on the production of 6-keto-PGFl~ by cells (A) and fragments (B) from bovine parathyroid glands. Cells or fragments were incubated for 60 minutes at 37 ° with 0.2% BSA, 0.5 mM MgSO4, and 1 mM EGTA and no added calcium (free calcium <10 -8 M) or the indicated concentrations of CaCI 2. Results show the mean±SEM for 12 observations in 4 experiments.
JANUARY 1985 VOL. 29 NO. 1
41
PROSTAGLANDINS
m
A.
E
300 Q.
200 i v
(/I ~p
u
4)
I00
o Q.
O,~
z"
o
t--
,7 .
300
b
t:l 0 n,"
LL Q. Ld I
~D I
I
I0 6
I 0 -5
ISOPROTERENOL Figure 4
42
, M
Effect of isoproterenol on 6-keto-PGFl~ production in cells (A) and fragments ~B). Cells or fragments were incubated for 30 or 60 minutes at 37 ° in standard medium with 0.2% BSA, 1.0 mM CaCl2, and 0.5 mM MgSO 4 with or without (-) isoproterenol. The results are the mean± SEM for 9 observations in 3 experiments (A) or 6 observations in 2 experiments (B).
JANUARY
1 9 8 5 V O L . 29 NO. 1
PROSTAGLANDINS DISCUSSION The present results demonstrate that both dispersed cells and fragments from bovine parathyroid glands produce a variety of prostaglandins. In both cells and fragments, the relative amounts of 6-keto-PGFle , PGE 2 and thromboxane B 2 produced were similar: 6-keto-PGFle, a stable non-enzymatic degradation product of prostacyclin (PGI2) , was produced in greatest quantity with production of lesser quantities of PGE 2 and thromboxane B 2 (the degradation product of thromboxane A2). The production of these prostaglandins by gland fragments, therefore, may reflect a predominant contribution by parathyroid cells themselves. Although the dispersed cell preparations used in these experiments were 90-95% pure, it is conceivable that the red blood cells which are the major contaminating cell type also produce prostaglandins. Other cell types were not routinely observed by electron microscopic evaluation of these preparations, including fat cells and platelets. The relative amount of PGF2~ produced by fragments, on the other hand, was more than that by cells, suggesting that this latter prostaglandin might be produced by a cell type(s) other than parathyroid cells per se. Because of the specificity of the antisera used to measure these compounds, it is unlikely that cross-reactivity could account for apparent production of any of these prostaglandins. The production of 6-keto-PGFle was stimulated by high calcium concentrations in both ceils and fragments, while the production of PGF2~ was also stimulated in cells. Phospholipase A 2 is known to be activated by calmodulin in a calcium-dependent manner in some cell types (14). Moreover, bovine parathyroid cells contain calmodulin (15) as well as calmodulin-dependent phosphodiesterase (15) and protein kinase(s) (16). It is possible, therefore, that the effects of calcium on prostaglandin production by dispersed cells and fragments from bovine parathyroid glands reflect the activation of this enzyme. It is not clear why there was less stimulation of 6-keto-PGFl~ production by calcium in dispersed cells than in fragments. A cell type other than parathyroid cells may be responsible for the bulk of the calcium-dependent accumulation of this metabolite. Alternatively, there may have been damage to this enzyme system during preparation of dispersed cells. Isoproterenol, on the other hand, which causes a 40-60-fold increase in cAMP accumulation in dispersed bovine parathyroid cells (17) had no effect on the production of 6-keto-PGFle by dispersed cells or fragments of parathyroid glands. In some systems cAMP has been found to inhibit (18) or stimulate (19) production of
JANUARY 1985 VOL. 29 NO. 1
43
PROSTAGLANDINS
prostaglandins. These mechanisms may not be present in parathyroid tissue or our experimental approach failed to detect them. It is of interest that bovine parathyroid cells respond in vitro to PGE 2 and PGF 2 , which stimulate (4) and inhibit (5), respectively, both cAMP accumulation and PTH release. It is possible, therefore, that local production of these prostaglandins might modulate parathyroid function. We previously found that indomethacin had no effect on calcium-regulated PTH release in dispersed bovine parathyroid cells (4,5). In these experiments, however, it is likely that there was sufficient dilution of prostaglandins produced by the dispersed cells to preclude biological effects at these cell densities. Kukreja, et al (2@), on the other hand, found no effect of indomethacin on serum calcium or iPTH in the rat, suggesting that endogenous production of prostaglandins did not acutely modify PTH release in this species. To our knowledge, similar studies are not available in the bovine species. Unlike PGE 2 and PGF2~, PGI2 had no acute effect on PTH release. Although we cannot ruIe out the possibility that PGI 2 was rapidly inactivated at the pH used in these experiments (pH 7.5), thereby precluding a biologic effect, a high concentration of the prostaglandin was employed (10-~M) and similar results were observed when it was added to the cells at 5 minutes intervals. Local production of this prostaglandin by the parathyroid gland, therefore, is of uncertain physiological significance. We have not tested the effects of thromboxanes on dispersed bovine parathyroid cells. The present results demonstrate that dispersed cells and fragments from bovine parathyroid glands produce large amounts of prostacyclin as well as readily measurable quantities of PGF2a , PGE2, and TxB 2. The production of 6-keto-PGFl~ and PGF2~ is stimulated by extracellular calcium but not by isoproterenol. Whether the endogenous production of prostaglandins within the parathyroid gland plays any role in the regulation of parathyroid function remains to be determined. ACKNOWLEDGEMENTS The authors gratefully acknowledge the expert secretarial work of Ms. Diane Rioux, Mrs. Nancy Orgill, and Mrs. Carmen Quintero as well as the excellent technical help of Joseph G. Thatcher, Edward J. Watson, and Robert Butters. This work was supported by PHS grants AM2591@ and AM3~028; Edward M. Brown is a recipient of Research Career Development Award AM~0627.
44
JANUARY 1985 VOL. 29 NO. 1
PROSTAGLANDINS
REFERENCES i. Peck, W.A., Klahr, S. Cyclic nucleotides in bone and mineral metabolism. In: Advances in Cyclic Nucleotide Research, Vol ii. (P. Greengard and G.A. Robison, eds.) Raven Press, New York, 1979. pp. 89-13~. 2. Heath, H., III. Biogenic amines and the secretion of parathyroid hormone and calcitonin. Endocr. Rev. 1:319-338. 198B. 3. Brown, E.M. PTH secretion in vivo and in vitro: regulation by calcium and other secretagogues. Mineral and Electrolyte Metabolism ~:130-15~. 1982. 4. Gardner, D.G., E.M. Brown, R.G. Windeck, and G.D. Aurbach. Prostaglandin E 2 stimulation of adenosine 3',5'-monophosphate accumulation and parathyroid hormone release in dispersed bovine parathyroid cells. Endocrinology I~3:577-582. 1978. 5. Gardner, D.G., E.M. Brown, R. Windeck, and G.D. Aurbach. Prostaglandin F2~ inhibits 3',5'-adenosine monophosphate accumulation and parathyroid hormone release from dispersed parathyroid cells. Endocrinology I~4:1-7. 1979. 6. Brown, E.M., J.G. Thatcher. Adenosine 3',5'-monophosphate (cAMP)-dependent protein kinase and the regulation of parathyroid hormone release by divalent cations and agents elevating cellular cAMP in dispersed bovine parathyroid cells. Endocrinology ~]~:1374-138B. 1982. 7. Levine, L. Levels of 13,14-dihydro-15-keto PGE 2 in some biological fluids as measured by radioimmunoassay. Prostaglandins ~4:1125-I130. 1977. 8. Swartz, S.L., G.H. Williams, N.K. Hollenberg, L. Levine, R.G.Dluhy, and T.J. Moore. Captopril-induced changes in prostaglandin production. J. Clin. Invest. 65:1257-1264. 198~. 9. Lackritz, R., D. Tulchinsky, K.J. Ryan, and L. Levine. Plasma prostaglandin metabolites in human labor. Am. J. Obstet. Gynecol. 131:484-489. 1978. i%. Snedecor, G.W., E.C. Cochran. Statistical Methods (ed 6). Ames Iowa State University Press, Iowa, 1967. p 296. ii. Dunnett, C.W. New tables for multiple comparisons with a control. Biometrics 2~:482-491. 1964. 12. Hassad, A., L. Levine. Stimulation of phospholipase and prostaglandin biosynthesis by mellitin in cell culture and in vivo. Res. Commum. Chem. Path. Pharmac. 18:5~7-157. 1977. 13. Brown, E.M., S. Hurwitz, G.D. Aurbach. Preparation of viable isolated bovine parathyroid cells. Endocrinology 99:1582-1588. 1976. 14. Wong, P.Y.-K., W.Y. Cheung. Calmodulin stimulates
JANUARY 1985 VOL. 29 NO. 1
45
PROSTAGLANDINS
15. 16. 17.
18.
19.
20.
human platelet phospholipase A2. Biochem. Biophys. Res. Commun. 9~:473-480. 1979. Brown, E.M. Calcium-regulated phosphodiesterase in bovine parathyroid cells. Endocrinology Ie7:1998-2003. 1980. Brown, E.M., J.G. Thatcher. Effects of calcium on the phosphorylation of endogenous proteins in sonicates of bovine parathyroid cells. Clin. Res. 31:381A. 1983. Brown, E.M., G.D. Gardner, R.A. Windeck, et al. Relationship of intracellular 3',5'-adenosine monophosphate accumulation to parathyroid hormone release from dispersed bovine parathyroid cells. Endocrinology 103:2323-2333. 1978. Malmsten, C., E. Granstrom, B. Samuelsson. cAMP inhibits synthesis of prostaglandin endoperoxide (PGG 2 in human platelets. Biochem. Biophys. Res. Commun. 68:569-576. 1976. Hamprecht, B., B.M. Jaffe, G.W. Philpott. Prostaglandin production by neuroblastoma, glioma, and fibroblast cell lines; stimulation by N6,O2-dibutyryl adenosine 3':5'-cyclic monophosphate. FEBS Letters 36:1193-1198. 1973. Shenerdiak, W.P., S.C. Kukreja, P.A. Johnson. Effect of prostaglandin El, E2 or indomethacin on serum parathyroid hormon~ ana'calcitonin in the rat. Prostaglandins Med. ~:641-645. 1981. Editor: P. Robertson
46
Received: 6-8-84
Accepted: 10-11-84
JANUARY 1985 VOL. 29 NO. 1