THEJOURNAL OF BIOLOGICAL CHEMISTRY Vol. 256, No. 20, Issue of October 25, pp. 10430-10434, 1981 Printed in U S. A .

Uptake and 25-Hydroxylationof Vitamin D, by Isolated Rat Liver Cells* (Received for publication, March 24, 1981, and in revised form, June 4, 1981)

Svein Dueland$, Inger Holmbergfj, Trond Berg, and Jan I. Pedersenl From the Institutefor Nutrition Research, School of Medicine, University of Oslo, Oslo 3, Norway, and the §Department of Clinical Chemistry and the Research Centre a t Huddinge Hospital, KarolinskaInstitutet, Stockholm, Sweden

MATERIALS AND METHODS The physiological roles played by hepatocytes and nonparenchymal cells of rat liver in the metabolism of Chemicals-Vitamin D, (cholecalciferol) and vitamin D2 (ergocalvitamin D3 have beeninvestigated. ciferol) obtained from Sigma Chemical Co., 25-hydroxyvitamin D, Tritium-labeled vitamin D3dissolved in ethanol was from Philips-Duphar B.V., Veenendaal, The Netherlands, [la,2a(n)administered intravenouslyto two rats. Isolation of the 3H]vitamin DS (12.3 Ci/mmol) and [23,24(n)-’H]25-hydroxyvitamin liver cells 30 and 70 min after the injection showed thatDS (94 Ci/mmol) from the Radiochemical Centre, Amersham, England, were purified by HPLC’ (see below) before use. 25-Hydroxy-[26vitamin D3 had been taken up both by the hepatocytes and by the nonparenchymal liver cells. The relative 2H:j]vitaminDSwas synthesized as described (6). [25,26,26,26,27,27,27DJ was a generous gift from Dr. Bjorkhem andDr. proportion of vitaminD3 that accumulated in the non- ‘H,]Vitarnin Larsson (Department of Clinical Chemistry, Huddinge Hospital, Hudparenchymal cells increased with time. Perfusion of the dinge, Sweden). Collagenase (type I) and pronase (grade B) were isolated rat liver with [‘HI vitamin D3 addedtothe purchased from Sigma and Calbiochem, respectively. All solvents perfusate confirmed the ability of both cell types to were analytical or HPLC grade. Other chemicals were high purity efficiently takeup vitamin D3 from the circulation. commercial materials. By a method based on high pressure liquid chroma- Animals-Male Wistar rats were fed an ordinary pellet diet and water ad libitum. The weight of the animals when used in the tography and isotope dilution-mass fragmentography it was found that isolated livercells in suspension had experiments was 150-200 g. Some rats were given a rachitogenic diet a considerable capacity to take up vitamin D3 from the after weaning ( 7 ) for at least 3 weeks before sacrifice. Rickets was by enlarged epiphyseal plates and by significantly lowered medium. About 2.5 fmol of vitamin DSwere foundto be Confirmed serum phosphorus concentrations. associated with each hepatocyte or nonparenchymal Preparation of Liver Cells-Isolated liver cells were obtained by cell after 1 h of incubation. 25-Hydroxylation in vitro a modification (8) of the method of Berry and Friend (9). In short, was found to be carried out only by the hepatocytes. the liver was removed from the animal and perfused in vitro, first The rate of hydroxylation was about the same whether with a calcium-free buffer for 5-10 min followed with a buffer conthe cells were isolated from normal or rachitic rats (3.5 taining 4 mM CaClz and 0.05% (w/v) collagenase. The perfusion rate and 4 pmol of 25-hydroxyvitamin D3 formed per h per was about 30 ml/min, andthe cells were separated after about10 min. lo6 cells, respectively).The possibility that the nonpar-The hepatocytes were separated from the nonparenchymal cells by enchymal cells might serve as a storagesite for vitamin differential centrifugation (10) in a medium containing 1% bovine serum albumin (fatty acid free). Bovine serum albumin was omitted, Da in the liveris discussed. however, from the medium used for preparing cells from rachitic rats.

(The reason for this omission was that in preliminary experiments with hepatocytes from these animals itwas noticed that bovine serum albumin to some extentinhibited the 25-hydroxylase activity.) In According to current concepts (1) vitamin Ds must first some of the experiments where the uptake of vitamin DBin liver cells undergo 25-hydroxylation in the liver to 25-hydroxyvitamin was studied in vivo, apure nonparenchymal cell preparation was D3, followed by la-hydroxylation in the kidneys to lar,25- obtained according to a published procedure (11) by incubating the dihydroxyvitamin D3 before it can exertits physiological cell suspension with 0.25% pronase at 37 “C for 60 min, which selecfunctions. Since theliver i s the only organ in which formation tively destroys the hepatocytes. The isolated cells were finally resusof 25-hydroxyvitamin Ds takes place to any extent (2, 3) this pended in the incubation buffer at a concentration of approximately 5 X lo6cells/ml. No cross-contamination was observed in the two cell organ plays a central role in the metabolism of the vitamin. populationsobtained from the livers of normal rats. In the cell Recently it has become evident that both the hepatocytes and preparations from the rachitic rats the mean contamination of the the nonparenchymal cells of the liver are involvedin the parenchymal cell preparations with nonparenchymal cells was 4% and metabolism of lipids and lipoproteins (4, 5 ) . In which of these that of the nonparenchymal cell preparation with parenchymal cells types of liver cells the hydroxylation of vitamin D3 takes place 2.5%. The viability of the isolated cells was estimated by the trypan is not known. In the present work we demonstrate that both blue (0.04%) exclusion test (12). The viability of the cells obtained from normal rats was 96% for 5 preparations. The mean number of t.he hepatocytes and the nonparenchymal cells of the ratliver cells from rachitic rats that excluded the dye was only 6276, with a have a high capacity to take up vitaminDn both in vivo and range of 52-78% (7 preparations). The lower viability of these cells in vitro. The further metabolism to 25-hydroxyvitamin Ds, could be explained both by the pathologic state of the animals and by the omission of bovine serum albumin from the medium during the however, takes place almost exclusively in the hepatocytes. preparation of the cells. Incubation, Extraction, a n d Chromatographic Procedures-The * This work wassupported by the Norwegian Research Council for Science and the Humanities. The costs of publication of this article isolated cells suspended in 2 ml of buffer were incubated in 50-ml were defrayed in part by the payment of page charges. This article Erlenmeyer flasks on a water bath at 37 “C under slow shaking (75 musttherefore be hereby marked “advertisement” in accordance oscillations/min). The incubation buffer contained 0.146 M NaCl, 5.4 mM KCI,0.8 mM MgS04, 2 mM CaCl?, 0.7 mM Na2HP04/KH2P04 with 18 U.S.C. Section 1734 solely to indicate this fact. 4 Present address, Department of Pharmacology, Institute of Phar’ The abbreviation used is: HPLC, high pressure liquid chromatogmacy, University of Oslo, Oslo 3, Norway. raphy. fi To whom all correspondence should be addressed.

10430

10431

Metabolism of Vitamin D3 by Isolated Rat Liver Cells buffer, pH 7.5,20 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid buffer, pH 7.5, and 10 mg/ml of bovine serum albumin. The osmolality was approximately 300 mosmol/liter. Cells from rachitic animals were incubated in the same medium except that it contained cytosolic fraction (4 mgof protein/ml)instead of bovine serum albumin. The cytosolic fraction had been prepared from a normal rat liver homogenate by centrifugation at 100,OOO X g for 90 min. The protein concentration of the resulting supernatant was 40 mg/ml. To 2 ml of the suspension of cells from rachitic ratsthe following additions were also made: 5.4 pmol of ATP, 6.4 pmol of malate, 1.3 pmol of NADP, 1.1 prnol of glucose &phosphate, 0.5 unit of glucose6-phosphate dehydrogenase, 20 pmol of MgCL, 1.1pmol of fructose, and 7.7 nmol of N,N'-diphenyl-p-phenylenediamine. (The stimulatory effect observed by these cofactors can be explained by activation of the 25-hydroxylase activity also in nonviable cells. No such stimulation was observed with cultured rat hepatocytes.) The reaction was started by the addition of 171 nmol of vitamin DB(cells from normal rats) or 520 nmol (cells from rachitic rats) in 20 pl or less of ethanol. A boiled cell suspension of identical composition was used as a blank incubation. The reaction was stopped after 2 h of incubation by the addition of 7 ml of methanol/chloroform (2:l). Deuterium-labeled 25hydroxyvitamin Da 100 ng, was added at thispoint. After extraction (7) the chloroform phase was fdtered through a0.5-pm Millipore filter FHLP 02500,was evaporated under nitrogen, and the residue was redissolved in the solvent used for HPLC. The formed 25-hydroxyvitamin D3 was separated on two consecutive HPLC systems (7). The entire sample was first chromatographed on a Zorbax-ODS column (4.6 x 250 mm, particle size 5 pm) with 7.5% H 2 0 in methanol as eluting solvent (1 ml/min). The fraction that corresponded to eluted 25-hydroxyvitamin Da (retention time, 11 min) was collected and rechromatographed on a Zorbax-Sil column (2.1 X 250 mm, particle size 5 p n ) with 2.5% isopropanol in hexane as solvent (0.8 ml/rnin). The eluted 25-hydroxyvitamin DB(retention time, 8.5 min) was converted intothe trimethylsilyl/t-butyldimethylsilylderivative, and the amount was determined by gas chromatography-mass spectrometry as described (13). In some experiments the formed 25-hydroxyvitamin DBwas quantitated by HPLC. In this case a Spherisorb-silica column (3 X 250 mm, particle size 5 pm) was used at thesecond chromatographic step with 2.5% isopropanol in hexane as eluting solvent (0.8 ml/min). The height of the UV-absorbing peak corresponding to 25-hydroxyvitamin 48.5 D, (retention time, 9.5 min) was compared to that of a standard. 49.5 Recovery was corrected for by the addition of tritium labeled 25hydroxyvitamin DB (l0,OOO cpm) prior to extraction. When the two methods of quantitation were compared identical results were obtained. All tritium-containing samples were counted at about 50% efficiency inaPackardTri-Carb liquid scintillation spectrometer. A standard amount of [3H]25-hydroxyvitaminDSwas routinely used to assure constant counting efficiency. A method based on isotope dilution-mass fragmentography (14) was used to quantitate theamount of vitamin DBthat hadbeen taken up by the cells during the incubation. In these experiments the cells were separated from the medium by centrifugation immediately after the incubations. The cell fraction (after being washed 3 times) and the medium were extracted separately. [25,26,26,26,27,27,27-'H7]vitamin DB (170 ng) was added to the cell fraction prior to extraction and purification by HPLC. The fraction corresponding to vitamin Da that eluted from the Zorbax-ODS column (retention time, 57 min) was collected and converted tothe trimethylsilyl derivative and analyzed by gas chromatography-mass spectrometry essentially as described (14). One channel of the multiple ion detector was focused at the ion at m/e 325 corresponding to the derivative of vitamin DS and the otherchannel at m/e 332 corresponding to the derivative of 'H,-labeled vitamin DS. The amount of unlabeled vitamin DS was calculated from the ratio between the height of the peak at m/e 325 and the height of the peak at m/e 332 with use of a standard curve. The standard curve was obtained by analysis of different standard mixtures of vitamin DS(0-50 ng) together with a fixed amount of 'H7labeled vitamin D3.* Uptake of Tritium-labeled Vitamin D3 into Liver Cells in Vivo and duringPerfusion of the Liver-Tritium-labeled vitamin D3 (approximately 0.75 X lo6 cpm, 50 pmol) dissolved in 100 p1 of ethanol was givenin one of the tail veins of rachitic rats. After varying periods of time the livers were perfused and the cells isolated and separated

I. Bjorkhem and A. Larsson, unpublished results.

as described above. In one experiment, tritium-labeled vitamin U:s was added to the perfusate (approximately lo6 cpm, 70 pmol), and the liver was perfused for 60 min. Prior to extraction of the cell preparations a known amount (310 ng) of vitamin Dz was added to correct for recovery during extraction and chromatography. The samples were extracted with chloroform/ methanol, 2:1, as above, and the upper layer was re-extracted once with chloroform. The extract was taken to dryness under a stream of Nz and redissolved in the solvent used for HPLC. The samples were separated on two consecutive HPLC systems, but the order was reversed compared to the description above. The sample was first separated on a Spherisorb-silica column (3 X 250 mm, particle size 5 pm) with 2.5% isopropanol in hexane as solvent (1ml/ml). From the distribution of radioactivity in the collected fractions (l/min) the relative amount of tritium-labeled vitamin D3 (retention time 3.2 min) and 25-hydroxyvitamin Da (retention time 7.3 min) could be determined. Only one-tenth of the fraction that contained vitamin DBand vitamin Dz was used for determination of radioactivity; the remainder was separated on a Zorbax-ODS column (4.6 X 250 mm) with 1.5% Hz0 in methanol as eluting solvent (1.2 ml/min). This system separated vitamin DPand vitamin DS (retention times, 13.2 and 14.4 min, respectively). The radioactivity in the vitamin D3 fraction was determined. After correcting for recovery of the added vitamin DZ the amount of radioactive vitamin DBoriginally taken up by the cells was calculated. (The assumption was made that the losses during extraction and chromatography were identical for vitamins D? and D?.) RESULTS

Uptake of PHIVitamin D3 in Liver Cells in Vivo-After TABLE I Uptake of ["H]uitamin D3 in liver cells in vivo The experimental procedure is described under "Materials and Methods." Number of isolated cells Time after injection

Hepatocytes min

30 70

Nonparenchymal cells

1O6

Recovered [ 'Hlvitamin D, Hepatocytes

Nonparenchymai cells

cpm/1o6 cells

31.2 110 27.4 330

460 230

Time (h) FIG. I. Uptake of vitamin Da by isolated rat liver cella. Hepatocytes (0,10.6 x lo6 cells) and nonparenchymal ceUs (0,5.1 x lo6 cells) from the liver of a normal rat were incubated in a volume of 2 ml as described under "Materials and Methods" in the presence of 171 nmol of vitamin DS for varying periods of time. The cells were separated from the medium, washed 3 times, and the amounts of vitamin D3 associated with the cells were determined by isotope dilution-mass spectrometry as described under "Materials and Methods.'' The results are given as amount of vitamin DRper lo6 cells.

Metabolism o f Vitamin Ds by Isolated Rat Liver Cells

10432

intravenous injection of [3H]vitamin D3 into 2 rachitic ratsIt approximately 58% of the radioactivity was recovered in the liver of one of the rats30 min after the injection, and thesame amount was also recovered from the liver of the other rat 70 min after the injection. The blood contained 6.9 and 4.8% of the radioactivity at these same times, respectively. In the blood, the fraction of the radioactivity that corresponded to 25-hydroxyvitamin Ds increased from 1.9% a t 30 min to 5.7% at 70 min. After removal of the livers and separation of the hepatocytes from the nonparenchymal cells it was found that [‘HI vitamin D3 had been taken up by both cell types (Table I). The data showed that the relative proportion of vitamin D3 that accumulated in the nonparenchymal cells increased with time.

could be argued thatthe pronase treatment used to obtain pure nonparenchymalcells (11)would lead to liberation of vitamin D3 from the hepatocytes and subsequent binding to the nonparenchymal cells. When the experiment reported in Table I was repeated and the cells were separated by the differential centrifugation method (lo), essentially the same results as those reported were obtained. Practically all the radioactivity recovered from the isolated cells cochromatographed with vitamin DR (96%) on HPLC. Theamountthat cochromatographed with 25-hydroxyvitamin D3 was very small (at 70 min, 3% in the hepatocytes and 1.4%in the nonparenchymal cells). Uptake of fH]Vitamin D3 during Perfusion of the Isolated Liver-An intact liver removed from a rachitic rat was perfused for 1 h with [3H]vitamin D3 added to the perfusion

TABLE I1 Uptake of t3H]vitamin D3 in liver cells after perfusion of rat liver in vitro The experimental procedures is given under “Materials and Methods.” The nonparenchymal cells were prepared by pronase treatment (cf. “Materials and Methods”). Recovered radioactivity Number of isoTotal lated cells Vitamin DJ 25-Hydroxyvitamin D:, cpm/lo6cells

cpm

106

6,200 Perfusion medium (58.5%) Hepatocytes NonDarenchvmaf cells

20,700 10,300 52 272.5 32.2

35,500 14,200 4,700

8

12

8 12 Tlme (min)

16

L

Time(mm)

cpm/1Ohcells

cpm

cpm/1Ohcells

38 I26

(17.4%) 1,600 (11.3%) 60 ( 1.3%)

6

(72.6%) 145

4,000 (86.3%)

i

1‘“ L

cpm

D

C

I

-

16

2

II

12 Tlme (mln)

L

8

1G

u L 8 12 Time (min)

16

FIG. 2. High pressure liquid chromatography on a Spheri- liver of a rachitic rat. The incubation time was 2 h. A , authentic 25sorb silica column (3 X 250mm) after prepurification on a hydroxyvitamin D3 (43 ng); B, complete incubation system with 2 X Zorbax-ODScolumnof the methanol/chloroform extract of lo7 parenchymal liver cells; C, blank incubation as in B, but the cell incubations of isolated liver cells and vitamin Da. The experi- suspension had been boiled for 10 min prior to incubation; D,complete mental conditions including extraction and chromatography are given ’ incubationsystem with 5 X lo6 nonparenchymal liver cells. The under “Materials and Methods.” The cells were isolated from the eluting solvent was 2.5% isopropanol in hexane, 0.8 ml/min. I

I

X

4 2

Time l h )

3

1

80

Z t

/f // 2

L

Time ( h )

I

7-1 FIG. 3. Effect of time on the formation of 25-hydroxyvitamin Da by liver cells isolated from normal ( A ) or rachitic (B) rats. The incubation conditions were as described under “Materials and Methods.” 10.6 X lo6 hepatocytes (0)and 5.1 X lofinonparenchymal cells (0)were used per incubation in A , and 10.6 X IO6 and 7.6 X IO6, respectively, in B. The results are given as amount of 25-hydroxyvitamin DRformed per lo6 cells. 6

Metabolism of Vitamin D3 by Isolated Rat Liver Cells

10433

extraction. The very low content of vitamin DBin these cells (Fig. 1) is another argument in favor of a true uptake. The uptake was time dependent, and the nonparenchymal cells appeared to have a somewhat higher capacity than the hepatocytes (Fig. 1). 25-Hydroxylation of Vitamin DBin Isolated Liver CellsWhen isolated hepatocytes were incubated in the presence of vitamin DGone more polar product could be detected by HPLC of the incubation extract. The retention time of this product was identical with authentic 25-hydroxyvitamin DB both by reversed-phase chromatography on the Zorbax-ODS column and by rechromatography on the silica column (Fig. 2B). Due to lack of sufficient amounts of material it was not possible to obtain a full mass spectrum of the product. The most prominent peaks in the mass spectrum of the trimethylsilyl/t-butyldimethylsilyl derivative of 25-hydroxyvitamin D3 are those at m / e 131, m / e 439, and m / e 586 (13). After prepurification on HPLC, the mass fragmentographic analysis showed that all these ions were present in the gas chromatographic peak corresponding to the derivative of the product isolated from the incubations with vitamin DB. Furthermore, within the experimental errors, the ratios between the intensities of these different ions were identical with the corresponding ratios obtained in an analysis of derivative of authentic 25-hydroxyvitamin DB.It is, therefore, reasonable to conclude that theidentity of the product is indeed 25-hydroxyvitamin D3. The quantitativedetermination by the mass fragI 1 I I I I mentographic method (13)of the amount of product formed yielded results identical with that based on the peak height after thesecond HPLC step(et: "Materials and Methods"). When nonparenchymal cells were incubated inthe presence of vitamin DB, none or only an insignificant amount of the more polar product was formed (Fig. 2 0 ) . This formation could entirely be explained by the contamination of this cell fraction with parenchymal cells. With cells isolated from normal rats the totalformation of 25-hydroxyvitamin DBwas linear with timeat least up to 4 hr (Fig. 3A). In separate experiments it was found that about 50% of the formed 25-hydroxyvitamin D, was released to the medium. Linearity with time was also observed with cells isolated from rachitic rats (Fig. 3 B ) . The rate of product c formation by the hepatocytes was within the same range 0 .4whether the cells were isolated from normal or from rachitic rats. Only after a prolonged time of incubation was formation of 25-hydroxyvitamin DBby the nonparenchymal cell preparations measurable (Fig. 3, A and B ) . Linearity with the number of cells was found to be satisfactory up to about 10' cells (Fig. 4). In some experiments with Number of cells ( x ~ O )- ~ FIG. 4. Effect of cell number onthe formation of 25-hydroxy- a higher number of cells the reaction leveled off; this was vitamin D3 by liver cells isolated from normal rat. The time of observed inparticular with hepatocytes isolated from the incubation was 2 h; other conditions were as described under "Mate- rachitic rats. nonparenchymal cells. rials and Methods." 0,parenchymal cells; 0, This lack of linearity at very high cell numbers was not

medium. After perfusion for 10-15 more min with the collagenase-containing buffer the cells were separated by use of the pronase method. About two-thirds of the radioactivity was recovered in the perfusion medium (Table 11). Of this, as much as 17.4% cochromatographed with 25-hydroxyvitamin D3indicating that theadded vitamin DBhad been extensively metabolized by the liver cells during the perfusion. A considerable amount of the radioactivity was recovered both in the hepatocytes and in the nonparenchymal cells (Table 11). When expressed per number of cells, this last cell type contained about 3 times as much radioactivityas thehepatocytes (Table 11). The fraction of the radioactivity that cochromatographed with 25-hydroxyvitamin DBcorresponded to about 11%in the hepatocytes but was rather insignificant in the nonparenchymal cells (Table 11). This may indicate that the 25-hydroxylation of vitamin DB primarily takes place in the hepatocytes. Uptake of Vitamin DBduring Incubation of Isolated Liver Cells-Considerable amounts of vitamin D3 were associated with boththe hepatocytes and thenonparenchymal cells after incubation in a mediumthat contained 86 nmol/ml of vitamin DB(Fig. 1).Before extraction both cell preparations had been washed 3 times, andit is reasonable to assume that the vitamin DB hadreally been taken upby the cells and not only bound to thecell surface. The zero-time samples were left on ice during the entire incubation period (up to 4 h) before

2

FIG. 5. Effect of vitamin D3 concentration on the formation of 25hydroxyvitamin Ds by hepatocytes isolated from normal ( A ) or rachitic

0"

:% x

-*

LrI/ / A

I

I

r

B C

I P

I

I

7" 7, -

0

c

8

E-

5 a5 P

''

flziz !-:

! ~ ~ ~ ~ us ~ ~ b ~~ ~ l t f~ i n ! ~~: ~ and Methods" for 2 h and with varying ; amounts of vitamin DS. The results are z 2 given as amount of 25-hydroxyvitamin s _a DJ formedper IO6 cells during 2 h of incubation. 9

g;

~

6

T T

A'?

L

N O L

02

+

I -

~

E

U

Vltarnin D, Inrnol)

I

I

200

GOO

Vitarnln D, (nrnol)

1 600

1 800

10434

Metabolism of Vitamin D3 by Isolated Rat Liver Cells

caused byhypoxia because continuous oxygenation of the medium did not influence the results. Presumably the explanation is that the higher the cell concentration the higher the tendencytoward aggregation. Lesssubstrate will thus be available for uptake per cell. The very small formation of 25hydroxyvitamin D3 observed with the nonparenchymal cell preparation was only measurable at very high cell numbers (Fig. 4). With a fixed number of parenchymal cells (9.4 X lo6) per incubation the formation of 25-hydroxyvitamin Ds was dependent on the amount of substrate added up to about 130 nmol per incubation whena saturation level was reached (Fig. 5).Reciprocal plotting of the data indicated apparent K, values of about 4 PM and 6 PM with hepatocytes from normal and rachitic rats, respectively. The maximum rates of the reaction were about 3.5 and 4 pmol X h" X cells in the two cases (Fig. 5, A and B ) . DISCUSSION

stored vitamin D3 might subsequently be transferred to the hepatocytes for hydroxylation. Two different vitamin D3 25-hydroxylases have been characterized in rat liver. One is localized to the endoplasmic reticulum (20-23) and the other in the mitochondria (7, 13, 24) The fist appears to havea high affinity but low capacity, while the second has a low affinity and a high capacity. This last one may be responsible for the continuous increase in circulating 25-hydroxyvitamin D3 after increasing dosage of vitamin D3 (25). Any feedback regulation of the 25-hydroxylase activity does not appear to be very effective and is not able to prevent toxic effects of large doses of vitamin Ds. Uptake and storage of vitamin D3in the nonparenchymalcells could be one way that the organism to some extent could protect itself from the toxic effects of large doses of vitamin DJ. Acknowledgment-We are grateful to Dr. Christian A. Drevon for help and advice with the preparation of liver cells.

Previous studies have established that vitamin D3 given intravenously to rats to a large extentis taken upby the liver REFERENCES (3, 15-17). The present results haveconfirmed these findings. 1. DeLuca, H. F., and Schnoes, H. K. (1976) Annu. Reu. Biochem. Furthermore, the results have shown that both the hepato45,631-666 cytes and the nonparenchymal liver cells of rachitic rats are 2. Ponchon, G., Keenan, A. L., and DeLuca, H. F. (1969) J. Clin. active in removing vitamin D3 from the circulation. If one Znuest. 48, 2032-2037 3. Olson, E.B., Jr., Knutson, J. C., Bhattacharyya, M. H., and assumes that about two-thirds of the liver cells are hepatoDeLuca, H. F. (1976) J. Clin. Znuest. 57, 1213-1220 cytes and one-third nonparenchymal cells (18), from the data in Table I one can estimate that30 min after the intravenous 4. Van Berkel, T. J . C., and Van Tol, A. (1978) Biochim. Biophys. Acta 530, 299-304 injection of labeled vitamin Da dissolved in ethanol about90% 5. Van Berkel, T. J. C., and Van Tol, A. (1979) Biochem. Biophys. of the radioactivity taken up by the liver is recovered in the Res. Commun. 89, 1097-1101 hepatocytes and the remainderin the nonparenchymal cells. 6. Bjorkhem, I., and Holmberg, I. (1976) Clin. Chim. Acta 68,215221 Seventy min after the injection the proportion of the radio7. Bjorkhem, I., Holmberg, I., Oftebro, H., and Pedersen, J. I. (1980) activity in the nonparenchymal cells had increased to about J. Biol. Chem. 255, 5244-5249 40%. 8. Seglen, P. 0. (1976) Methods Cell Biol. 13, 29-83 The importanceof the nonparenchymalcells for the uptake 9. Berry, M. N., and Friend, D. S. (1969) J . Cell Biol. 43, 506-520 of vitamin D3 was even more pronounced whenthe liver was 10. Nilsson, M., and Berg, T. (1977) Biochim.Biophys. Acta 497, perfused in vitro and with vitaminDR added to theperfusion 171-182 11. Berg, T., and Boman, D. (1973) Biochim. Biophys. Acta321,585medium (Table 11). 596 Also, the capacity of isolatedlivercells toaccumulate vitamin DSfrom the medium appears to be considerable (Fig. 12. Berg, T., Boman, D., and Seglen, P. 0. (1972) Exp. Cell Res. 72, 571-574 1). About 2.5 fmol of vitamin DRwere associated with each 13. Bjorkhem, I., and Holmberg, I. (1978) J . Biol. Chem. 253, 842hepatocyte or nonparenchymal cell after 1 h of incubation 849 (Fig. 1). 14. Bjorkhem, I., and Larsson, A. (1978) Clin. Chim. Acta 88, 559567 The findings that vitaminDSis efficiently taken up bothby the hepatocytes and by the nonparenchymalcells suggest that 15. Norman, A.W., and DeLuca, H.F. (1963) Biochemistry 2, 11601168 both cell types may be of importance in the metabolism of 16. Neville, P. F., and DeLuca, H. F. (1966) Biochemistry 5, 2201vitamin D3. It should be kept in mind, however that giving 2207 vitamin DSdissolved in ethanol is rather unphysiological and 17. Roianasathit. S..and Haddad, J.G. (1976) Biochim. BioDhvs. that the situation maybe different under normal conditions. Acta 421, i2-21 25-Hydroxylation is the most important metabolic conver- 18. Van Berkel, T. J. C. (1979) Trends Biochem. Sci. 4, 202-205 sion of vitamin DJthat occurs in the liver (1-3). In a previous 19. Reitano, J. F., Reed, M. A., Rostron, P.L., Intenzo, C. M., and Capuzzi, D. M. (1977) Mol. Cell. Biochem. 15, 213-217 study it was shown that isolated liver cells are able to carry 20. Bhattacharyya, M. H., and DeLuca, H. F. (1974) Arch. Biochern. out this reaction (19). No attempt was made, however, to Biophys. 160, 58-62 separate the parenchymal from the nonparenchymal liver 21. Delvin, E. E., Arabian, A,, and Glorieux, F. H. (1978) Biochem. J . cells. The incubation experiments reported here strongly in172,417-422 dicate that the 25-hydroxylation takes place only in the he- 22. Sulimovici, S.,Roginsky, M. S.,Duffy, J. L., and Pfeifer, R. F. (1979) Arch, Biochem. Biophys. 195,45-52 cells are patocytesand at comparablerateswhetherthe 23. Madhok, T. C., and DeLuca, H. F. (1979) Biochem. J . 184, 491derived from rachitic or from normal rats. For the moment 499 we can only speculate on what function the nonparenchymal 24. Bjorkhem, I., and Holmberg, I. (1979) J. Biol. Chem. 254, 9518cells may play inthe metabolismof vitamin D3. One possibility 9524 is that these cells may serve as a site of storage for vitamin 25. Shepard, R. M., and DeLuca, H. F. (1980) Arch. Biochem. Biophys. 202,43-53 DR.When the need for more 25-hydroxyvitamin D3 arises the ,