Trm JOURNAL cm BIOLOGICAL CHEMISTRY Vol. 247, No. 8, Issue of April 25, pp. 232&2335, lQi2

Printed in U.S.A.

The Metabolism of Vitamin D, in the Chick*

(Received for publication, September 13, 1971)

MARK R,. HAUSSLER~ AND HOWARD RASMUSSEN

From the Department of Biochemistry, University of Pennsylvania, Philadelphia, Pennsylvania 19104

SUMMARY

Physiological doses of radioactive vitamin D3 are adminis- tered to vitamin D-deficient chicks 16 hours prior to the preparation of lipid extracts of small intestine, plasma, kidney, and bone. Analysis of the vitamin D metabolite pattern is carried out via column chromatography on silicic acid, Sepha- dex LH-20, and Celite (liquid-liquid partition), countercurrent distribution and treatment with periodic acid. The predomi- nant form of the vitamin in the intestine is a polar metabolite (Peak 4BI) which is homogeneous in all four separation systems, migrates in the area of known dihydroxy-D3-vita- mins, loses 46% of its tritium in the course of metabolic formation from [l(~,Za-3Hz]vitamin Da and is insensitive to cleavage by periodate. These data are consistent with the recent identification of Peak 4Br as 1 ,25-dihydroxy-vitamin Da by Holick, Schnoes, and DeLuca (1971) (Proc. Naf. Acad. Sci. U. S. A. 68, 803) and the original proposal by Haussler, Myrtle, and Norman (1968) (J. BioZ. Chem. 243, 4055) that this metabolite is the active form of vitamin D in the intestine.

The major metabolite present in plasma, kidney, and bone is 25-hydroxy-vitamin Da, but significant amounts of 1,25- dihydroxy-vitamin DS are detected in all three sites. The plasma dihydroxy-vitamin Do-metabolite profile differs markedly from that of the kidney and bone in that the pre- dominant species is 25,26-dihydroxy-vitamin Da, with 21,25- dihydroxy-vitamin DB also being present in measurable quantities. Virtually all dihydroxy-vitamin DB in kidney and bone exists as the 1,25-dihydroxy derivative. The presence of 1,25-dihydroxy-vitamin D3 in kidney strengthens the proposal that it is produced metabolically from 25-hydroxy- vitamin Da by a renal enzyme. The association of 1,25- dihydroxy-vitamin Da with bone and its considerable activity in promoting skeletal dissolution, raises the question as to whether 25-hydroxy-vitamin D3 or the 1,25-dihydroxy-sterol represents the active form directing bone resorption.

Considerable effort has been devoted in the past 5 years to the study of the metabolism of vitamin Da. Blunt et al. (1, 2) first identified the major metabolite of vitamin Da in plasma as

* This research was supported by Public Health Service Grants AM-39992 and AM-09650.

f Present address, Department of Biochemistry, College of Medicine, University of Arizona, Tucson, Arizona 85724. To whom all correspondence should be directed.

25-hydroxy-vitamin DP and indicated that it was 1% times as active as vitamin Da in curing rickets and acted faster than the native vitamin in promoting intestinal calcium transport and bone resorption. Since 25-01-I-D& acted directly on isolated intestine (3) and bone in tissue culture (4), and was incorrectly thought to represent the form associated with the intestinal nuclear fraction (5), DeLuca postulated that 25-OEI-D3 was the metabolically active form of vitamin Da (6-8). Hsussler et al. (9, 10) showed that a sterol more polar than 25.OII-D3 was specifically associated with the t,arget intestinal mucosa. This more polar form, denoted Peak 4B, had activity greater than vitamin D and appeared in the gut prior to the onset of vitamin D-mediated intestinal absorption of calcium. More- over, Peak 4B was found to be selectively associated with the chromatin fraction of the nucleus via a non-histone-protein re- ceptor molecule (11). Since vitamin D had been hypothesized to function through the induction RNA and protein synthesis (12, 13), this localization of Peak 413 in the nuclear chromatin was consistent with the hypothesis that Peak 4B was the form of vitamin D initiating events at the molecular level in the in- testine.

Kodicek and co-workers (14-16) confirmed the findings of Haussler et al. (9) and, in addition, discovered that [I-or-W]vi- tamin Da loses virtually all of its tritium during metabolism to Peak 4B (denoted Peak P by these workers). They also re- ported that 25-OH-D3 was an intermediate in the conversion of vitamin DB to Peak 4B (17) and suggested that Peak 4B might contain an oxy-function at carbon 1 in addition to the hydroxyl group at carbon 25.

Further indication that Peak 4B represented the active form of the vitamin in the gut was provided when Haussler et al. (18, 19) demonstrated that Peak 413 acted 3 times faster than either 25.OH-D3 or vitamin Da and was at least 5 times as active as the native vitamin in stimulating intestinal calcium transport. Positive identification of Peak 4B as 1,25-dihydroxy-vitamin DI (1,25-di-OH-D:) was accomplished by Holick et al. (20) us- ing extracts from 1450 chick guts. Lawson et al. (21) and Nor- man et al. (22) confirmed this identification employing metab- olite generated in vitro using the renal enzyme capable of cat- alyzing the formation of Peak 4B from 25-OH-D3 (23).

Although it is now apparent that 1,25-di-OH-D3 is the active form which mediates intestinal calcium absorption, its exact

1 The abbreviations used are: 25.OH-D3, 25-hydroxy-vitamin Da; 1,25-di-OH-D3, 1,25-dihydroxy-vitamin D, (also referred to as Peak 4B1, Peak P, and intestinal Peak V); 21,25-di-OH-D3, 21,25-dihydroxy-vitamin D 3 ; 25,26-di-OH-Da, 25,26-dihydroxy- vitamin DB; CCD, countercurrent distribution; lc, partition co- efficient.

2328

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1972 M. R. Haussler and H. Rasmussen

mechanism of action at the level of the intestine remains un- clear. Previous studies with vitamin Da indicated that the vi- tamin (metabolite) functioned via induction of RNA and pro- tein synthesis (12, 13). Vitamin D increases intestinal RNA synthesis (24) and the template efficiency of intestinal chroma- tin (25). Wasserman and co-workers (26-28) have isolated and extensively characterized a calcium-binding protein from intestine which is induced by vitamin D. In addition, vitamin D increases the activity of alkaline phosphatase (29) in the in- testinal brush border membrane, and this enzyme is capable of hydrolyzing ATP and pyrophosphate (30-32). The functional involvement of these induced proteins in the transport of cal- cium and the exact site and mode of action of 1,25-di-OH-D3 in this system remain to be elucidated. Direct experiments with 1,25-di-OH-Da, preferably with isolated systems, should lead to the answers to these questions.

Unfortunately, previous studies of vitamin D metabolism have centered on one specific tissue. Investigations by Haus- sler et al. (9, 10) were concerned primarily with the physiolog- ically important metabolites in the intestine. Extensive stud- ies by DeLuca and co-workers were initially limited to char- acterization of plasma metabolites and led to the identification of 25.OH-D3 (l), 21,25-dihydroxy-vitamin Dz (21,25-di-OH- D3) (33), and 25,26-dihydroxy-vitamin D3 (25,26-di-OH-D3) (34). These workers have not reported the unequivocal de- tection of the intestinally active metabolite (I, 25.di-OH-Ds) in plasma or in tissues other than intestine. Another question which has not been answered is whether 21,25-di-OH-Da and 25,26-di-OH-D3 a,re associated with tissues which carry out calcium translocation.

The purpose of t,he present report is to present a comprehen- sive study of the metabolites of vitamin DS in the tissues of the chick which play a role in calcium homeostasis. The signifi- cance of 1,25-di-OH-D3 in the intestine is confirmed and ex- tended and new evidence is provided which points to the impor- tance of 1,25-di-OH-D3 in bone resorption.

MATERIALS Ah-D METHODS

Animals-White leghorn cockerels (Moyer’s Chicks, Quaker- town, Pa.) were utilized in all experiments. Vitamin D-defi- cient chicks were raised on a diet described elsewhere (13) and were used for investigation when they became rachitic during their fourth week of development.

Specialized Chemicals-[4-14ClVitamin Ds (specific activity 32.3 mCi per mmole) and [la!,2a-3Hz]vitamin Da (specific activ- ity 577 mCi per mmole) were obtained from Amersham-Searle. [26,27-3H2]25-OH-D3 (specific activity 315 mCi per mmole) is a product of New England Nuclear, and nonradioactive 25- OH-D3 was a gift from Dr. A. W. Norman. Crystalline vita- min D3 was purchased from Calbiochem. Periodic acid (Hs IOs) was secured from Mann Research Laboratories. ~-EC- dysone was a kind gift from Dr. John Siddall. Silicic acid (Bio-Sil HA, minus 325 mesh) was purchased from Calbiochem, and Celite (Johns-Manville) was obtained from the A. H. Tho- mas Co., Philadelphia. All other chemicals and solvents used were reagent grade. Solvents employed in countercurrent dis- tribution, Sephadex LH-20 chromatography, and liquid-liquid partition chromatography on Celite were glass-distilled just prior to the experiment.

Extraction Techniques---Labeled sterols were dissolved in 0.2 ml of 1,2-propanediol and were administered either orally or

intracardially. The chicks were killed by decapitation, the blood was collected with the aid of a funnel, and the small in- testine, kidneys, and tibia were removed immediately. The small intestine was rinsed with isotonic sucrose by forcing the solution through the gut with a syringe; pairs of tibia and fibula were freed of adhering muscle and slit endwise to remove the bone marrow. Plasma was prepared from heparinized blood by centrifugation at 10,000 X g for 10 min.

Extraction of the small intestine, plasma, kidneys, or bone with acetone-dichloroethane (2:1) was carried out in a l-gallon Waring Blendor at a setting of medium for 2 min. Four vol- umes of acetone-dichloroethane per g wet weight of tissue were used in each extraction. The initial acetone-dichloroethane extract was filtered, and, following a second identical extraction of the residue, the combined extracts were taken to dryness with a flash evaporator. The lipid residue was then solubilized in a minimal volume of diethyl ether, clarified by centrifugation for 10 min at 10,000 x g at 5”, and then used for chromatograph separations.

Xilicic Acid Column Chromatography-Silicic acid was ac- tivated by heating to 120” for 24 hours just prior to use. Si- licit acid (25 g) was suspended in hexane and formed into a column (1.8 x 17.5 cm) with the aid of 7 pounds nitrogen pres- sure for packing. The sample was applied in a small volume of hexane-diethyl ether (1: 1)) and elution was carried out by the following schedule: 125 ml of hexane-ether (1: I), Fraction 1; 100 ml of diethyl ether, Fractions 2 to 5; 150 ml of diethyl ether- dichloroethane (1 :I), Fractions 6 to 11; and 125 ml of acetone, Fraction 12. Each column was run under 4 pounds of nitrogen pressure and with a flow rate of 5 ml per min.

Countercurrent Distribution-Distribution studies were carried out on a H. 0. Post, model 2B, loo-tube automated machine; 100 transfers were performed with a lo-ml mobile (upper) phase and a lo-ml stationary (lower) phase. A solvent system of ethyl acetate-hexane-ethanol-water (5:15:7.5: 12.5) was uti- lized in all CCD runs.

Sephadex LH-20 Column Chromatography-Liquid-gel parti- tion chromatography on Sephadex LH-20 was performed using the method of Holick and DeLuca (35).

Celite Column Chromatography-Celite was washed with con- centrated HCl and organic solvents, and the fine particles were removed prior to use (36). Each column contained 11 g of Celite prepared in the following fashion: 5 volumes of 10% ethyl acetate in hexane were equilibrated with 1 volume of 45% water in ethanol. The lcwer water-ethanol phase served as a sta- tionary phase and 11 ml were mixed into the 11 g of Celite. The Celite was then suspended in excess ethyl acetate-hexane (upper phase) and packed with a rod into a homogeneous column (1 X 36 cm). Samples were applied in 0.4 mi of ethyl acetate-hexane and the column was eluted with ethyl acetate-hexane as a mo- bile phase. One to two pounds nitrogen pressure were applied to achieve a flow rate of 0.4 ml per min.

Periodate Oxidation-Reactions of metabolites with periodic acid were carried out by dissolving 12.5 mg of H&IO6 in 5 ml of lower phase of the CCD solvent system (water-ethanol). The metabolite was then added and allowed to react for 15 min at room temperature. The reaction was terminated by adding the reaction mixture to position “0” of the CCD machine and im- mediately initiating the transfer procedure. Periodate remained in the initial tube while metabolites migrated along the train to their appropriate positions after 100 transfers. Preliminary

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

2330 Vitamin D Metabolites in Chick Vol. 247, No. 8

INTESTINE 4BI

(6 6 %

KIDNEY

2 4 6 8 IO 12

6

BONE

25-OH-D3 - 3 ~!62%)

) 2 4 6 8 1012

FRACTION NUMBER

FIG. 1. Silicic acid column chromatography of lipid extracts of intestine, plasma, kidney, and bone from rachitic chicks treated with 20 i.u. of [4-Wlvitamin Da 16 hours prior to killing. The chicks from which the intestines were obtained were dosed intra- cardially; all other chicks received labeled vitamin Da orally. Extraction and chromatography were carried out as described in the text. Each column represents the chromatography of pooled extracts from 20 chicks. As is detailed in the text, Peaks 4B are eluted with 100% acetone. Recovery of applied radioactivity was Sl%, 89%, 76%, and 71y0 in the intestine, plasma, kidney, and bone columns, respectively; further elution of the columns with in- creasing concentrations of methanol in acetone eluted two to three minor, more polar peaks. The more polar metabolites have been noted in other studies (9, 10, 33, 34), but they are quantitatively unimportant in the present context.

experiments with model compounds which were not expected to be altered by this treatment, such as 25.OH-D3, indicated that there was conservation of these steroids. Steroids with vicinal hydroxyl groups, such as a+ecdysone, were at least 80% cleaved under these conditions.

Radioactivity Determinations-Tritium and 14C radioactivity were determined by counting samples in a Packard Tri-Carb liquid scintillation system. Entire fractions or appropriate aliquots of the sterols were dried in the vial under a stream of air. The samples were solubilized in 10 ml each of a toluene- based mixture containing 4 g of 2,5-diphenyloxazole (PPO) and 50 mg of 1,4-bis[2-(5-phenyloxazolyl)]benzene (POPOP) per liter of toluene. Samples were counted to 2g0 error and counts per min of tritium or 14C, or both, were converted to disintegra- tions per min by the dual-isotope, internal standardization method as previously described (9).

RESULTS

In order to assess the quantitative importance of metabolites of vitamin Da after a physiological dose of the vitamin, a dose level of 20 i.u.2 was chosen. [4J4C]Vitamin Da was employed

2 One international unit of vitamin Dz is equivalent to 65 pmoles

I - : 2 s 300 1 -

t

I a v 200 -

100 - 1 0, I -I

25-OH-D;

(K=15

(K=0.9) I I I I I I I I I I I I I I I I

1’ 5

0 20 40 60 80 100

TUBE NUMBER FIG. 2. CCD of IGlabeled Peak 4Br and standard [3H]25-

OH-Dz. W-Labeled Peak 4Br was secured following silicic acid chromatography in an experiment similar to that pictured in Fig. 1 and combined with commercially obtained [26,27-3H2]25-OH-D3. The metabolites were separated via CCD as described in the text; recovery was 84y0 for 1% and 95y0 for 3H.

to eliminate misinterpretation of data because of tritium loss during metabolism of the various preparations of [3H]vitamin D3. In the experiments summarized in Fig. 1, chicks were given either oral or intracardial doses of 20 i.u. of [4-14C]vitamin Da and were killed 16 hours later. Following extraction of the tissues in question, chromatographic evaluation of the metab- olite pattern was performed with silicic acid chromatography. As previously reported (9, 10, 18, 19), the major metabolite in the intestine is Peak 4Br. In contrast to the intestine, the plasma, kidney, and bone contain primarily 25.OH-D8. The 25-OH-D3 in all four sites has been identified by chromatography with standard preparations of radioactive and nonradioactive 25.OH-Da (10, 18, 19). However, the plasma, kidney, and bone also have significant quantities of radioactivity migrating sim- ilarly to Peak 4Br. Pending further characterization by addi- tional chromatography, these metabolites are designated Peak 4B with a subscript identifying the organ from which the metab- elite(s) is obtained.

Further investigation of the Peaks 4B from the various tis- sues was accomplished by chromatography on CCD, Sephadex LH-20, and Celite. Fig. 2 illustrates distribution of Peak 4Br on CCD with added [26,27-3H2]25-OH-Da. The reference 25- OH-Da migrates far along the train with a partition coefficient (k) of 15, while the more polar Peak 4Br forms a symmetrical peak near the center of the train with a k of 0.9. The shape of the Peak 4Br pattern is virtually identical with the theoretical distribution pattern predicted for one compound with a k of 0.9.

When Peak 4Br is analyzed in the identical CCD system, a significantly different pattern results (Fig. 3, upper portion).

Peak 4Br yie-ds a diffuse and skewed peak with an average k of 1.8, indicating that Peak 4Br is probably a composite of several metabolites. Since Peak 4Bp has a larger k than Peak 4Br, it is less polar than Peak 4Br and qualitatively different as pre-

or 0.025 pg. The chick requires 5 to 10 i.u. per day as a minimum physiological dose.

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1972 M. R. Haussler and H. Rasmussen 2331

T 200- PEAK 4ep

UW;;~D Kgl.8

5 E T

400-y PEAK 40~ HT04-TREATED

FIG. 3. CCD of Peak 4Br before and after treatment with peri- odic acid. A samnle of 3H-labeled Peak 4Bp secured in an experi- ment analogous to that pictured in Fig. 1 was divided into two equal parts. One portion was run without prior treatment (upper graph) ; the other was previously treated for 15 min with periodic acid as described in the text and then distributed in the CCD machine employing the standard solvent system (lower graph). Recovery of radioactivity was 86yo in both runs.

viously established by tritium loss experiments (14). Addi- tional information as to the nature of Peak 4Bp is provided by periodate oxidation prior to CCD. As is seen in the lower half of Fig. 3, the majority of Peak 4Bp is sensitive to cleavage by periodate. The migration pattern of the reaction products provides a fortuitous opportunity to identify the predominant metabolite present in Peak 4Bp. The product, which migrates with a k of 15.5, is very similar to 25-OH-Da in its chromato- graphic behavior (see Fig. 2). The only sterol which will yield a 25 hydroxy-D3-like compound upon periodate treatment is 25,26-di-OH-D3, with the product being 25.keto, 26-normethyl- vitamin Da. Thus, at this dose level, the major dihydroxy- vitamin Ds in rachitic chick plasma is 25,26-di-OH-D%.

Peak 4Br was also examined after periodate oxidation, both to show that it differs from 25,26-di-OH-D3 of plasma and to probe the possibility that vicinal hydroxyl formation at carbon 2 and carbon 3 could explain the loss of tritium (14). Fig. 4 pictures Peak 4Br run on the standard CCD system before and after treatment with periodate. Peak 4Br is virtually insensi- tive to cleavage except for a small fraction of about 4y0 which represents 25,26-di-OH-D3. The insensitivity of Peak 4Br to periodate oxidation indicates that carbon 2 does not contain an

200 K=0.9

5 5

200- K.0.9

PEAK 481

HI04-TREATED

K =15.5

0 20 40 60 80 100 TUBE NUMBER

FIG. 4. CCD of Peak 4Br before and after treatment with periodic acid. 3H-Labeled Peak 4Br was obtained via silicic acid chromatography as shown in Fig. 1. The 3H-labeled Peak 4Br was then handled exactly as described for the experiment with Peak 4Bp (Fig. 3). Recovery of 3H applied to the CCD was 85% in each run.

oxy-function and further points to carbon 1 as the site for a second additional hydroxyl function (14). These data are con- sistent with a structure of 1,25-di-OH-D3 for Peak 4Br, and this assumption has been substantiated by the recent identifica- tion of Peak 4Br as 1,25-di-OH-D3 in three laboratories (20-22).

Additional comparison between Peak 4Br and Peak 4Bp was accomplished via liquid-gel partition chromatography on Sepha- dex LH-20. Simultaneous chromatography of 14C-labeled Peak 4Bp and 3H-labeled Peak 4Br obtained in experiments similar to the one pictured in Fig. 1 is shown in Fig. 5. Peak 4Br again migrates as a single peak, whereas Peak 4B, is resolved into at least three metabolites. 21,25-di-OH-Da and Peak Vi,, as com- ponents of plasma Peak 4B, are tentatively located by compari- son with the results of Holick and DeLuca (35). Excellent re- coveries and separations are obtained by this method with one significant ‘exception. The majority of Peak 4Bp (25,26-di- OH-Da) was not resolved from Peak 4Br (1,25-di-OH-Ds). This fact puts severe limitations on this method for the detec- tion of 1,25-di-OH-D3 in plasma samples and precludes its use for unequivocally elucidating the species of di-OH-D, present in other tissues such as bone and kidney.

Celite rechromatography of Peak 4B proved to be the most powerful tool in resolving and identifying the various dihydroxy- Da-vitamins. Fig. 6 illustrates chromatography of 14C-labeled Peak 4Br and 3H-labeled Peak 4Bp together on the same col-

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

2332 Vitamin D Metabolites in Chick Vol. 247, No. 8

/ -

-

I 0

- 14C-4Bp --- “H-4Br 461

I&\ : : I I

\ : 1

1 : 1

FRACTION NUMBER (IDI’d)

FIG. 5. Liquid-gel partition chromatography of 3H-labeled Peak 4Br and 14C-labeled Peak 4Bp. Peaks 4Br and 4Bp were ob- tained by silicic acid column chromatography and then simul- taneously chromatographed on Sephadex LH-20 according to the procedure of Holick and DeLuca (35). Recovery of 1% and tritium applied to the column was 95% and 92oj,, respectively. The small init,ial peak emerging in Fraction 7 represents slight contamination of both Peak 4Br and Peak 4Bp with 25-OH-D3.

- INTEST.‘4C-4B

T - -- PLASMA 3H-40

25,26Di-OH-D3 T 72000

I II I,25 Di-OH-D3 -7 0 I

:: n I L I I I I , I I I 5 F ::

000 L . I

m

I 0

FRACTION NUMBER (5ml)

FIG. 6. Liquid-liquid partition chromatography of W-labeled Peak 4Br and 3H-labeled Peak 4Bp. Peaks 4Br and 4Bp were harvested from silicic acid columns (Fig. 1) and then simultane- ously chromatographed on Celite as described in the text. Por- tions of the 3H-labeled Peak 4Bp were saved for use as markers in the Celite columns illustrated in Figs. 7 and 8. The initial peak emerging in 1 hold-up volume (Fractions 3 and 4) is 25-OH-Da contamination of the Peak ~BJ and Peak 4Bp. Recoveries of radioactivity were 74y0 for 3H and 72yo for r4C.

umn. Celite liquid-liquid partition clearly shows that Peak 4Br is a single metabolite (1,25-di-OH-Ds), whereas Peak 4Bp is resolved into its three primary dihydroxy components: 21,25- di-OH-Ds, 25,26-di-OH-Da, and 1,25-di-OH-Da. The minor metabolite seen in Fig. 5, Peak Vi,, apparently migrates with 21,25-di-OH-D3 in this system. Rechromatography of this

combined peak in a Celite system specifically designed for 21,25- di-OH-D3 (37) tends to support this conclusion,3 but positive

3 M. R. Haussler, unpublished results.

T - KIDNEY 14C-4B

---- PLASMA 3H-4 6 T 25,26DI-OH-D3

1 ‘2000 ‘;

.‘I I I .L

I I 4 I

5 400- !

:

; 1,25 Di OH-D,

F :

s: :

ii! I

;

:

2 c 200 - 5

2

0’ 5 IO 20 25 30

FRACTION NUMBER (5ml)

FIG. 7. Liquid-liquid partition chromatography of W-labeled Peak 4Bx and 3H-labeled Peak 4Bp on Celite. Recoveries of radioactivity were 70% for 3H and 65% for r4C.

identification of 21, 25-di-OH-D3 awaits further experimenta- tion. 25,26-di-OH-D3 is identified by the fact that it is the major portion of 4Bp and it is periodate-sensitive. Of great importance is the detection of 1,25-di-OH-D3 as a minor com- ponent of Peak 4B,. Plasma 1,25-di-OH-D3 is identified via

its exact migration with [3H]1,25-di-OH-D3 from the intestine. The presence of 1,25-di-OH-D3 in plasma is expected because it must traverse from its endocrine site of production, the kidney, to its ultimate target tissue, the gut. However, its detection in measurable quantities provides evidence that a diagnostic test can be developed for this metabolite in blood.

Fig. 7 pictures the chromatography of ‘%-labeled peak 4BK along with added 3H-labeled Peak 4Bp to serve as a marker. Clearly the majority of kidney Peak 4B exists as 1,25-di-OH-D3; a small amount of 25,26-di-OH-D3 is also found in the kidney. Since 1,25-di-OH-D3 is postulated to be produced exclusively by a renal enzyme system (23), this result supports that concept by showing that 1,25-di-OH-D3 is present in the kidney in viva.

Because the skeleton is considered a second target organ for vitamin D, it was of interest to examine the nature of the di- hydroxy-Dp-vitamins present at this site. Fig. 8 illustrates data which indicate that virtually all of the bone Peak 4B con- sists also of 1, 25-di-OH-D3. Very small amounts of 25,26-di- OH-D3 and 21 ,25-di-OH-D3 are present in the bone. The finding of significant quantities of 1,25-di-OH-D3 in bone raises the question as to whether the intestinally active steroid might also represent the active form of vitamin Da at the level of the skeleton. Biological assays on 1, 25-di-OH-D3 in terms of bone dissolution were carried out in order to answer this question. As can be seen in Table I, 1,25-di-OH-Da has considerable activity in promoting bone resorption. 1,25-di-OH-D3, isolated from the guts of 400 rachitic chicks which had previously received 50 i.u. of [3H]vitamin Ds, was purified by solvent partitions as pre- viously described (19) and then via silicic acid and Celite chro- matography as in Figs. 1 and 6. The purified 1,25-di-OH-D3 was tested initially at a level of 65 pmoles with chicks main- tained on a standard rachitogenic diet (Table I). Only 10 hours after administration, this low dose produced a significant rise in plasma Ca++. However, since the chicks were main- tained on a normal diet with respect to calcium, it is not possible

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1972 M. R. Haussler and H. Rasmussen 2333

-BONE “C-48

- -- PLASMA 3H-4B

25,2601-OH-O3 T -2000 :I -i- ,I

: :

: : I

: I

1,25Oi-OH-D1

TABLE II Absolute amount of vitamin D3 and metabolites in intestine, plasma,

kidneys, and bone 16 hours after 20 i.u. (1300 pmoles) of vitamin D3

Data is taken from Fig. 1, and 6 to 8 and converted to picomoles of metabolite per one chick tissue. 1,25-di-OH-D3 metabolite is corrected for a reduction of 46% in specific activity when calcu- lating pmoles from [3H]l,25-di-OH-D3.

pm&s Stew1 per chick tissue I

: I ,

21.25 Di-OH-D3: t - Vitamin

D3 21 , 25.di- OH-D, +Vb

25.OH-D3 1 , 25-di- OH-Da

‘2,5,5,~&$- 3

2.74 3.10 9.00 0.32 0.01 6.62 40.90 1.76 3.21 1.32 1.32 2.11 1.18 0.26 0.07 6.81 23.30 4.75 1.23 0.14

Intestine ...... Plasmaa. ...... Kidney ...... Bone*. ........ 5 IO I5 20 25 30

FRACTION NUMBER (5ml) - n Calculated assuming a 50% yield of blood via killing by de-

capitation and collection. * Computed from data on pairs of tibia and fibula and assuming

the skeleton represents 15yo of the body weight.

FIG. 8. Liquid-liquid partition chromatography of W-labeled Peak 4Br, and 3H-labeled Peak 4Bp on Celite. Recoveries of radio- activity were 73y0 for 3H and 68% for 1%.

TABLE I

Biological activity of 1,15-&-OH-D3 in promoting bone resorption in chicks

Animals employed in this experiment were grown for 3 weeks on the standard rachitogenic diet and then either used or trans- ferred for 3 days to a similar diet containing <O.l% calcium. Chicks transferred to the low calcium diet showed a further drop in plasma calcium concentration below the rachitic value. In- crease in plasma calcium produced in these chicks grown on the low calcium diet is regarded as a true measure of sterol-induced bone resorption (2, 38). Vitamin Da and metabolites were ad- ministered orally in 0.2 ml of 1,2-propanediol, and plasma calcium was determined 24 hours later with a Technicon Auto-Analyzer (39). Each number represents the average of five animals f S.E.M.

than 25-OH-D3 in skeletal resorption, these data suggest that 1,25-di-OH-Da plays a metabolic role in calcium translocation at the level of bone.

An over-all summary of the amounts of the various physiolog- ical metabolites of vitamin Da present in intestine, plasma, kid- ney, and bone is provided in Table II. 1,25-di-OH-D3 dom- inates the intestinal pattern and is concentrated in this organ in relation to other sites. The plasma carries primarily 25-OH- Da, but considerable amounts of all other metabolites are pres- ent. The plasma is unique in that it is the only location where 21,25-di-OH-Da and 25,26-di-OH-Da are found in appreciable amounts. The kidney contains primarily 25-OH-D3, presum- ably as a precursor for the production of 1, 25-di-OH-D3 at, this site. It is not known whether 25-OH-D3 or 1,25:di-OH-D3, or both, have physiological effects on the calcium-transporting system in the kidney itself. The bone metabolite pattern is characterized by a predominance of 25-OH-D3 and substantial amounts of 1,25-di-OH-D3. It is possible that both metabo- lites regulate bone function.

DISCUSSION

The present communication reports the combined use of highly refined chromatographic systems for the separation of vitamin Da metabolites and physiological doses of the vitamin to examine the metabolism of vitamin D in the rachitic chick. Special attention is directed toward vitamin D target organs as well as other sites of calcium regulation. It is concluded that there are at least four biologically significant metabolites of vita- min DB. Fig. 9 details the sequence of formation of these me- tabolites from the parent vitamin. The initial step involves hydroxylation at carbon 25. This reaction is reported to be catalyzed primarily by a liver enzyme (41), but we have ob- tained recent evidence4 that the kidney is another site of 25- OH-D3 formation from vitamin Da. The kidney is therefore capable of converting native vitamin Da completely to 1,25-di- OH-D,. Although 25-OH-D3 was previously thought to be the active form of the vitamin, it is now generally accepted that this

4 M. R. Haussler, manuscript in preparation.

:ek%tive ctivityO of me- abolite

E a

t .-

Chick diet Steroid treatment Plasma ca++

mg %

5.8 zk 0.5 9.5 f 0.3 7.2 f 0.4

3.8 f 0.4 4.3 f 0.6 5.8 f 0.4 7.4 f 0.7

Rachitogenic

Low calcium

3250 pmoles of vitamin D3 65 pmoles of 1,25-di-OH-

D2

325 pmoles of vitamin Ds 325 pmoles of 25-OH-D, 325 pmoles of 1,25-di-OH-

Dt

1.0 4.0 7.2

L

= Estimated from increment of plasma calcium increase, with that produced by vitamin Da set at 1.0.

* Assay carried out 10 hours after sterol in this case only.

to attribute the rise in plasma Ca+f solely to bone resorption. In a second experiment (Table I), a low calcium diet was utilized and 1,25-di-OH-D3 proved to be more active than vitamin Da in promoting bone resorption. In this single experiment, 1,25- di-OH-D3 is estimated to be almost twice as active as 25-OH-D3 in mediating skeletal resorption. Coupled with a recent report by Tanaka and DeLuca (40), that 1 ,25-di-OH-D3 acts faster

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

2334 Vitamin D Metabolites in Chick Vol. 247, No. 8

OH IA

HO’

ENZYME

OH

1,25-DI-OH-D3 (PEAK 4BI)

&OH r ‘/ HdV 25.26-Di-OH-D,

FIG. 9. Metabolism of physiological amounts of vitamin Da in _ _ the chick. The stereochemical orientation of the hydroxyl group on carbon 1 of 1,25-di-OH-D3 has not been firmly established, but it is most likely 01.

metabolite represents an intermediate in the synthesis of the final functional sterol, 1, 25.di-OH-D3. Production of 1,25di- OH-Da occurs exclusively in the kidney (23), qualifying the kidney as an endocrine organ which produces the sterol hormone that controls calcium transport.

The site(s) of production of the 21,25- and 25,26-di-OH-D3 is not known, and these metabolites are found chiefly in the plasma. The present report documents the fact that neither form is present in significant amounts in any of the tissues involved in calcium transport (Figs. 6 to 8). Yet they do appear in plasma even after low physiological doses of vitamin D3, indicating that they probably play some part in the over-all metabolic mecha- nism of vitamin D action. Their submaximal biological ac- tivities (33, 34) suggest that they may be involved in metabolic elimination and control of 25.OH-D3 function rather than having direct physiological actions in a target tissue. The minor plasma metabolite, Peak Vs, has not been characterized, and therefore its role and position in the metabolic scheme are unknown.

Other metabolites more polar than 25-OH-D3 have been re- ported (33-35, 42, 43), but these forms appear when the dose of vitamin DB is increased above the minimum physiological re- quirement of the animal. Accumulation of esters of vitamin D) in the liver also occurs when larger amounts of vitamin D are administered (44). Suda et al. (34) report that their plasma Peak V could contain as many as seven separate metabolites. However, they employed both a larger dose of vitamin D and a gradient of methanol to elute Peak V from silicic acid; this pro- cedure yields several of the more polar forms which are not eluted with acetone (Fig. 1). Thus, the increased heterogeneity of their Peak V compared to our Peak 4B, is not surprising. Further metabolites of the dihydrosy-Ds-vitamins will no doubt be discovered when radioactive 1,25-di-OH-D3 becomes readily available.

Previous investigations of the metabolism of vitamin D have been hampered by incomplete separation of metabolites, lack of evaluation by rechromatography, and use of varied systems in which the patterns could not be compared. Recently, the use of

Sephadex LH-20 columns (35) has been employed as an effective means of separating vitamin D3, 25.OH-D3, and several of the more polar metabolites. But as has been pointed out in the present study (Fig. 5), resolution of 25,26-di-OH-D3 from 1,25- di-OH-D3 cannot be achieved with this system. Combination of this technique with Celite chromatography or with periodate treatment (to assess the amount of 25,26-di-OH-Da) would be required to scrutinize a mixture of vitamin Da and its known metabolites. Based on our initial studies with countercurrent distribution, we have devised a liquid-liquid partition separation procedure on columns of Celite which is capable of separating all of the heretofore identified dihydroxy-Ds-vitamins. A sol- vent system similar to that employed for CCD (Figs. 2 and 3) is utilized, with the Celite column increasing strikingly the num- ber of “theoretical plates.” Once the routine homogeneous packing of the columns is mastered, this technique becomes per- fectly reproducible. It has an advantage over previous Celite systems (9, 33) in that the metabolites emerge (earlier) within 2 to 6 hold-up volumes without considerable loss of resolving power. Combined with a preliminary separation of vitamin DD, 25-OH-D3, and the dihydroxy-Do-fraction on silicic acid or Sephadex LH-20, this method permits the complete chromato- graphic evaluation of a given metabolite mixture.

We have substantiated our identification of the metabolites in the Celite chromatography system (Figs. 6 to 8) by additional chromatography not detailed under “Results.” The initial peak emerging in Fractions 3 and 4 on Celite migrates identically with 25.OH-Da in three chromatography systems (1, 18, 35). The second peak is resolved into 21,25-di-OH-D3 and Peak VP, either by chromatography on Sephadex LH-20 as in Fig. 5 or on a Celite system devised for 21,25-di-OH-D3 (37). The iso- lated 25,26-di-OH-D3 peak is the only peak susceptible to perio- date oxidation prior to rechromatography on either Sephades LH-20 or Celite. Furthermore when the 4Ep fraction from an experiment such as that pictured in Fig. 5 is isolated and chro- matographed in Celite, it is resolved into 26,25-di-OH-D3 and 1,25-di-OH-D3 as expected. The cross-checks provided by this additional chromatography strengthen the assignment of struc- tures to the various peaks.

The detection of 1, 25-di-OH-D3 in plasma (Fig. 6) has several important implications. The metabolite is presumably in transit from the kidney to the target intestine and possibly to the bone. The fact that 1,25-di-OH-D3 circulates via the blood argues that it is both the active form and the circulating active form. The later designation has been put on 25-OH-D3 by DeLuca and co-workers (20, 42) primarily because 25-OH-D3 is quantitatively more dominant in the plasma. In addition, the detection of significant amounts of 1,25-di-OH-D3 in plasma sug- gests that a competitive binding assay or radioimmunoassay can be devised for this metabolite in blood samples. Such an assay would be of great utility in diagnosing defects in vitamin D metabolism in patients with clinical disorders in calcium homeo- stasis.

The proposal by Fraser and Kodicek (23) that 1,25-di-OH-D3 is synthesized in the kidney is supported by our finding of a high renal concentration of this metabolite in viva (Fig. 7). This concept explains the observation that patients with chronic renal failure and those on hemodialysis show decreased absorp- tion of calcium from the gut and suffer from secondary hyper- parathyroidism and bone demineralization (45-47). It is likely that 1, 25.di-OH-D3 will become a valuable supplement to main-

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1972 M. R. Haussler and H. Rasmussen 2335

tain a normal calcium balance in such cases where the metabolic formation of 1, 25-di-OH-D3 is disrupted by kidney disease and uremia.

Detection of 1,25-di-OH-Da in bone implies that the intesti- nally active metabolite may also function at the skeleton. The localization of 1,25-di-OH-D3 in bone is not as striking as is seen in the intestine (Table II), but sufficient quantities are present to warrant a search for specific receptors as has been done in the intestine (9, 11). Preliminary data indicate that specific receptor systems may exist in bone for both 25.OH-D3 and 1,25-di-OH-D3 (48) ; this perhaps represents one receptor with a low level of specificity (i.e. binds both forms). Further binding studies are required before a conclusion can be reached. The other approach which led to acceptance of 1,25-di-OH-D3 as the active form in the intestine is the assessment of the rela- tive biological potency and kinetics of action of the metabolites in question. As can be seen in Table I, 1, 25.di-OH-D3 has con- siderable activity in promoting bone resorption and is more ac- tive than 25.OH-D3 in this single point assay. Recent findings of Tanaka and DeLuca (40) indicate that 1 ,25-di-OH-D3 acts more rapidly than 25-OH-D3, but both metabolites have ap- proximately equal potency. Additional experiments with iso- lated bone in culture (4) are necessary to elucidate the active metabolite in the skeleton. Perhaps 25.OH-D3 is able to func- tion as a more efficient substitute for 1, 25-di-OH-D3 in the bone system than it is in the gut. Thus, a mass-action effect of 25- OH-Da to simulate 1 ,25-di-OH-D3 may explain its direct ac- tivity in bone (4) and in the isolated intestine (3).

A general picture of the metabolism of vitamin D now exists, but a complete understanding of this vitamin awaits the elucida- tion of numerous details related to the production and function of the various metabolite forms. Little information is available concerning the enzymes which catalyze metabolic alterations of the native vitamin. The discovery of physiological and pharma- cological factors which control these enzyme reactions could be of great importance to our over-all knowledge of calcium me- tabolism. Finally, the biochemical mechanisms of action of the hormonal form(s) of vitamin D remain as poorly understood phenomena and will be the topic of considerable future research.

Acknowledgments-We would like to thank Douglas Boyce, Philip Esocoff, Ulder Tillman, and Gabriel Tucker, III for their expert technical assistance during various phases of this study.

1.

2.

3. 4.

5. 6. 7. 8. 9.

10.

11.

REFERENCES

BLUNT, J. W., DELUCA, H. F., AND SCHNOES, H. K. (1968) Biochemistry 7, 3317

BLUNT, J. W., TANAICA, Y., END DELUCA, H. F. (1968) Proc. Nat. Acad. Sci. U. S. A. 61, 1503

OLSON, E. B., END DELUCA, H. F. (1969) Science 166,405407 TRUMMEL, C. L., RAISZ, L. G., BLUNT, J. W., AND DELuc~,

H. F. (1969) Science 163, 1450-1451 DELUCA, H. F. (1969) Amer. J. Clin. Nub. 22, 412 DELUCA, H. F. (1969) Arch. Zntern. Med. 124, 442 DELUCA, H. F. (1969) Fed. Proc. 28, 1678 DELUCA, H. F. (1969) Cakijied Tissue Res. 4, 283 HAUSSLER, M. R., MYRTLE, J. F., AND NORMAN, A. W. (1968)

J. Biol. Chem. 243, 4055-4064 MYRTLE, J. F., HAUSSLER, M. R., AND NORMAN, A. W. (1970)

J. Biol. Chem. 246, 1190-1196 HAUSSLER, M. R., AND NORMAN, A. W. (1969) Proc. Nut.

Acad. Sci. Il. S. A. 62, 155

12. ZULL, J. E., CZ.~RNOWSKA-MISZT~L, E., AND DELUCA, H. F. (1965) Science 149. 182-184

13. NORMAN, A. W. (lQk6) Amer. J. Physiol. 211, 829 14. LAWSON, D. E. M., WILSON, P. W., AND KODICEK, E. (1969)

Nature 222, 171-172 15. LAWSON, D. E. M., WILSON, P. W., BARICER, D. C., AND

KODICEK, E. (1969) Biochem. J. 116, 263 16. KODICEK, E., LAWSON, D. E. M., AND WILSON, P. W. (1970)

Nature 226, 763-764 17.

18.

19.

20.

21.

22.

23. 24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45. 46.

47.

48.

LAWSON, D. E. M., WILSON, P. W., AND KODICEK, E. (1969) Biochem. J. 115, 269

HAUSSLER, M. R., LITTLEDIKE, E. T., BOYCE, D. W., AND RASMUSSEN, H. (1970) Fed. Proc. 29, 368

HAUSSLER, M. R., BOYCE, D. W., LITTLEDIICE, E. T., AND RASMUSSEN, H. (1971) Proc. fIrat. Acad. Sci. U. S. A. 68. 177

HOLICK, M. F., SCHNO&, H. K., AND DELUCA, H. F. (1971) Proc. Nat. Acad. Sci. U. S. A. 68.803

LAWSON, D. E. M., FRASER, D. RI, KODICEK, E., MORRIS, H. R., AND WILLIAMS, D. H. (1971) Nature 230, 228-230

NORMAN, A. W., MYRTLE, J. F., MIDGETT, It. J., NOWICKI, H. G., WILLIAMS, V., AND POP&K, G. (1971) Science 173, 51-54

FRASER, D. R., AND KODICEK, E. (1970) Nature 228, 764-766 STOHS, S. J., ZULL, J. E., AND DELUCA, H. F. (1967) 8io-

chemistry 6, 1304 HALLICK, 12. B., AND DELUCA, H. F. (1969) Proc. Nut. Acad.

Sci. U. S. A. 63, 528 WASSERMAN, R. H., AND TAYLOR, A. N. (1966) Science 162,

791-793 WASSERMAN, R. H., AND TAYLOR, A. N. (1968) J. Biol. Chem.

243, 3987-3993 WASSERMAN, R. H., CORRADINO, R. A., AND TAYLOR, A. N.

(1968) J. Biol. Chem. 243, 3978-3986 HAUSSLER, M. R., AND NAGODE, L. A. (1969) J. Cell Biol. 43,

51a MARTIN, D. L., MELANCON, M. J., AND DELucA, H. F. (1969)

Biochem. Biophys. Res. Commun. 36, 819 NAGODE, L. A.,.HAUSSLER, M. R., BOYCE, D. W., PECHET, M.,

AND RASMUSSEN, H. (1970) Fed. Proc. 29. 368 HAUSSLER, M. R.,‘NAG~DE,‘L. A., AND RASMUSSEN, H. (1970)

Nature 228, 1199 SUDA, T.: DELUCA, H. F., SCHNOES, H. K., PONCHON, G.,

TANAKA, Y., AND HOLICK, M. F. (1970) Biochemistry 9, 2917-2922

SUDA, T., DELUCA, H. F., SCHNOES, H. K., TANAKA, Y., AND HOLICK, M. F. (1970) Biochemistry 9, 4776-4780

HOLICK, M. F., AND DELUCA, H. F. (1971) J. Lipid Res. 12, 460

ENGEL, L. L., CAMERON, C. B., STOFFYN, A., ALEXANDER, J. A., KLEIN, O., AND TROFINMOW, N. D. (1961) Anal. Bio< them. 2, 114

BOYLE, I. T., GRAY, R. W., AND DELUCA, H. F. (1971) Proc. Nat. Acad. Sci. U. S. A. 68, 2131

HIBBERD, K. A., AND NORMAN, A. W. (1969) Biochem. Pharma- col. 18, 2347

FEINBLATT, J., BELANGER, L. F., AND RASMUSSEN, H. (1970) Amer. J. Physiol. 218, 1624

TANAKA, Y., AND DELUCA, H. F. (1971) Arch. Biochem. Bio- phys. 146, 574

HORSTING, M., AND DELUCA, H. F. (1969) Biochem. Biophys. Res. Commun. 36, 251

OMDAHL, J., HOLICK, M., SUDA, T., TANAKA, Y., AND DELUCA, H. F. (1971) Biochemistry 10, 2935

COUSINS, R. J., DELUCA, H. F., SUDA, T., CHEN, T., AND TANAKA, Y. (1970) Biochemistry 9, 1453-1459

LUND, J., DELUCA, H. F., AND HORSTING, M. (1967) Arch. Biochem. Biophys. 120, 513

MERRILL, J. P. (1970) N. Engl. J. Med. 282,953 GREGORY, D. H., AND MESSNER, R. P. (1969) J. Lab. C&n.

Med. 74, 464 KATZ, A. I., HAMPERS, C. L., AND MERRILL, J. P. (1969)

Medicine 43, 333 HAUSSLER, M. R. (1968) Doctoral Thesis. TJniversitv of Cali-

fornia, Riverside

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Mark R. Haussler and Howard Rasmussen in the Chick3The Metabolism of Vitamin D

1972, 247:2328-2335.J. Biol. Chem. 

  http://www.jbc.org/content/247/8/2328Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/247/8/2328.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on June 9, 2020http://w

ww

.jbc.org/D

ownloaded from