- Email: [email protected]
Cyclooxygenase products of arachidonic acid metabolism by mouse bone in organ culture
418
Biochimica et Biophysics @ Elsevier/North-Holland
Acta, 620 (1980) 418-428 Biomedical Press
BBA 57681
CYCLOOXYGENASE PRODUCTS OF ARACHIDONIC BY MOUSE BONE IN ORGAN CULTURE
EDWARD
F. VOELKEL,
ARMEN
H. TASHJIAN,
Jr. and LAWRENCE
Laboratory of Toxicology, Harvard School of Public cology, Harvard Medical School, Boston, MA 02115, Brandeis University, Waltham, MA 02154 (U.S.A.) (Received
Fehruary
13th,
ACID METABOLISM
LEVINE
Health and Department of Pharmaand Department of Biochemistry,
1980)
Key words: Prostaglandin cyclooxygenase; Throm boxane; (Mouse bone)
Arachidonic
acid; Indomethacin;
Prostacyclin;
Summary The products of endogenous and exogenous arachidonic acid metabolism via the cyclooxygenase pathway in mouse bone in organ culture were identified and quantitated by the use of high performance liquid chromatography and radioimmunoassay. Production of prostaglandins E2, Fza, and IZ from endogenous substrate was stimulated by incubation of bone with epidermal growth factor and the tumor promoter 12-0-tetradecanoyl-phorbol-13-acetate. Addition of arachidonic acid to the culture medium not only resulted in the accuBone metabmulation of prostaglandins Ez, F,, , and IZ but also thromboxane. olized prostaglandins Ez and F,, to their respective 13,14-dihydro-15-ketoderivatives. Prostaglandin I*, measured as 6-keto-prostaglandin F,, was synthesized by bone, and metabolic products of prostaglandin IZ or 6-keto-prostaglandin Fr,, either 6,15diketo-prostaglandin F,, or 13,14dihydro-6,15-diketoprostaglandin F 1Lv,were also detected in the culture media. Formation of cyclooxygenase products of endogenous and exogenous arachidonic acid metabolism (both basal and stimulated) and bone resorption were inhibited by indomethatin. Bone as a tissue responded biochemically not only to exogenous prostaglandins and agents that enhance endogenous prostaglandin production but also to exogenous arachidonic acid by biosynthesis of prostaglandins, prostacyclin and thromboxane. Furthermore, bone metabolized these cyclooxygenase products to their more stable metabolites.
Address Health,
correspondence 665
Huntington
to:
Armen
Avenue,
H.
Boston.
Tashjian, MA
Jr.,
02115.
Laboratory U.S.A.
of Toxicology,
Harvard
School
of Public
419
Introduction Exogenous prostaglandins were first shown to stimulate the resorption of bone in culture in 1970 [ 11. We suggested that it was excessive production of prostaglandin Ez by certain tumors that caused the hypercalcemia of malignant disease in these model systems [ 2-51. Recently, it has been shown that rat bone can metabolize arachidonic acid to several products [6]. Earlier, we had demonstrated that prostaglandin Ez was synthesized by bone and suggested that it acted locally to stimulate bone resorption [ 7-91; among the most active stimulators of prostaglandin Ez production by bone were the tumor-promoting phorbol esters [ 71 and epidermal growth factor [8]. In our previous studies, only prostaglandin E2 had been measured, and de novo biosynthesis was demonstrated by inhibition of the prostaglandin cyclooxygenase by indomethatin. In this report we show that bone can metabolize both endogenous and exogenous arachidonic acid via the cyclooxygenase pathway not only to the primary prostaglandins (e.g., prostaglandins Ez and F2@)but also to prostacyclin and thromboxane. In addition, prostaglandin Ez, prostaglandin F,, and either prostaglandin IZ or 6-keto-prostaglandin F,, are converted by bone to their more stable metabolites. Materials and Methods Organ culture of bone. Neonatal mouse calvaria were placed in organ culture as described previously [lo]. For all the experiments described in this report, the medium was Dulbecco’s modified Eagle’s medium supplemented with 5 mg/ml bovine albumin (albumin medium). Calvaria were preincubated in albumin medium for 24 h before experimental treatments were begun. The medium was changed at this time, and fresh control medium or medium containing a specific treatment was added and incubation continued for an additional 48 h. Bone resorption was determined by measuring the release of calcium from the calvaria into medium at the end of the 48-h treatment period
[lOI* Measurement of medium calcium. The concentration of total calcium in bone culture medium was measured by automatic fluorimetric titration with a Corning calcium analyzer, Model 940. Measurement of metabolites of arachidonic acid. Medium from 3 to 8 separate cultures (6-16 ml) treated identically was pooled and acidified to pH 3.5 by addition of 1.0 M HCl. The acidified medium was processed on small XAD-2 columns (Isolab) which had been prewashed with 20-25 ml of distilled HzO. The initial effluent was rechromatographed on the same column. The column was then washed with 20-25 ml of distilled Hz0 and then eluted with 20-25 ml of ethanol. The ethanol eluate was evaporated to dryness under Nz gas, the residue was taken up in l-2 ml of ethanol and concentrated to about 100 ~1 for fractionation by high-performance liquid chromatography. For high-performance liquid chromatography we used a Waters liquid chromatograph, model ALC/GPC 244 (Milford, MA) equipped with a model 6000A Solvent Delivery System. The column employed was a 30 cm X 3.9 mm Fatty Acid Analysis Column used in the reversed phase mode [ 111. The arachi-
420
donic acid metabolites were eluted isocratically with a solvent system containing CH~CN/C~H~/CH~COOH~H~O (230 : 2 : 1 : 767, v/v). The flow rate was 2.0 ml/min per tube. 60 tubes were collected for each run. Each fraction was lyophilized to dryness and resuspended in 0.5 ml of Gel-Tris buffer (0.01 M Tris-HCI, 0.14 M NaCl, and 1 mg/ml gelatin, pH 7.4) for analysis by radioimmuno~say. The high-perforate liquid ~hromato~aphy elution profiles for Flcu, thromboxane Bz, 6-keto-prostaglandin Fla, 6,15-diketo-prostaglandin F,, prostaglandin Fza, prostaglandin Ez, 13,14-dihydro-15-keto-prostaglandin E, obtained with nonradioactive and 13,14-dihydro-l5-keto-prostaglandin standard preparations and serologic measurements have been described previously [ 11,12] and were, in the experiments reported here, respectively, fraction numbers B-11, 12-16, 18-21, 22-24, 28-32, 36-41, and 43-46. The chromatographic properties of the nonradioactive standards were confirmed by chromatography of the appropriate labeled compounds. Average recovery for each of these me~bolites was 60% [13]. The data for culture medium samples were not corrected for recovery. Under the chromatographic conditions used, arachidonic acid was retarded on the column and did not appear in the effluent until after 90 min (fraction number >60). The total amount of metabolite in each peak was obtained by summing the values for each individual fraction comprising that peak. Radioimmunoassay. The appropriate fractions (20-400 ~1 each) corresponding to the elution volumes of the seven known standards were analyzed quantitatively by selective radioimmunoassays [ 11,121. Assays were performed using the following protocols. 6-Keto-pros~l~d~ F la, thromboxane Bz, prostaglandin FZa, prostaglandin EZ, 13,14-dihydro-15-keto-prostaglandin Fzo, and 13,143Hdihydro-15-keto-prostaglandin E 2 were measured using the homologous labeled compound as tracer and antibody against the homologous ligand. 6,15Diketo-prostaglandin Fi, was measured using 13,14-dihydro-15-keto[ 5,6,8,9, ll,12,14(n)-3H]-prostagl~d~ FZa: and an antiserum which recognizes a 15keto function as immunodominant, anti-13,14-dihydro-l&keto-prostaglandin F2, 1141. Statistical method. The results of each experiment were subjected to analysis of variance, and the standard errors (S.E.) were calculated from the residual error term of that analysis. Where variances were not heterogeneous and the experiments were of the same design, the results were pooled and appropriate statistical significance was calculated. Materials. Epidermal growth factor was purchased from Collaborative Research, Inc., Waltham, MA. 12-O-Tetradecanoyl-phorbol-13-acetate was obtained from Consolidated Midland Corp., Brewster, NY. Bovine albumin and indomethacin were from Sigma Chemical Co., St. Louis, MO. The XAD-2 resin columns were from Isolab, Inc., Akron, OH. Chemicals were of reagent or highperformance liquid chromatography grade and were obtained from Fisher Scientific, Boston, MA. The standard pros~gl~dins and other metabolites of arachidonic acid were gifts from the Upjohn Co., Kalamazoo, MI. The radiolabeled prostaglandins and metabolites used were purchased from New England Nuclear Corp., Boston, MA and the Amersham Co., Chicago, IL.
421
Results We have previously shown that epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate stimulate bone resorption in vitro by a mechanism that involves the production of prostaglandin Ez [ 7,8]. In these earlier experiments prostaglandin E, was measured serologically by direct assay of unfractionated culture medium. In order to determine whether other cyclooxygenase products of arachidonic acid were formed, we cultured mouse calvaria in the absence and presence of epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate and assayed the media, using a combination of high-performance liquid chromatography and radioimmunoassay, for prostaglandin FZcu,prostaglandin Ez, their 13,14-dihydro-15-keto metabolites, and two degradation products of the labile prostacyclin prostaglandin IZ. Positions of the chromatographic peaks for each of the six products are shown in Fig. 1 for control cultures and bones treated with epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate. The quantitative data from three separate experiments are presented in Fig. 2. Because of the uniformity of the design and results in each experiment and the lack of statistical heterogeneity in variances, data from individual experiments were pooled, analyzed statistically, and summarized in Table I. Both 12-O-tetradecanoyl-phorbol-13-acetate and epidermal growth factor stimulated bone resorption as shown by the increase in medium calcium concentration (Table I). 12-0-Tetradecanoyl-phorbol-13-acetate and epidermal growth factor each elicited large (7-8-fold) increases in the production of Gk-PGh,
6,15k-PGb,
PGE,
PGFm
13,14HJ5k-PG&,
]0.07~0
05-
13,14Hi15k-PGI
25!
0.6
23
FRACTION
30
40
45
NUM6ER
Fig. 1. Elution profiles on high-performance liquid chromatography of samples of culture media from control bones and bones treated with epidermal growth factor or 12-O-tetradecanoyl-phorbol-13-acetate. Groups of 5-7 calvaria per treatment were cultured for 48 h in the absence of additives (control) or with epidermal growth factor (50 ng/ml) or 12-O-tetradecanoyl-phorbol-13-acetate (25 ng/ml). At the end of the experiment, medium was harvested from each culture and one or two pools per treatment group were prepared, extracted and chromatographed. The effluent fractions were assayed serologically for the metabolites shown. PG. prostagkxiin.
422 Gk-PGh,
6,15k-PG\,
PGE,
13,i4H~15k~PGF2, 13,14Hi15k-PGE,
20-
C T
E
C
T
E
C
T
E
Fig. 2. Concentrations of six cyclooxygenase metabolites of arachidonic acid in bone culture medium. Groups of 5-7 calvaria per treatment were cultured for 48 h in the absence of additives (C for control) or with 25 ng/ml 12-O-tetradecanoyl-phorbol-13-acetate (T) or 50 &ml epidermal growth factor (E). Medium was harvested, processed and assayed as described in Fig. 1. Three separate experiments of the same design were performed. Medium from each treatment group for each experiment was combined into one or two separate pools. Each point gives the value for one such pool. PG. prostaglandin.
fraction (about 20-30%) of this prostaglandin E2; a small but significant prostaglandin E, was metabolized to 13,14-dihydro-15-keto-prostaglandin Ez by bone. Smaller increases in the production of prostaglandin Fzcv and its 13,14dihydro-15-keto metabolite were noted. In response to 12-O-tetradecanoylphorbol-13-acetate and epidermal growth factor bone elicited a large increase (about lo-fold) in the production of prostaglandin IZ as measured by accumulation of its products 6-keto- and 6,15diketo-prostaglandin F,,. Although not shown in these experiments, we have previously demonstrated [7,8] that the effects of both epidermal growth factor and 12-0-tetradecanoyl-phorbol-13acetate on bone resorption and on the production of prostaglandin E, by bone are inhibited in parallel by low concentrations of the cyclooxygenase inhibitor indomethacin. In the experiments just described, epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate were used as agents to stimulate prostaglandin synthesis from endogenous substrate in bone. Because exogenous arachidonic acid has been shown to enhance bone resorption by a pathway that is blocked by indomethacin [9], we have added exogenous arachidonic acid to bones in culture and have measured and compared thg. products formed to those elicited in response to epidermal growth factor and 12-0-tetradecanoyl-phorbol-13acetate. The positions of the chromatographic peaks for each of the six products described previously as well as that for thromboxane Bz in cultures treated with arachidonic acid alone or arachidonic acid plus indomethacin are shown in Fig. 3. The results presented in Table II demonstrate that arachidonic acid enhanced bone resorption in a dose-dependent manner and that this effect, but not that of exogenous prostaglandin E,, was inhibited completely by indo-
I ACID
IN MEDIUM
OF CALVARIA
CULTURED
WITH
12-O-TETRADECA-
*P < 0.05. ** P < 0.01. ***p < 0.001.
None 12-O-Tetradecanoylphorbol-13acetate (25 ng/ml) EGF (50 &ml)
Treatment
10.1 r 0.32 *** 12.7 t 0.28 ***
7.4 r 0.30
Medium calcium
1.0 1.3
? 0.12 *** f 0.14 ***
0.11 * 0.11
2.5 2.5
r 0.39 * + 0.39 *
0.91 * 0.30
GJB-diketo PGFlo,
@g/ml)
6-k&o-
acid metabolites
PGFlo,
Arachidonic
0.07 ? 0.016 0.09 * 0.019 *
0.02 t 0.016
PGFza
0.29 f 0.04 ** 0.34 r 0.05 **
0.04 + 0.04
PGEz
0.019 r 0.003 0.032 f 0.003 ***
0.013 f 0.003
1 5-keto-PGFZ,
13.14-dihydro-
0.06 ? 0.015 0.10 * 0.017 ***
0.02 * 0.017
13,14-dihy&o15-keto-PGEZ
The data given are mean values pooled from three separate experiments of the same design. Each experiment was analyzed separately and the mean values and variances were not heterogeneous. The values given are mean k S.E.. and statistical significance of the experimental values as compared to control are given and are indicated by asterisks (see below). Groups of 5-7 bones were incubated in albumin medium alone or in medium containing 12-O-tetradecanoyl-phorbol-13-acetate or epidermal growth factor for 48 h. At the end of each experiment, medium was collected and assayed for calcium concentration or for metabolites of arachidonic acid by high performance liquid chromatography and radioimmunoassay. PG. prostaglandin; EGF, epidermal growth factor.
CONCENTRATIONS OF CALCIUM AND METABOLITES OF ARACHIDONIC NOYGPHORBOL-13-ACETATE AND EPIDERMAL GROWTH FACTOR
TABLE
424
6k-PGF,a .___
TX&
6,15k-PGF,,,
PGFza
I L_ 4
13,14Hi15k-PGb,
PGE,
L
30
23
20
15
-10
13,14Hi15k-FGJ
_I L_.ldLLL 40
45
FRAC?TOIV NUM6ER Fig. 3. Elution profiles on high-performance liquid chromatography of culture media from bones treated with arachidonic acid alone or arachidonic acid plus indomethacin. Groups of 6-8 calvaria per treatment were cultured for 48 h with 10 Hglml arachidonic acid (AA) or 10 pg/ml arachidonic acid plus 0.5 pg/ml indomethacin (AA + Indo). At the end of the experiment, medium was harvested, pooled, extracted, chromatographed and assayed. TX, thromboxane; PG. prostaglandin.
methacin. The data from experiments using 5 or 10 pg/ml of exogenous arachidonic acid are presented in Table III. In both experiments arachidonic acid stimulated bone resorption and this effect was inhibited by indomethacin. The patterns of arachidonic acid metabolites formed in the two experiments were qualitatively similar, although the effects were generally larger in Expt. II in which the higher concentration of arachidonic acid was used. Exogenous arachidonic acid stimulated the formation and release of prostaglandin E,, prostaglandin FZa, their 13,14-dihydro-15-keto metabolites and presumably also thromboxane AZ as measured by its product thromboxane Bz. The increases in arachidonic acid metabolites, as the increases in bone resorption, were inhibited by indomethacin. The magnitudes of the effects of 10 pg/ml of arachidonic acid were comparable to those elicited by 12-O-tetradecanoyl-phorbol-13-acetate (25 ng/ml) and epidermal growth factor (50 ng/ml) except that
425 TABLE
II
CONCENTRATIONS
OF
CALCIUM
DONIC ACID IN THE ABSENCE
IN THE
MEDIUM
AND PRESENCE
OF CALVARIA
CULTURED
WITH
ARACHI-
OF INDOMETHACIN
The data given are mean values pooled from four separate experiments of similar design. Each experiment was analyzed statistically and the mean values and variances were not heterogeneous. The values given are means ? S.E., and statistical significance of experimental values as compared to control are given as indicated by the asterisk. Groups of 4-l bones were incubated in albumin medium alone or with the additives indicated for 48 h. At the end of each experiment, medium was collected and the concentration of calcium measured. Treatment
None Indomethacin Prostagiandin E2 Arachidonic acid Arachidonic acid Arachidonic acid and indomethacin Arachidonic acid and indomethacin Prostaghmdin E2 and indomethacin
Dose
Medium calcium
wg/mD
(mg/mD
0.50 0.10 5.0 10.0 5.0 0.50 10.0 0.50 0.10 0.50
6.9 6.6 10.4 8.6 10.1
f r + ? r
0.22 0.22 0.21 0.38 0.25
* * *
6.9 f 0.42 7.4 * 0.28 10.5
* 0.27
*
* P < 0.001.
12-0-tetradecanoyl-phorbol-13-acetate and epidermal growth factor did not enhance the production of thromboxane AZ, at least not to the extent that measurable levels of thromboxane Bz accumulated in the medium (data not shown). Once again there was a large (5-20-fold) increase in the production of prostaglandin I* as measured by accumulation of 6-keto-prostaglandin F,, and this effect was also comparable in magnitude to those elicited by 12-O-tetradecanoyl-phorbol-13-acetate and epidermal growth factor (Table I). Discussion From the results of the experiments presented here we conclude that mouse calvaria, as representative of bone as a tissue, can produce four major cyclooxygenase products of arachidonic acid, i.e., prostaglandin E?, prostaglandin F 201, thromboxane A*, and prostaglandin IZ. Each product was identified by its characteristic elution position from reversed phase high-performance liquid chromatography and quantitated by serologic assay. This approach enables discrimination between the mono- and dienoic prostaglandins of the E series which are not distinguished by the antibodies alone [ 141. Each cyclooxygenase product formed was further metabolized in the bone culture system. Prostaglandin Ez and prostaglandin F,, were converted to their 13,14-dihydro-15keto metabolites indicating the presence in bone of 15-hydroxyprostaglandin dehydrogenase and prostaglandin A13-reductase. Prostaglandin IZ accumulated as 6-keto-prostaglandin F,, by a step which is likely to be nonenzymic [15] as well as 6,15diketo-prostaglandin F,, (or 13,14-dihydro-6,15-diketo-prostaglandin F1,; our high-performance liquid chromatography procedure and radioimmunoassay probably do not distinguish these two metabolites), a product of
OF CALCIUM AND METABOLITES
OF ARACHIDONIC
ACID IN MEDIUM OF CALVARIA
CULTURED
WITH ADDED
ARACHIDONIC
* P < 0.01. ** P < 0.001.
Expt. II None Arachidonic acid (10 /.&g/ml) Arachidonic acid + indomethacin
Expt. I None Arachidonic acid (5 pg/ml) Arachidonic acid + indomethacin
Treatment
*
6.6 f 0.47 10.9 * 0.47 ** 6.8 f 0.52
6.9 f 0.40 8.6 ? 0.40 6.9 ? 0.43
(mg/dI)
Medium calcium
0.04 0.90 0.03
0.03 0.15
0.84 1.4 1.1
6.15-diketoPGFlo,
G-ketoPGFla
Thromboxane B2
Arachidonic acid metabolites @g/ml)
<0.005 0.12 0.02
0.013 0.025 <0.005
PGFza
0.025 0.06 0.03
PGE,
PGF2,
13.14dihydro15-k&o-
13,14dihydro15ketoPGE2
The data from two separate experiments are presented. Values for medium calcium are means * S.E. for 6-8 bones per group. and statistical significance of the experimental values compared to control are given as indicated by the asterisks. Values for arachidonic acid metabolites were determined on pooled samples of medium and no S.E. was calculated. Bones were incubated in albumin medium alone or in medium containing added arachidonic acid (5 or 10 &ml in Expt. I and II, respectively) alone or arachidonic acid plus indomethacin (0.5 pg/mI in both experiments). At the end of each experiment, medium was collected and assayed for calcium concentration or for metabolites of arachidonic acid by high performance liquid chromatography and radioimmunoassay. PG, prostaglandin.
CONCENTRATfONS ACID
TABLE III
427
15-hydroxyprostaglandin dehydrogenase (and A13-reductase) activity. When bones were stimulated to resorb by treatment with 12-O-tetradecanoyl-phorbol-l3-acetate or epidermal growth factor (Table I), concentrations of prostaglandin Ez, prostaglandin Fzcu, and prostaglandin IZ (measured as 6-keto-prostaglandin F,,) increased 4-lo-fold. Relatively small fractions of the stimulated synthesis of prostaglandin E2 and prostaglandin Fzo, accumulated as their respective 13,14-dihydro-15-keto metabolites, about 30% in the case of prostaglandin Fzcv and about 25% for prostaglandin E2. These values are considerably lower than the accumulation of 13,14-dihydro-l5-keto-prostaglandin Ez or prostaglandin F,, in the plasma of mice [16] or rabbits [17] bearing prostaglandin-producing tumors. It is possible that our measurements of 13,14-dihydro-15-keto-prostaglandin Ez underestimate the’ production of this metabolite because it has been reported to be degraded further in aqueous medium to ll-deoxy-13,14-dihydro-l5-keto-11,16-cycle-prostaglandin Ez [ 201. The values for accumulated 6,15-diketo-prostaglandin F,, were high and were calculated from low amounts of inhibition of the immune system. These results may reflect predominance in bone of the 15-hydroxyprostaglandin dehydrogenase pathway of prostaglandin IZ metabolism. Alternatively, our quantitative analysis of the 6,15-diketo metabolite of prostaglandin IZ, which uses a heterologous immune system, may overestimate the true value of 6,15-diketo-prostaglandin F1, due to cross-reactions with our antibody by additional unidentified metabolites. The absolute values will need to be confirmed by a homologous radioimmunoassay for 6,15diketo-prostaglandin F,, which is being developed. When bones were stimulated to resorb with 10 pg/ml of exogenous arachidonic acid (Table III), the general pattern of cyclooxygenase products formed was similar to that generated from endogenous substrate exept that significant amounts of thromboxane Bz were detected. It is possible that the K, for the thromboxane synthetase in bone is high and that insufficient substrate was generated from endogenous sources even after stimulation with 12-O-tetradecanoyl-phorbol-13-acetate and epidermal growth factor. The measured thromboxane Bz was not due to contamination by arachidonic acid because arachidonate is greatly retarded on the column and, even more importantly, its appearance was eliminated by incubation with arachidonic acid plus indomethacin (Table III). It is possible that thromboxane B2 was not detected in experiments using endogenous arachidonic acid substrate because the low amounts of thromboxane AZ generated were bound to protein in the medium and thus protected from conversion to thromboxane Bz [18]. In the presence of exogenous substrate in the medium, it is possible that the arachidonic acid could displace thromboxane AZ from protein and so allow its conversion to thromboxane Bz. Because bone is a heterogeneous tissue, we cannot determine from these experiments which cell type or types in bone were responsible for the biosynthetic and degradative pathways we have identified. It is likely that this will not be answered until homogenous populations of the different types of bone cells are available for in vitro studies. Raisz and his colleagues [6] have found that fetal rat bone in culture produces and releases into medium 6-keto-prostaglandin F,, when treated with
428
rabbit serum plus complement. They suggest on the basis of several biological experiments that prostaglandin Iz may be an important mediator of bone resorption [6]. However, their effects of prostaglandin I2 are small (a 5-12s increase in 45Ca release) or require very high concentrations of prostaglandin II (10m5 M), and show unusual dose-response relationships. Crawford et al.[ 191, in studies of rat osteosarcoma cells, have found that prostaglandin Iz was only about 3 and 10% as active as prostaglandin Ez in stimulating cyclic AMP formation and adenylate cyclase activity, respectively. In their cell culture system, a good correlation has been found between several bone resorption-stimulating factors and cyclic AMP formation or adenylate cyclase activity [ 191. Finally, in the rabbit VX2 carcinoma model in which hypercalcemia and increased bone resorption are mediated by a prostaglandin mechanism [4,5,17], we have found no elevation in plasma of the concentration of 6-keto-prostaglandin F,, despite up to 50-fold increases in the concentration of 13,14-dihydro-15-keto-prostaglandin Ez [ 131. Acknowledgements
This investigation was supported in part by research grants from the USPHS (AM 10206 and CA 17301). We thank Iftekhar Alam for advice on the analytic procedures and Yolanda Santo for expert assistance. L.L is an American Cancer Society Research Professor of Biochemistry (Award PRP-21). This is publication No. 1333 from the Department of Biochemistry, Brandeis University, Waltham, MA, U.S.A. References Klein,
L.G.
(1970)
Tashjian,
D.C.
A.H.,
and
Jr., Voelkel,
Raisz,
E.F..
Levine,
Tashjian,
A.H.,
Jr.. Voelkel,
E.F..
Goldhaber,
VoeIke:,
E.F.,
Tashjian,
A.H.,
Endocrinology
Jr.,
86,1436-1440
L. and
Goldhaber,
P. and
Franklin,
R..
Levine,
P. (1972)
J. Exp.
L. (1974)
Fed.
Wasserman,
E.
and
Med.
Proc.
Levine,
136.
33,
1329-1343
81-86
L. (1975)
Metabolism
24,
973-986 5
Tashjian,
6
Raisz,
A.H..
L.G.,
Jr. (1978)
Cancer
Vanderhoek,
J.Y.,
Res.
38.41384141
Simmons,
H.A.,
Kream.
B.E.
and
Nicolaou,
K.C.
(1979)
Prostaglandins
17.905--914 7
Tashjian.
A.H.,
Jr.. Ivey,
8
Tashjian,
A.H.,
Jr. and
9
Tashjian,
A.H..
eds.).
pp.
173-179,
10
Ivey,
J.L.,
11
Alam,
Jr. (1978)
Wright,
Levine. Alam.
I., VoeIkeI.
14
Pang,
S-S.
Publ.
Corp.,
Sun.
F.F.,
cycIin 16
D.R. K.
12
L. and
New
A.H..
A.H..
L. (1977)
in The
Abstr.
(Horton VA
Anal.
Prostaglandins
J. Clin.
Biochem.
Med.
Jr. and
Prostaglandins
Res.
Arlington,
Jr. (1976)
L. (1979)
Tashjian,
B.M.,
J.R.
A.H.,
Tiss. Inc.,
L. (1978) Biophys.
Commun. J.E..
Invest. 93,
16, 85.
Tarpley,
58,
221-232
966-975 T.M.
and
Davis,
Med.
4, 227-237
W.F.,
1327-1338
339-345
3.295-304
Levine,
L. (1980)
ProstagIandins
ProstagIandins
(Ramwell,
Wang,
MaIik,
P., ed.),
Vol.
3. pp.
41-76,
Plenum
York
Taylor,
(Vane,
Tashjian,
Levine.
CaIcif.
Tasjian,
Levine,
Levine,
Biochem.
Retrieval
I. (1979)
E.F.,
B. and
L. (1978)
in Suppl.
and
and
Alam,
and
Delclos.
Information
I.. Ohuchi,
13
15
J.L., Levine,
and
Jr..
McGuire,
J.C.,
BergstrGm,
Voelkel,
S.,
E.F.
eds.),
and
P. Y-K.
pp.
Levine.
119-131. L.
K.U.
Raven
(1977)
and
Press.
Biochem.
McGiff. New
Biophys.
J.C.
(1979)
in Prosta-
York Res.
Commun.
74,199-
207 17
Tashjian,
18
Fitzpatrick.
19
Crawford,
20
GranstrGm. pp.
4243
A.H..
Jr.. VoeIkel.
F.A. A.,
and
Atkins.
E. and
E.F.
German, D.,
Kindahl,
and
R.R.
and
Martin,
H. (1979)
Levine, (1977) T.J.
L. (1977)
Prostaglandins
ProstagIandins (1978)
Progr.
4th
Biochem. Int.
14,
14,
309-317
881-889
Biophys.
ProstagIandin
Res. Conf.,
Commun. Washington.
82,
1195-1201 DC.
Abstracts,
Biochimica et Biophysics @ Elsevier/North-Holland
Acta, 620 (1980) 418-428 Biomedical Press
BBA 57681
CYCLOOXYGENASE PRODUCTS OF ARACHIDONIC BY MOUSE BONE IN ORGAN CULTURE
EDWARD
F. VOELKEL,
ARMEN
H. TASHJIAN,
Jr. and LAWRENCE
Laboratory of Toxicology, Harvard School of Public cology, Harvard Medical School, Boston, MA 02115, Brandeis University, Waltham, MA 02154 (U.S.A.) (Received
Fehruary
13th,
ACID METABOLISM
LEVINE
Health and Department of Pharmaand Department of Biochemistry,
1980)
Key words: Prostaglandin cyclooxygenase; Throm boxane; (Mouse bone)
Arachidonic
acid; Indomethacin;
Prostacyclin;
Summary The products of endogenous and exogenous arachidonic acid metabolism via the cyclooxygenase pathway in mouse bone in organ culture were identified and quantitated by the use of high performance liquid chromatography and radioimmunoassay. Production of prostaglandins E2, Fza, and IZ from endogenous substrate was stimulated by incubation of bone with epidermal growth factor and the tumor promoter 12-0-tetradecanoyl-phorbol-13-acetate. Addition of arachidonic acid to the culture medium not only resulted in the accuBone metabmulation of prostaglandins Ez, F,, , and IZ but also thromboxane. olized prostaglandins Ez and F,, to their respective 13,14-dihydro-15-ketoderivatives. Prostaglandin I*, measured as 6-keto-prostaglandin F,, was synthesized by bone, and metabolic products of prostaglandin IZ or 6-keto-prostaglandin Fr,, either 6,15diketo-prostaglandin F,, or 13,14dihydro-6,15-diketoprostaglandin F 1Lv,were also detected in the culture media. Formation of cyclooxygenase products of endogenous and exogenous arachidonic acid metabolism (both basal and stimulated) and bone resorption were inhibited by indomethatin. Bone as a tissue responded biochemically not only to exogenous prostaglandins and agents that enhance endogenous prostaglandin production but also to exogenous arachidonic acid by biosynthesis of prostaglandins, prostacyclin and thromboxane. Furthermore, bone metabolized these cyclooxygenase products to their more stable metabolites.
Address Health,
correspondence 665
Huntington
to:
Armen
Avenue,
H.
Boston.
Tashjian, MA
Jr.,
02115.
Laboratory U.S.A.
of Toxicology,
Harvard
School
of Public
419
Introduction Exogenous prostaglandins were first shown to stimulate the resorption of bone in culture in 1970 [ 11. We suggested that it was excessive production of prostaglandin Ez by certain tumors that caused the hypercalcemia of malignant disease in these model systems [ 2-51. Recently, it has been shown that rat bone can metabolize arachidonic acid to several products [6]. Earlier, we had demonstrated that prostaglandin Ez was synthesized by bone and suggested that it acted locally to stimulate bone resorption [ 7-91; among the most active stimulators of prostaglandin Ez production by bone were the tumor-promoting phorbol esters [ 71 and epidermal growth factor [8]. In our previous studies, only prostaglandin E2 had been measured, and de novo biosynthesis was demonstrated by inhibition of the prostaglandin cyclooxygenase by indomethatin. In this report we show that bone can metabolize both endogenous and exogenous arachidonic acid via the cyclooxygenase pathway not only to the primary prostaglandins (e.g., prostaglandins Ez and F2@)but also to prostacyclin and thromboxane. In addition, prostaglandin Ez, prostaglandin F,, and either prostaglandin IZ or 6-keto-prostaglandin F,, are converted by bone to their more stable metabolites. Materials and Methods Organ culture of bone. Neonatal mouse calvaria were placed in organ culture as described previously [lo]. For all the experiments described in this report, the medium was Dulbecco’s modified Eagle’s medium supplemented with 5 mg/ml bovine albumin (albumin medium). Calvaria were preincubated in albumin medium for 24 h before experimental treatments were begun. The medium was changed at this time, and fresh control medium or medium containing a specific treatment was added and incubation continued for an additional 48 h. Bone resorption was determined by measuring the release of calcium from the calvaria into medium at the end of the 48-h treatment period
[lOI* Measurement of medium calcium. The concentration of total calcium in bone culture medium was measured by automatic fluorimetric titration with a Corning calcium analyzer, Model 940. Measurement of metabolites of arachidonic acid. Medium from 3 to 8 separate cultures (6-16 ml) treated identically was pooled and acidified to pH 3.5 by addition of 1.0 M HCl. The acidified medium was processed on small XAD-2 columns (Isolab) which had been prewashed with 20-25 ml of distilled HzO. The initial effluent was rechromatographed on the same column. The column was then washed with 20-25 ml of distilled Hz0 and then eluted with 20-25 ml of ethanol. The ethanol eluate was evaporated to dryness under Nz gas, the residue was taken up in l-2 ml of ethanol and concentrated to about 100 ~1 for fractionation by high-performance liquid chromatography. For high-performance liquid chromatography we used a Waters liquid chromatograph, model ALC/GPC 244 (Milford, MA) equipped with a model 6000A Solvent Delivery System. The column employed was a 30 cm X 3.9 mm Fatty Acid Analysis Column used in the reversed phase mode [ 111. The arachi-
420
donic acid metabolites were eluted isocratically with a solvent system containing CH~CN/C~H~/CH~COOH~H~O (230 : 2 : 1 : 767, v/v). The flow rate was 2.0 ml/min per tube. 60 tubes were collected for each run. Each fraction was lyophilized to dryness and resuspended in 0.5 ml of Gel-Tris buffer (0.01 M Tris-HCI, 0.14 M NaCl, and 1 mg/ml gelatin, pH 7.4) for analysis by radioimmuno~say. The high-perforate liquid ~hromato~aphy elution profiles for Flcu, thromboxane Bz, 6-keto-prostaglandin Fla, 6,15-diketo-prostaglandin F,, prostaglandin Fza, prostaglandin Ez, 13,14-dihydro-15-keto-prostaglandin E, obtained with nonradioactive and 13,14-dihydro-l5-keto-prostaglandin standard preparations and serologic measurements have been described previously [ 11,12] and were, in the experiments reported here, respectively, fraction numbers B-11, 12-16, 18-21, 22-24, 28-32, 36-41, and 43-46. The chromatographic properties of the nonradioactive standards were confirmed by chromatography of the appropriate labeled compounds. Average recovery for each of these me~bolites was 60% [13]. The data for culture medium samples were not corrected for recovery. Under the chromatographic conditions used, arachidonic acid was retarded on the column and did not appear in the effluent until after 90 min (fraction number >60). The total amount of metabolite in each peak was obtained by summing the values for each individual fraction comprising that peak. Radioimmunoassay. The appropriate fractions (20-400 ~1 each) corresponding to the elution volumes of the seven known standards were analyzed quantitatively by selective radioimmunoassays [ 11,121. Assays were performed using the following protocols. 6-Keto-pros~l~d~ F la, thromboxane Bz, prostaglandin FZa, prostaglandin EZ, 13,14-dihydro-15-keto-prostaglandin Fzo, and 13,143Hdihydro-15-keto-prostaglandin E 2 were measured using the homologous labeled compound as tracer and antibody against the homologous ligand. 6,15Diketo-prostaglandin Fi, was measured using 13,14-dihydro-15-keto[ 5,6,8,9, ll,12,14(n)-3H]-prostagl~d~ FZa: and an antiserum which recognizes a 15keto function as immunodominant, anti-13,14-dihydro-l&keto-prostaglandin F2, 1141. Statistical method. The results of each experiment were subjected to analysis of variance, and the standard errors (S.E.) were calculated from the residual error term of that analysis. Where variances were not heterogeneous and the experiments were of the same design, the results were pooled and appropriate statistical significance was calculated. Materials. Epidermal growth factor was purchased from Collaborative Research, Inc., Waltham, MA. 12-O-Tetradecanoyl-phorbol-13-acetate was obtained from Consolidated Midland Corp., Brewster, NY. Bovine albumin and indomethacin were from Sigma Chemical Co., St. Louis, MO. The XAD-2 resin columns were from Isolab, Inc., Akron, OH. Chemicals were of reagent or highperformance liquid chromatography grade and were obtained from Fisher Scientific, Boston, MA. The standard pros~gl~dins and other metabolites of arachidonic acid were gifts from the Upjohn Co., Kalamazoo, MI. The radiolabeled prostaglandins and metabolites used were purchased from New England Nuclear Corp., Boston, MA and the Amersham Co., Chicago, IL.
421
Results We have previously shown that epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate stimulate bone resorption in vitro by a mechanism that involves the production of prostaglandin Ez [ 7,8]. In these earlier experiments prostaglandin E, was measured serologically by direct assay of unfractionated culture medium. In order to determine whether other cyclooxygenase products of arachidonic acid were formed, we cultured mouse calvaria in the absence and presence of epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate and assayed the media, using a combination of high-performance liquid chromatography and radioimmunoassay, for prostaglandin FZcu,prostaglandin Ez, their 13,14-dihydro-15-keto metabolites, and two degradation products of the labile prostacyclin prostaglandin IZ. Positions of the chromatographic peaks for each of the six products are shown in Fig. 1 for control cultures and bones treated with epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate. The quantitative data from three separate experiments are presented in Fig. 2. Because of the uniformity of the design and results in each experiment and the lack of statistical heterogeneity in variances, data from individual experiments were pooled, analyzed statistically, and summarized in Table I. Both 12-O-tetradecanoyl-phorbol-13-acetate and epidermal growth factor stimulated bone resorption as shown by the increase in medium calcium concentration (Table I). 12-0-Tetradecanoyl-phorbol-13-acetate and epidermal growth factor each elicited large (7-8-fold) increases in the production of Gk-PGh,
6,15k-PGb,
PGE,
PGFm
13,14HJ5k-PG&,
]0.07~0
05-
13,14Hi15k-PGI
25!
0.6
23
FRACTION
30
40
45
NUM6ER
Fig. 1. Elution profiles on high-performance liquid chromatography of samples of culture media from control bones and bones treated with epidermal growth factor or 12-O-tetradecanoyl-phorbol-13-acetate. Groups of 5-7 calvaria per treatment were cultured for 48 h in the absence of additives (control) or with epidermal growth factor (50 ng/ml) or 12-O-tetradecanoyl-phorbol-13-acetate (25 ng/ml). At the end of the experiment, medium was harvested from each culture and one or two pools per treatment group were prepared, extracted and chromatographed. The effluent fractions were assayed serologically for the metabolites shown. PG. prostagkxiin.
422 Gk-PGh,
6,15k-PG\,
PGE,
13,i4H~15k~PGF2, 13,14Hi15k-PGE,
20-
C T
E
C
T
E
C
T
E
Fig. 2. Concentrations of six cyclooxygenase metabolites of arachidonic acid in bone culture medium. Groups of 5-7 calvaria per treatment were cultured for 48 h in the absence of additives (C for control) or with 25 ng/ml 12-O-tetradecanoyl-phorbol-13-acetate (T) or 50 &ml epidermal growth factor (E). Medium was harvested, processed and assayed as described in Fig. 1. Three separate experiments of the same design were performed. Medium from each treatment group for each experiment was combined into one or two separate pools. Each point gives the value for one such pool. PG. prostaglandin.
fraction (about 20-30%) of this prostaglandin E2; a small but significant prostaglandin E, was metabolized to 13,14-dihydro-15-keto-prostaglandin Ez by bone. Smaller increases in the production of prostaglandin Fzcv and its 13,14dihydro-15-keto metabolite were noted. In response to 12-O-tetradecanoylphorbol-13-acetate and epidermal growth factor bone elicited a large increase (about lo-fold) in the production of prostaglandin IZ as measured by accumulation of its products 6-keto- and 6,15diketo-prostaglandin F,,. Although not shown in these experiments, we have previously demonstrated [7,8] that the effects of both epidermal growth factor and 12-0-tetradecanoyl-phorbol-13acetate on bone resorption and on the production of prostaglandin E, by bone are inhibited in parallel by low concentrations of the cyclooxygenase inhibitor indomethacin. In the experiments just described, epidermal growth factor and 12-O-tetradecanoyl-phorbol-13-acetate were used as agents to stimulate prostaglandin synthesis from endogenous substrate in bone. Because exogenous arachidonic acid has been shown to enhance bone resorption by a pathway that is blocked by indomethacin [9], we have added exogenous arachidonic acid to bones in culture and have measured and compared thg. products formed to those elicited in response to epidermal growth factor and 12-0-tetradecanoyl-phorbol-13acetate. The positions of the chromatographic peaks for each of the six products described previously as well as that for thromboxane Bz in cultures treated with arachidonic acid alone or arachidonic acid plus indomethacin are shown in Fig. 3. The results presented in Table II demonstrate that arachidonic acid enhanced bone resorption in a dose-dependent manner and that this effect, but not that of exogenous prostaglandin E,, was inhibited completely by indo-
I ACID
IN MEDIUM
OF CALVARIA
CULTURED
WITH
12-O-TETRADECA-
*P < 0.05. ** P < 0.01. ***p < 0.001.
None 12-O-Tetradecanoylphorbol-13acetate (25 ng/ml) EGF (50 &ml)
Treatment
10.1 r 0.32 *** 12.7 t 0.28 ***
7.4 r 0.30
Medium calcium
1.0 1.3
? 0.12 *** f 0.14 ***
0.11 * 0.11
2.5 2.5
r 0.39 * + 0.39 *
0.91 * 0.30
GJB-diketo PGFlo,
@g/ml)
6-k&o-
acid metabolites
PGFlo,
Arachidonic
0.07 ? 0.016 0.09 * 0.019 *
0.02 t 0.016
PGFza
0.29 f 0.04 ** 0.34 r 0.05 **
0.04 + 0.04
PGEz
0.019 r 0.003 0.032 f 0.003 ***
0.013 f 0.003
1 5-keto-PGFZ,
13.14-dihydro-
0.06 ? 0.015 0.10 * 0.017 ***
0.02 * 0.017
13,14-dihy&o15-keto-PGEZ
The data given are mean values pooled from three separate experiments of the same design. Each experiment was analyzed separately and the mean values and variances were not heterogeneous. The values given are mean k S.E.. and statistical significance of the experimental values as compared to control are given and are indicated by asterisks (see below). Groups of 5-7 bones were incubated in albumin medium alone or in medium containing 12-O-tetradecanoyl-phorbol-13-acetate or epidermal growth factor for 48 h. At the end of each experiment, medium was collected and assayed for calcium concentration or for metabolites of arachidonic acid by high performance liquid chromatography and radioimmunoassay. PG. prostaglandin; EGF, epidermal growth factor.
CONCENTRATIONS OF CALCIUM AND METABOLITES OF ARACHIDONIC NOYGPHORBOL-13-ACETATE AND EPIDERMAL GROWTH FACTOR
TABLE
424
6k-PGF,a .___
TX&
6,15k-PGF,,,
PGFza
I L_ 4
13,14Hi15k-PGb,
PGE,
L
30
23
20
15
-10
13,14Hi15k-FGJ
_I L_.ldLLL 40
45
FRAC?TOIV NUM6ER Fig. 3. Elution profiles on high-performance liquid chromatography of culture media from bones treated with arachidonic acid alone or arachidonic acid plus indomethacin. Groups of 6-8 calvaria per treatment were cultured for 48 h with 10 Hglml arachidonic acid (AA) or 10 pg/ml arachidonic acid plus 0.5 pg/ml indomethacin (AA + Indo). At the end of the experiment, medium was harvested, pooled, extracted, chromatographed and assayed. TX, thromboxane; PG. prostaglandin.
methacin. The data from experiments using 5 or 10 pg/ml of exogenous arachidonic acid are presented in Table III. In both experiments arachidonic acid stimulated bone resorption and this effect was inhibited by indomethacin. The patterns of arachidonic acid metabolites formed in the two experiments were qualitatively similar, although the effects were generally larger in Expt. II in which the higher concentration of arachidonic acid was used. Exogenous arachidonic acid stimulated the formation and release of prostaglandin E,, prostaglandin FZa, their 13,14-dihydro-15-keto metabolites and presumably also thromboxane AZ as measured by its product thromboxane Bz. The increases in arachidonic acid metabolites, as the increases in bone resorption, were inhibited by indomethacin. The magnitudes of the effects of 10 pg/ml of arachidonic acid were comparable to those elicited by 12-O-tetradecanoyl-phorbol-13-acetate (25 ng/ml) and epidermal growth factor (50 ng/ml) except that
425 TABLE
II
CONCENTRATIONS
OF
CALCIUM
DONIC ACID IN THE ABSENCE
IN THE
MEDIUM
AND PRESENCE
OF CALVARIA
CULTURED
WITH
ARACHI-
OF INDOMETHACIN
The data given are mean values pooled from four separate experiments of similar design. Each experiment was analyzed statistically and the mean values and variances were not heterogeneous. The values given are means ? S.E., and statistical significance of experimental values as compared to control are given as indicated by the asterisk. Groups of 4-l bones were incubated in albumin medium alone or with the additives indicated for 48 h. At the end of each experiment, medium was collected and the concentration of calcium measured. Treatment
None Indomethacin Prostagiandin E2 Arachidonic acid Arachidonic acid Arachidonic acid and indomethacin Arachidonic acid and indomethacin Prostaghmdin E2 and indomethacin
Dose
Medium calcium
wg/mD
(mg/mD
0.50 0.10 5.0 10.0 5.0 0.50 10.0 0.50 0.10 0.50
6.9 6.6 10.4 8.6 10.1
f r + ? r
0.22 0.22 0.21 0.38 0.25
* * *
6.9 f 0.42 7.4 * 0.28 10.5
* 0.27
*
* P < 0.001.
12-0-tetradecanoyl-phorbol-13-acetate and epidermal growth factor did not enhance the production of thromboxane AZ, at least not to the extent that measurable levels of thromboxane Bz accumulated in the medium (data not shown). Once again there was a large (5-20-fold) increase in the production of prostaglandin I* as measured by accumulation of 6-keto-prostaglandin F,, and this effect was also comparable in magnitude to those elicited by 12-O-tetradecanoyl-phorbol-13-acetate and epidermal growth factor (Table I). Discussion From the results of the experiments presented here we conclude that mouse calvaria, as representative of bone as a tissue, can produce four major cyclooxygenase products of arachidonic acid, i.e., prostaglandin E?, prostaglandin F 201, thromboxane A*, and prostaglandin IZ. Each product was identified by its characteristic elution position from reversed phase high-performance liquid chromatography and quantitated by serologic assay. This approach enables discrimination between the mono- and dienoic prostaglandins of the E series which are not distinguished by the antibodies alone [ 141. Each cyclooxygenase product formed was further metabolized in the bone culture system. Prostaglandin Ez and prostaglandin F,, were converted to their 13,14-dihydro-15keto metabolites indicating the presence in bone of 15-hydroxyprostaglandin dehydrogenase and prostaglandin A13-reductase. Prostaglandin IZ accumulated as 6-keto-prostaglandin F,, by a step which is likely to be nonenzymic [15] as well as 6,15diketo-prostaglandin F,, (or 13,14-dihydro-6,15-diketo-prostaglandin F1,; our high-performance liquid chromatography procedure and radioimmunoassay probably do not distinguish these two metabolites), a product of
OF CALCIUM AND METABOLITES
OF ARACHIDONIC
ACID IN MEDIUM OF CALVARIA
CULTURED
WITH ADDED
ARACHIDONIC
* P < 0.01. ** P < 0.001.
Expt. II None Arachidonic acid (10 /.&g/ml) Arachidonic acid + indomethacin
Expt. I None Arachidonic acid (5 pg/ml) Arachidonic acid + indomethacin
Treatment
*
6.6 f 0.47 10.9 * 0.47 ** 6.8 f 0.52
6.9 f 0.40 8.6 ? 0.40 6.9 ? 0.43
(mg/dI)
Medium calcium
0.04 0.90 0.03
0.03 0.15
0.84 1.4 1.1
6.15-diketoPGFlo,
G-ketoPGFla
Thromboxane B2
Arachidonic acid metabolites @g/ml)
<0.005 0.12 0.02
0.013 0.025 <0.005
PGFza
0.025 0.06 0.03
PGE,
PGF2,
13.14dihydro15-k&o-
13,14dihydro15ketoPGE2
The data from two separate experiments are presented. Values for medium calcium are means * S.E. for 6-8 bones per group. and statistical significance of the experimental values compared to control are given as indicated by the asterisks. Values for arachidonic acid metabolites were determined on pooled samples of medium and no S.E. was calculated. Bones were incubated in albumin medium alone or in medium containing added arachidonic acid (5 or 10 &ml in Expt. I and II, respectively) alone or arachidonic acid plus indomethacin (0.5 pg/mI in both experiments). At the end of each experiment, medium was collected and assayed for calcium concentration or for metabolites of arachidonic acid by high performance liquid chromatography and radioimmunoassay. PG, prostaglandin.
CONCENTRATfONS ACID
TABLE III
427
15-hydroxyprostaglandin dehydrogenase (and A13-reductase) activity. When bones were stimulated to resorb by treatment with 12-O-tetradecanoyl-phorbol-l3-acetate or epidermal growth factor (Table I), concentrations of prostaglandin Ez, prostaglandin Fzcu, and prostaglandin IZ (measured as 6-keto-prostaglandin F,,) increased 4-lo-fold. Relatively small fractions of the stimulated synthesis of prostaglandin E2 and prostaglandin Fzo, accumulated as their respective 13,14-dihydro-15-keto metabolites, about 30% in the case of prostaglandin Fzcv and about 25% for prostaglandin E2. These values are considerably lower than the accumulation of 13,14-dihydro-l5-keto-prostaglandin Ez or prostaglandin F,, in the plasma of mice [16] or rabbits [17] bearing prostaglandin-producing tumors. It is possible that our measurements of 13,14-dihydro-15-keto-prostaglandin Ez underestimate the’ production of this metabolite because it has been reported to be degraded further in aqueous medium to ll-deoxy-13,14-dihydro-l5-keto-11,16-cycle-prostaglandin Ez [ 201. The values for accumulated 6,15-diketo-prostaglandin F,, were high and were calculated from low amounts of inhibition of the immune system. These results may reflect predominance in bone of the 15-hydroxyprostaglandin dehydrogenase pathway of prostaglandin IZ metabolism. Alternatively, our quantitative analysis of the 6,15-diketo metabolite of prostaglandin IZ, which uses a heterologous immune system, may overestimate the true value of 6,15-diketo-prostaglandin F1, due to cross-reactions with our antibody by additional unidentified metabolites. The absolute values will need to be confirmed by a homologous radioimmunoassay for 6,15diketo-prostaglandin F,, which is being developed. When bones were stimulated to resorb with 10 pg/ml of exogenous arachidonic acid (Table III), the general pattern of cyclooxygenase products formed was similar to that generated from endogenous substrate exept that significant amounts of thromboxane Bz were detected. It is possible that the K, for the thromboxane synthetase in bone is high and that insufficient substrate was generated from endogenous sources even after stimulation with 12-O-tetradecanoyl-phorbol-13-acetate and epidermal growth factor. The measured thromboxane Bz was not due to contamination by arachidonic acid because arachidonate is greatly retarded on the column and, even more importantly, its appearance was eliminated by incubation with arachidonic acid plus indomethacin (Table III). It is possible that thromboxane B2 was not detected in experiments using endogenous arachidonic acid substrate because the low amounts of thromboxane AZ generated were bound to protein in the medium and thus protected from conversion to thromboxane Bz [18]. In the presence of exogenous substrate in the medium, it is possible that the arachidonic acid could displace thromboxane AZ from protein and so allow its conversion to thromboxane Bz. Because bone is a heterogeneous tissue, we cannot determine from these experiments which cell type or types in bone were responsible for the biosynthetic and degradative pathways we have identified. It is likely that this will not be answered until homogenous populations of the different types of bone cells are available for in vitro studies. Raisz and his colleagues [6] have found that fetal rat bone in culture produces and releases into medium 6-keto-prostaglandin F,, when treated with
428
rabbit serum plus complement. They suggest on the basis of several biological experiments that prostaglandin Iz may be an important mediator of bone resorption [6]. However, their effects of prostaglandin I2 are small (a 5-12s increase in 45Ca release) or require very high concentrations of prostaglandin II (10m5 M), and show unusual dose-response relationships. Crawford et al.[ 191, in studies of rat osteosarcoma cells, have found that prostaglandin Iz was only about 3 and 10% as active as prostaglandin Ez in stimulating cyclic AMP formation and adenylate cyclase activity, respectively. In their cell culture system, a good correlation has been found between several bone resorption-stimulating factors and cyclic AMP formation or adenylate cyclase activity [ 191. Finally, in the rabbit VX2 carcinoma model in which hypercalcemia and increased bone resorption are mediated by a prostaglandin mechanism [4,5,17], we have found no elevation in plasma of the concentration of 6-keto-prostaglandin F,, despite up to 50-fold increases in the concentration of 13,14-dihydro-15-keto-prostaglandin Ez [ 131. Acknowledgements
This investigation was supported in part by research grants from the USPHS (AM 10206 and CA 17301). We thank Iftekhar Alam for advice on the analytic procedures and Yolanda Santo for expert assistance. L.L is an American Cancer Society Research Professor of Biochemistry (Award PRP-21). This is publication No. 1333 from the Department of Biochemistry, Brandeis University, Waltham, MA, U.S.A. References Klein,
L.G.
(1970)
Tashjian,
D.C.
A.H.,
and
Jr., Voelkel,
Raisz,
E.F..
Levine,
Tashjian,
A.H.,
Jr.. Voelkel,
E.F..
Goldhaber,
VoeIke:,
E.F.,
Tashjian,
A.H.,
Endocrinology
Jr.,
86,1436-1440
L. and
Goldhaber,
P. and
Franklin,
R..
Levine,
P. (1972)
J. Exp.
L. (1974)
Fed.
Wasserman,
E.
and
Med.
Proc.
Levine,
136.
33,
1329-1343
81-86
L. (1975)
Metabolism
24,
973-986 5
Tashjian,
6
Raisz,
A.H..
L.G.,
Jr. (1978)
Cancer
Vanderhoek,
J.Y.,
Res.
38.41384141
Simmons,
H.A.,
Kream.
B.E.
and
Nicolaou,
K.C.
(1979)
Prostaglandins
17.905--914 7
Tashjian.
A.H.,
Jr.. Ivey,
8
Tashjian,
A.H.,
Jr. and
9
Tashjian,
A.H..
eds.).
pp.
173-179,
10
Ivey,
J.L.,
11
Alam,
Jr. (1978)
Wright,
Levine. Alam.
I., VoeIkeI.
14
Pang,
S-S.
Publ.
Corp.,
Sun.
F.F.,
cycIin 16
D.R. K.
12
L. and
New
A.H..
A.H..
L. (1977)
in The
Abstr.
(Horton VA
Anal.
Prostaglandins
J. Clin.
Biochem.
Med.
Jr. and
Prostaglandins
Res.
Arlington,
Jr. (1976)
L. (1979)
Tashjian,
B.M.,
J.R.
A.H.,
Tiss. Inc.,
L. (1978) Biophys.
Commun. J.E..
Invest. 93,
16, 85.
Tarpley,
58,
221-232
966-975 T.M.
and
Davis,
Med.
4, 227-237
W.F.,
1327-1338
339-345
3.295-304
Levine,
L. (1980)
ProstagIandins
ProstagIandins
(Ramwell,
Wang,
MaIik,
P., ed.),
Vol.
3. pp.
41-76,
Plenum
York
Taylor,
(Vane,
Tashjian,
Levine.
CaIcif.
Tasjian,
Levine,
Levine,
Biochem.
Retrieval
I. (1979)
E.F.,
B. and
L. (1978)
in Suppl.
and
and
Alam,
and
Delclos.
Information
I.. Ohuchi,
13
15
J.L., Levine,
and
Jr..
McGuire,
J.C.,
BergstrGm,
Voelkel,
S.,
E.F.
eds.),
and
P. Y-K.
pp.
Levine.
119-131. L.
K.U.
Raven
(1977)
and
Press.
Biochem.
McGiff. New
Biophys.
J.C.
(1979)
in Prosta-
York Res.
Commun.
74,199-
207 17
Tashjian,
18
Fitzpatrick.
19
Crawford,
20
GranstrGm. pp.
4243
A.H..
Jr.. VoeIkel.
F.A. A.,
and
Atkins.
E. and
E.F.
German, D.,
Kindahl,
and
R.R.
and
Martin,
H. (1979)
Levine, (1977) T.J.
L. (1977)
Prostaglandins
ProstagIandins (1978)
Progr.
4th
Biochem. Int.
14,
14,
309-317
881-889
Biophys.
ProstagIandin
Res. Conf.,
Commun. Washington.
82,
1195-1201 DC.
Abstracts,