IT HAS BEEN previously demonstrated

Effects of Prednisone on Vitamin D Metabolism in Man LOUIS V. AVIOLI,1 STANLEY J. BIRGE, AND SOOK WON LEE Department of Medicine, The Jewish Hospital ...
Author: Jessica Park
0 downloads 2 Views 653KB Size
Effects of Prednisone on Vitamin D Metabolism in Man LOUIS V. AVIOLI,1 STANLEY J. BIRGE, AND SOOK WON LEE Department of Medicine, The Jewish Hospital of St. Louis and Washington University School of Medicine, St. Louis, Missouri 63110 ABSTRACT. Vitamin D3-3H metabolism was studied in 4 normal adult subjects before and during periods of prednisone administration. The biological activity of plasma vitamin D 3 and its metabolites was assayed by in vitro measurements of 45Ca accumulation by duodenal slices. In each instance prednisone induced a decrease in the plasma half-time of vitamin D3-3H, a decrease in the accumulation of a biologically active vitamin D 3 metabolite and the rapid appearance of chloroform-soluble biologically inac-

I

T HAS BEEN previously demonstrated in man that glucocorticoids have a dramatic effect in lowering serum calcium in situations characterized by vitamin D overdosage (1, 2) and/or hypersensitivity to vitamin D (3, 4), apparently by reducing intestinal calcium absorption. The results of recent investigations in animals also reveal that cortisol administration decreases (5-9) and bilateral adrenalectomy increases (10) intestinal calcium transport and absorption. Although evidence obtained in clinical and animal experimentation indicates that glucocorticoids are "antagonistic" to the action of vitamin D, it is still uncertain whether this "antagonism" is due to direct interference with the metabolism of vitamin D or to glucocorticoid-induced metabolic changes in certain tissues ordinarily responsive to the vitamin. The purpose of this report is to present evidence that in man the administration of a synthetic glucocorticoid,

tive metabolites without concomitant alterations in water-soluble vitamin D 3 metabolites. These experiments support the conclusion that the observed antagonism of glucocorticoids to the action of vitamin D in man is related to the rapid turnover of vitamin D3, a diminished production of a biologically active vitamin D metabolite and the subsequent decrease in the intestinal absorption of calcium. (J Clin Endocr 28: 1341, 1968)

prednisone, leads to a derangement in vitamin D metabolism and a diminished production of an extremely potent biologically active metabolite which normally stimulates the intestinal absorption of calcium.

Materials and Methods

All subjects were studied on a clinical research center after 14 days of adaptation to diets containing 800 U of vitamin D/day. After a 16-hr overnight fast and 1 hr before breakfast, 10 /xCi of radiochemically pure 1,23 H-vitamin D3 with a specific activity of 80 /zCi/mg was dissolved in 1.0 ml of absolute ethanol, administered orally to 4 normal healthy adult volunteers with a calibrated syringe, and rinsed down with 250 ml of milk. Serial heparinized blood samples were obtained at 5, 15, 30 and 45 min and at 1, 2, 4, 8, 12, 16 and 24 hr. For the subsequent 4 days blood was obtained at 12-hr intervals. Two weeks following the initial control vitamin D33 H study, when plasma, urine and fecal radioactivity measurements were below background, each subject was placed on 30 mg prednisone/ day for 10 days. On the fifth day of prednisone therapy a repeat 5-day oral vitamin D3-3H study was performed. Received April 2, 1968; accepted May 24. All plasma samples were subjected 3to comSupported in part by Grants AM 11247 and AM bustion for determination of total H, and 11674 from the National Institute of Arthritis and Metabolic Diseases and Clinical Research Center methanol-chloroform extraction and thin-layer chromatographic procedures according to techGrant FR 00036. 1 Career Research Development Awardee (6- niques published previously (11). In each case aliquots of the chloroform-soluble extract were K3-GM-22 676-03). 1341

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 January 2017. at 04:22 For personal use only. No other uses without permission. . All rights reserved.

AVIOLI, BIRGE AND LEE

1342

Before Prednisone III (79%) DPM 3000

IV(20%)

1500

During Prednisone DPM 3000

1500

0

in (65%)

IVa IV (13%) (14%)

V (8%)

A,

20 40 60 FRACTIONS (10ml)

FIG. 1. A. Silicic acid chromatography of the 24-hr plasma chloroform extract from subject JH. The chromatogram was developed with a hyperbolic gradient elution of diethyl ether in n-hexane on multibore columns according to previously published techniques (11). Tritium was measured in aliquots of all column fractions. B. Effect of prednisone on the distribution of 24-hr plasma chloroform extract from subject JH. The chromatogram was developed as noted above in (A). Tritium was measured in aliquots of all column fractions. applied across the origin of silica gel G thinlayer plates which were then chromatographed in a solvent of 10% acetone in n-hexane (v/v). After drying, the plates were sprayed with 0.20% KMnO4 in 1% Na2CO3 to locate the crystalline vitamin D3 nonradioactive marker which had also been placed at the origin at each side of the plate with the radioactive sample. The radioactivity was then located by scraping off successive 0.5 cm segments of the silicic acid with a microscope slide and the scrapings placed in a counting vial containing 18 ml of a liquid scintillation counting solution. The latter consisted of 3.0 g of 2,5-diphenyloxazole (PPO) and 100 mg of 1,4-bis (5-phenyl-2-

Volume 28

oxazoyl)-benzene (POPOP) per liter of toluene. As noted previously (11), the crystalline marker vitamin D3 always migrated the same distance on the silica gel G chromatogram whether chromatographed alone or with a lipid extract of the biological sample. The radioactivity in chloroform extracts of plasma samples migrating with the same Rf as the stable vitamin D markers was labeled "radioactive vitamin D 3 ." Chloroform extracts of 24-hr plasma samples were also dried in a flash evaporator under a stream of nitrogen, the residue dissolved in Skellysolve B (a petroleum fraction boiling at 65-67 C), applied to multibore silicic acid columns and chromatographed, using diethyl ether-petroleum ether gradient elution procedures previously described (11). Aliquots of each fraction were removed, evaporated under a stream of nitrogen and dissolved in 18 ml of a toluene liquid scintillation counting solution which contained 2 g PPO and 100 mg POPOP per liter. As noted in Fig. 1A, 24 hr following an oral dose of vitamin D3, 3 components were normally obtained with this chromatographic procedure. Identical chromatographic fractions from the 24-hr plasma samples of the 4 patients were pooled and the solvent evaporated under a stream of nitrogen. The residue was dissolved in ethanol-propylene glycol and administered by stomach tube to rachitic rats. The biological activity of the various chromatographic fractions was then assayed according to the oxygendependent accumulation of 45Ca by the duodenal slices of rachitic rats as previously described by Schachter and co-workers (12). The bioassay data are described as "net" accumulations of calcium in duodenal slices incubated under O2. The oxygen-dependent accumulation of calcium (/zmole/g45 slice) was calculated from the expression: [( Ca concentration in slices under O2 (cpm/g) —45Ca concentration in slices under45 N2 (cpm/g)] X l / [initial specific activity of Ca in the medium (cpm/jiimole Ca)]. The dosages of the chromatographed material assayed by this technique were calculated on the assumption that the specific activity of the metabolites within the various chromatographed fractions was the same as the original radioactive vitamin D3.

Results In the control studies and those performed during prednisone administration, plasma vitamin D3-3H rose gradually following the oral D3-3H dose, attaining peak levels within seven to eight hours in

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 January 2017. at 04:22 For personal use only. No other uses without permission. . All rights reserved.

September 1968 PREDNISONE EFFECTS ON VITAMIN D METABOLISM TABLE 1. Plasma half-time of vitamin D3-3H in normal subjects before and during prednisone administration* Subject

Before prednisone (hi-)

During prednisone (hr)

RM JH MB MT

27 30 24 23

15 18 10 12

* Vitamin D3-3H in chloroform-soluble plasma extracts was isolated on silica gel G thin-layer plates. The plates were chromatographed in a solvent of 10% acetone in n-hexane (v/v) as previously described (11). The radioactivity migrating with the same Rf as standard crystalline nonradioactive vitamin D 3 markers was labeled "radioactive vitamin D3." Half-times were calculated by analysis of semilogarithmic plots of plasma "radioactive vitamin D 3 " (expressed as % administered dose) disappearance with time.

all subjects and decreasing exponentially thereafter for the remainder of the study. As shown in Table 1, in each instance prednisone induced a significant fall in the plasma half-time of vitamin D3-3H. The results of silicic acid chromatography of chloroform extracts obtained from 24-hour plasma samples in the untreated subjects are noted in Table 2. The material in peak 3 has been previously isolated from both rat (14) and human (11) plasma and identified as unaltered vitamin D3 with potent biological antirachitic activity. As noted in Fig. 2, the substance in peak 3 from human plasma is as effective as crystalline vitmin D3 in restoring the active calcium transport mechanism in duodenal slices of rachitic rats. The polar vitamin D3-3H metabolite in peak 4 previously isolated from both rat (14) and human (11, 12) plasma has also been shown to possess antirachitic activity. Recently, DeLuca and co-workers have demonstrated that the peak 4 metabolite fraction obtained from rat plasma also acts twice as fast as the parent vitamin D3 in stimulating the intestinal transport of calcium (15). As illustrated in Fig. 2, the oxygen-dependent accumulation of calcium was also increased by peak 4 substance obtained from human plasma with

1343

0.4r Peak IV

Vit. D 3 ° Peek III a PeaklVa PeakV

I 2 3 DAYS AFTER TEST DOSE

FIG. 2. CVdependent accumulation of Ca by duodenal slices at various intervals following 0.25 IU of peak 3 of crystalline3 vitamin D 3 and amounts (vitamin D3- H) and vitamin D3-3H metabolites (peaks 4, 4a, 5) corresponding to 0.25 IU of the parent vitamin according to radioactivity. The amount of each compound tested was estimated on the basis of the specific activity of the parent D33 H: 176,000 dpm=l ng. Duodenal slices from vitamin D-depleted rats were tested using the experimental methods described by Schachter and co-workers (12) except that the incubation medium was modified to contain 2X10"4M CaCU and 0.004M KC1.

observed increments greater than those achieved by crystalline vitamin D3 or peak 3. Heretofore, peak 5 (Table 2) has not been observed in man following intravenous vitamin D3-3H administration (11). This substance presently appears to represent a metabolite unique to oral vitamin D3-3H studies. As noted in Fig. 2, the material in peak 5 was biologically inactive. The increment in vitamin D3 fractional turnover rate observed during prednisone administration was associated with decreased plasma concentrations of peaks 3 and 4 and a rapid appearance of inactive chloroform-soluble vitamin D3 metabolites (Fig. IB and 2). Comparative analysis of total radioactivity in whole plasma and chloroform-soluble extracts of plasma in both control and prednisone-treated states were similar, with over 96% of the plasma radioactivity chloroform-soluble in each instance. During the control studies performed before prednisone administration,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 January 2017. at 04:22 For personal use only. No other uses without permission. . All rights reserved.

AVIOLI, BIRGE AND LEE

1344

Volume 28

TABLE 2. Effect of prednisone on the distribution of radioactivity in plasma chloroform extracts during silicic acid chromatography on multibore columns* Subject —

% Total radioactivity Peak 3

Peak 4a

RM JH MB MT

76.8 79.0 70.6 75.1

0 0 0 0

Mean ±SE

75.4 ±1.8

0 0

RM JH MB MT

64.5 65.0 62.9 66.0

17. 0 13. 0 14. 5 10. 0

During prednisone 10.7 14.0 11.2 12.7

Mean ±£E

64.6 ±0.65

13. 6 ±1. 5

12.2 ±0.75

P

Peak 4

Before prednisone 21.6 20.0 26.8 23.7 23.0 ±1.5