Clinical Endocrinology (1979) 11,491-495.

M E T A B O L I S M OF I N T R A V E N O U S L Y A D M I N I S T E R E D CHOLECALCIFEROL IN M A N J. M . B A R R A G R Y , M . W . F R A N C E , B. J . B O U C H E R R . D. C O H E N

AND

Metabolic and Endocrine Unit and Department of Clinical Chemistry. The London Hospital Medical College, London E l (Received5 January 1979; revised30 April 1979; accepted 14 May 1979)

SUMMARY

Following intravenous injection of ('H) cholecalciferol into healthy subjects the disappearance of label from the plasma was followed by the reappearance ('rebound') of ('H) radioactivity associated exclusively with cholecalciferol. Lipoprotein fractionation of plasma revealed an increasing association of label with protein- rather than lipoprotein during the rebound phase. We conclude that the rebound of plasma radioactivity reflects the transfer of label from lipoprotein to Vitamin D-binding globulin in the liver followed by its release into plasma. Relatively few studies of the behaviour of intravenously administered cholecalciferol have been made in man (Avioli et al., 1967; Mawer et al., 1971), and little is known about its fate in the first 8 hours after injection. In an attempt to investigate the intestinal absorption of cholecalciferol using a double isotope technique, previously undescribed information about the fate of intravenously administered cholecalciferol in man was obtained and is reported here. SUBJECTS Five healthy males (30-57 years) whose plasma 25-hydroxyvitamin D (25(OH)D) levels ranged from 36 to 75 nmol/l (normal range 20-100 nmol/l) were studied. Each fasted for 12 hours before the study. Venous blood samples were drawn through an indwelling cannula before and at intervals after administration of label, and collected in heparinized tubes. Experiment (f): Subjects A and B consumed a 50 g lipid meal containing 0.6 pCi of (I4C) cholecalciferol (total vitamin D content 300 I.U.). Simultaneously 1 pCi of (3H) cholecalciferol (1 5 1.U. of vitamin D) dispersed in 3 ml of 10% Intralipid was injected intravenously over a 10 min period. Correspondence: Professor R. D. Cohen, The London Hospital, London El IBB. 0300-0664/79/1100-0491602.00

0 1979 Blackwell Scientific Publications 49 1

J . M . Barragrj) et a l .

492

Experinten1 (2): Subjects C and D received 1 pCi of (3H)cholecalciferol in 3 ml of lo1:;, Intralipid intravenously. No lipid meal or oral label was administered. Experiment ( 3 ) ; An intravenous dose of 3 pCi of ('H) cholecalciferol in 3 ml of 10% lntralipid was administered to subject E without a lipid meal or oral label. Plasma samples drawn were subjected to lipoprotein fractionation. METHODS ( l a , 2 ~ ~ ( n ) - ~vitamin H) D,, 12.3 Ci/mmol, and (14-4) vitamin D,, 32.3 mCi/mmol, were obtained from the Radiochemical Centre (Amersham, U.K.). Lipid extraction of plasma in ch1oroform:methanol was performed using the method of Folch e f al. (1957) and labelled chloecalciferol was separated from its polar metabolites (mainly 25(OH)D) using silica gel column chromatography (Barragry el al., 1978). Radioactive counting was performed on a Nuclear Enterprises NE83 12 Liquid Scintillation Spectrometer. The plasma content of radioactivity was calculated and expressed as a percentage of the dose of label administered assuming a plasma volume of 5% of body weight. Plasma lipoprotein fractionation was performed by ultracentrifugation (Hatch, 1968) with separation of chylomicra, VLDL, LDL and HDL from the other plasma proteins. RESULTS E.wperiment (1): In subject A there was an initial phase of rapid disappearance of ( 3 H ) cholecalciferol from the plasma so that only 3.5% of the dose administered was present in the plasma 10 min after intravenous injection. This was followed by an abrupt reappear-

J 400

500

Ti me ( rnin )

Fig. 1. Experiment (I), subject A: plasma response to the intravenous administration of ('H) cholecalciferol and the oral administration of (I4C) cholecalciferol. 0-+ (3H) cholecalciferol. 0+ ('H) polar metabolites, m(14C) cholecalciferol, m- - 4 (I4C) polar metabolites. ~

Intravenous ckolecalcijerol in inan

49 3

ance of (3H) cholecalciferol in the plasma to reach a peak value of 18% at 3 hours, falling rapidly and then more gradually to 7.5% at 7 h. Following oral administration of (I4C) cholecalciferol the plasma ( I4C) cholecalciferol response rose gradually to 5% of the administered dose at 7 hours. The concentration of polar metabolites of (I4C) and (3H) cholecalciferol in the plasma increased steadily without a rebound to reach values of 8% and 10% of the dose administered at 7 hours (Fig. I). Qualitatively similar results were obtained in subject B. Experirnenr (2): A similar response was seen in subjects C and D following administration of ('H) cholecalciferol alone. In subject C there was an initial sharp decline in plasma ('H) cholecalciferol from 250/, of dose at 2 niin to 3.5% at 20 min, rebounding to 20% at 5 hours and declining to 6%at 10 hours. In subject D 45:4 of the administered dose of (3H) cholecalciferol was present in the plasma at 2 rnin but had diminished to undetectable levels at 30 min. There was then an abrupt rebound to 36% at 2 hours, falling gradually to 18% at 5 hours. In both subjects a progressive rise in plasma (3H) polar metabolite response without any rebound was seen. E.vperirnen/ ( 3 ) : In subject E following intravenous injection of (3H) cholecalciferol there was an initial phase of rapid removal of label, falling from 57% of dose at 5 min to 15.8% at 35 min, a rebound of plasma (3H)cholecalciferol to 22.8% at 60 min, decreasing to 16% at 3 hours. On lipoprotein fractionation (Table 1) plasma ('H) cholecalciferol was seen to reside in the chylomicron, HDL and protein fractions and a rebound of radioactivity occurred in all three fractions. Thus chylomicron-associated ('H) cholecalciferol rebounded at 60 rnin to 367; of its level at 5 rnin, HDL-associated label to 16"d and protein-associated label at 74:/, of their respective levels at 5 min. ('H) cholecalciferol associated with protein accounted for 39% of the total amount of ('H) cholecalciferol present in the plasma 5 min after injection and for 7lx at 60 min. DISCUSSION This study demonstrates that the disappearance of OH) cholecalciferol from the plasma Table 1 . Experiment (3): Distribution of ('H) cholecalciferol among plasma lipoproteins and protein after intravenous injection. Results are expressed as a percentage of dose of label administered. Lipoprotein fractionation Total HDL Protein radioactivity Time Chylomicron VLDL LDL %dose %dose %dose %dose "/,dose :;/,dose (min)

5 10

4.6 2. I

25 35 45 60 90 I20 I no

0.4 0.2 I .7 1.6 0.4 0.0

1.o

494

J . M . Barragry et al.

after an intravenous dose is interrupted by the reappearance of label in the plasma within the first few hours. In the subjects taking a simultaneous oral dose of (I4C)cholecalciferol with a lipid meal there was no evidence that the rebound was different from that seen in subjects given intravenously OH) cholecalciferol alone. This rebound is variable in its onset, magnitude and duration, and chromatographic separation of vitamin D metabolites shows that the rebound of plasma (3H) radioactivity is associated exclusively with cholecalciferol. On lipoprotein fractionation of the plasma it can be seen that there is an increasing association of (3H) cholecalciferol with non-lipid-containing protein rather than lipoprotein and this association is greatest at the peak of the rebound. This pattern closely resembles the distribution of radioactivity among serum proteins in the rat after intravenous injection of (3H) cholecalciferol (Rikkers & De Luca, 1967). On the basis of these findings it is reasonable to infer that the non-lipoprotein fraction of plasma bearing the label contains the specific vitamin D transport protein, vitamin D-binding globulin (VDBG). This a-globulin functions as a carrier and solubilizer of cholecalciferol and its polar active metabolites (Haddad & Walgate, 1976). The initial phase of rapid disappearance of label from the plasma is probably the result of rapid dilution of label in the vascular and extravascular spaces combined with uptakes by tissues, especially the liver (Neville & De Luca, 1966), which is the site of VDBG synthesis (Haddad & Walgate, 1976). A previous study has revealed a similar multiphasic pattern of decay of plasma ('H) cholecalciferol radioactivity in human subjects following intravenous injection of (3H) (Avioli el al.,1967). This was considered to reflect different metabolic pools with varied rates of turnover or certain physiological processes essential for the transport of the vitamin. A further study suggested that in the rat the rebound of plasma (3H)radioactivity was the result of hepatic 25-hydroxylation and release of (jH)25(OH)D (Ponchon & De Luca, 1969). However vitamin D deficient rats were used in the latter study and under these circumstances conversion into 25(OH)D is the major metabolic path for the disposal of administered cholecalciferol (Mawer el al.,1971). In contrast, in the vitamin D replete state the initial disappearance of administered cholecalciferol is caused by distribution of unchanged cholecalciferol to the tissues, including the liver (Mawer ef al., 1972). All subjects in the present study were vitamin D replete and levels of (3H)polar metabolites (including 25(OH)D) were low during the phase of rebound of (3H)cholecalciferol. We conclude that in vitamin D replete man, administered OH) cholecalciferol is transported by lipoprotein (Schachter el af.,1964) to the liver where much of it is transferred to protein, presumably VDBG, and is released again into the plasma where it largely accounts for the observed rebound of radioactivity. The subsequent decay of label is probably due partly to uptake by vitamin D stores and partly to metabolic transformation into its 25-hydroxymetabolite in the liver. It is evident from the foregoing that estimation of the plasma half-life of cholecalciferol can be calculated only from approximately 8 hours onwards. ACKNOWLEDGEMENT

We wish to thank Mr L. Owen, Radio-pharmacist, The London Hospital, for preparation of labelled cholecalciferol. REFERENCES AVIOLI, L.V., LEE, S.W., McDONALD, J.E., LUND, J . & DE LUCA. H.F. (1967) Metabolism of vitamin

Intravenous cholecalciferol in man

49 5

D3-’H in human subjects: Distribution in blood, bile, faeces and urine. Journalof Clinicallnuestigation, 46, 983-992. BARRAGRY, J.M., FRANCE, M.W., CORLESS, D., GUTPA, S.P., SWITALA, S., BOUCHER, B.J. & COHEN, R.D. (1978) Intestinal cholecalciferol absorption in the elderly and in younger adults. Clinical Science and Molecular Medicine, 55, 213-220. FOLCH, J., LEES, M. & SLOANE-STANLEY, G.H. (1957) A simple method for the isolation and purification of total lipids from animal tissues. Journal of Biological Chemistry, 226,497-509. HADDAD, J.G. & WALGATE, J. (1976) 25-hydroxyvitamin D transport in human plasma. Journal of Biological Chemistry, 251,48034809. HATCH, F.T. (1968) Practical methods for plasma lipoprotein analysis. Aduances in Lipid Research, 6, 1-68. MAWER, E.B., LUMB, G.A., SCHAEFER, K. & STANBURY, S.W. ((1971) The metabolism of isotopically labelled vitamin D3 in man. Clinical Science. 40.39-53. MAWER, E.B., BACKHOUSE, J., HOLMAN, C.A., LUMB, G.A. & STANBURY, S.W. (1972) The $istrihution and storage of vitamin D and its metabolites in human tissues. Clinical Science, 43,413-431. NEVILLE, P.F. & DE LUCA, H.F. (1966) The synthesis of ( U 3 H ) vitamin D’ and the tissue localization of a 10 I.U. dose per rat. Biochemistry, 5,2201-2209. PONCHON, G . ’ &DE LUCA, H.F. (1969) The role of the liver in the metabolism of vitamin D. Journal of Clinical Inuestigarion. 48, 1273-1279. RIKKERS, H. & DE LUCA, H.F. ( I 967) An in uiuo study of the carrier proteins of 3H-vitamins D3 & D4 in rat serum. American Journal of Physiology, 213,38&386. SCHACHTER, D., FINKELSTEIN, J.D. & KOWARSKI, S. (1964) Metabolism of vitamin D: I , Preparation of radioactive vitamin D and its intestinal absorption in the rat. Journal of Clinical Investigation, 43, 787-795.