JOURNAL OF CELLULAR PHYSIOLOGY 135:545-550 (1988)

Demonstration of 1 a,25-Dihydroxyvitamin D3 Receptor-Like Molecule in ST 13 and 3T3 L1 Preadipocytes and its Inhibitory Effects on Preadipocyte Differentiation

M A Y U M I S A W * AND AKlYOSHl H I R A C U N Department of Oncology, Tokyo Metropolitan institute of Medical Science,

Honkomagome 3-18-22, Bunkyo-Ku, Tokyo 113, lapan

The active metabolite of vitamin D3, 1 a,25-dihydroxyvitamin D3 (1,25(OH),D,), inhibited morphologic and enzymatic expression during differentiation of preadipocyte to adipocyte. In the presence of -6.4-20 x M 1 ,25(OH)>Di, the triacylglycerol accumulation was only 50% of that of fully differentiated control cells. High-affinity binding sites for 1,25-dihydroxyvitamin Di were detected in two preadipose cell lines. The 1 ,25(OH)2D3 binding component sediments at 3.3 S in 4-24% (wiv) sucrose gradients prepared in hypertonic buffer. Binding assay revealed that N,,, was 70 fmolimg protein and 90 frnolimg protein, and Kd value was 170 p M and 37 p M in cell lines ST 13 and 3T3 L1, respectively. We also found that differentiated adipocytes did not contain spe- cific receptors for 1 ,25(OH)*D3. 1 ,25(OHI2D3, 1 (OH)D3, 24,25(OH)>D3, and 24(OH)D3 all suppressed differentiation of preadipocytes to adipocytes, and the dose required closely reflected the affinities of the various metabolites and the synthetic derivative for 1 ,25(OH)>D3 receptor. It i s suggested that the action of vitamin Di on preadipocyte differentiation may result from a receptor-mediated event

We have established a clonal preadipose cell line (ST 13) from ddN mouse and have discovered the physio- logic modulator that controls terminal differentiation of preadipocyte to adipocyte (Hiragun et al., 1980). We have previously demonstrated that retinoids inhibited ST 13 preadipocyte differentiation a t physiologic con- centration and have shown the presence of intracellu- lar retinoid binding proteins in the cytosol of ST 13 cells (Sato et al., 1980, 1988). Prostaglandins Az, Dz, Ez, El and Fza induced differentiation of ST 13 pre- adipose cells to adipose cells (Sato et al., 1982). Inter- feron effectively inhibited differentiation of ST 13 (unpublished observations) and 3T3 L1 preadipose cells to adipose cells (Keay and Grossberg, 1980). Ignotz and Massague (1985) reported that TGF-P inhibited con- version of preadipocytes 3T3 L1 to adipocytes, and this cell line contained high-affinity receptors to the TGF-P.

Recently, we found that the differentiation of ST 13 and 3T3 L1 preadipocytes to adipocytes was reversibly inhibited by 1,25-dihydroxyvitamin D3, and we dem- onstrated the presence of high-affinity binding sites in both preadipose cell lines. Intracellular receptor for 1,25-dihydroxyvitamin D3 was present in many tis- sues, and a wide variety of biologic responses to this vitamin were observed in these target tissues. It in- duces differentiation of mouse (Miyaura et al., 1981) and human (Abe et al., 1981) promyelocytic leukemia cells and mouse epidermal cells (Hosomi et al., 1983) and inhibits differentiation of Friend erythroleukemia cells (Suda et al., 1984). These effects are believed to be mediated by an intracellular receptor for 1,25-

dihydroxyvitamin D3. Structural study of vitamin D3 receptor protein revealed a cysteine-, lysine-, and arginine-rich region possessing high homology with both the steroid receptor family and the avian erythro- blastosis gag-erb A gene product (McDonnell et al., 1987).

The present work demonstrates for the first time the presence of the vitamin D3 receptor-like macromole- cule in ST 13 and 3T3 L1 preadipose cell lines. 3T3 L1 cells were established from Swiss 3T3 mouse embryo fibroblast (Green and Kehinde, 1974) and ST 13 from stromal tissue of a ddN mouse mammary carcinoma. The origin of these cell lines differs, in that 3T3 L1 has an embryonal origin and ST 13 an adult origin. How- ever, the principal differentiation characteristics are similar for both cell lines. In this report, we describe the effect of vitamin D3 on terminal differentiation of preadipocyte to adipocyte, utilizing these two cell lines.

MATERIALS AND METHODS Chemicals

1a25-(OH)2-[22,23-3HlD3(120 Ciimmol) and lmethyl- 3H]thymidine(45Ci/mmol) were obtained from Amer-

Received August 20, 1987; accepted February 10, 1988. *To whom reprint requests/correspondence should be addressed. Abbreviations used: 1,25(OH)2D3, la,25-dihydroxyvitamin DB; l(OH)D,,la-hydroxyvitamin D3; 25(OH)D3, 25-hydroxyvitamin D,; 24,251(OH)~D~, 24-R,25-dihydroxyvitamin D3; RII, the regu- latory subunit of CAMP-dependent type I1 protein kinase.

0 1988 ALAN R. LISS, INC

SAT0 AND HIRAGUN 546 sham Corp. (Arlington Heights, IL). Hydroxylapatite (Bio-Gel HTP) was purchased from BioRad Laborato- ries (Richmond, CA). Vitamin D3 derivatives, i.e., 1,25(OH)zD3, 24,25(OH)zD3, 25(OH)D3, and 1(OH)D3, were kindly provided by Drs. Masafumi Fukushima and Junko Abe of Chugai Pharmaceutical (Tokyo, Japan). They were dissolved in ethanol.

Cell culture and induction of adipose conversion Stocks of ST 13 cells and 3T3 L1 cells were main-

tained in Dulbecco’s modified Eagle’s medium-Ham’s F-12 medium (1:l) supplemented with 10% calf serum and 60pgiml kanamycin (growth medium). For induc- tion of adipose conversion, cells were plated a t a density of 4 x lo3 cellsicm2 with the growth medium and were re-fed with medium supplemented with 10% fetal bovine serum and 10 pgiml insulin (induction medium) on the following day (day 1). The medium was changed every 3 days up to the 14th day, when adipose conversion was evaluated.

Assay of 1,25(OH)& receptor ST 13 cells and 3T3 L1 cells were washed with

Ca2 + -Mg2 + -free phosphate-buffered saline (PBS-) and were scraped off using a rubber policeman. Cells were disrupted by sonication and centrifuged in hypertonic buffer containing 0.3 M KC1, 2 mM EDTA, 0.5 mM dithiothreitol, 10 mM molybdate, 1 mM phenylmethyl- sulfonyl fluoride, and 10 mM Tris-HC1, pH 7.4 (KTEDM buffer) a t 105g a t 60 min. The supernatant, containing 0.2 mg protein10.2 ml, was incubated at 25°C for 60 min with concentrations of 0.03-0.9 nM [3Hll,25(OH)zD3. Nonspecific binding was assessed in parallel samples by incubation of a 300-fold excess of radio-inert 1,25(OHI2D3. Bound and free l3H11,25 (OH)zD3 were separated by the hydroxylapatite method (Wecksler and Norman, 1979). For sucrose density analysis, KTEDM buffer extracts (0.3 mg protein10.3 ml) were incubated with a saturating con- centration (1nM) of 1,25(OH)2[3HlD3 a t 25°C for 60 min. After dextran charcoal adsorption of free 1,25(OH)2[3HlD3, the reaction mixture was layered onto linear 4-24% (wlv) sucrose gradients prepared in KTEDM buffer and centrifuged a t 2°C for 18 hr a t 230,OOOg.

Immunological detection of the regulatory subunit of CAMP-dependent type I1 protein

kinase (RII) Rabbit anti-RII serum was prepared using purified

bovine heart RII as the antigen. The antiserum was purified by affinity chromatography on a bovine heart RII-conjugated Affigel-11 column. Specific IgG was eluted with 4 M MgC12 in PBS-. The IgG was dialized against PBS- and concentrated prior to use. The spec- ificity of anti-RII-IgG was determined by immunoblot analysis and immunoprecipitation experiments as pre- viously described (Sato et al., 1988). Control immuno- globlins were separated from nonimmune rabbit serum by affinity chromatography on DEAE Affigel Blue.

Other procedures Estimation of triacylglycerol and protein content

was performed as described previously (Sato et al., 1980). To measure the synthesis of DNA, ST 13 mono-

3T3 L1 I A- 10%

t

I B’ ST 13

O I I I I 6 ’ 8 I I ’ 2 4 6 8 ’ I ’ lo3’ I ’

Days after cell seeding

Fig. 1. Effects of 1,25(OH)2D3 on ST 13 and 3T3 L1 cell growth. Cells were inoculated a t 3 x lo3 cells/24-multiwell dish (1.8 em2) on day 0 in induction medium containing M or lo-@ M 1,25(OH)zD3. Cell viability was > 99%. W , control (0.1% ethanol); 0, M 1,25(OH)2D3; o lo-* M ~ , ~ E J ( O H ) ~ D ~ . Each point represents the mean value of three individual experiments.

layers incubated for 1 hr with 5 pCi r3H]thymidine in Dulbecco’s modified Eagle’s medium were washed with PBS- and solubilized with 0.5% SDS. Cell lysates were transferred to tubes, placed on ice, and treated with 10% trichloroacetic acid. Precipitates were collected on nitrocellulose filters, and radioactivity was determined by liquid scintillation counting.

RESULTS Effects of 1,25(OH)zD3 on cell growth

and morphology Concentrations of 1,25(OH)zD3 that almost com-

pletely blocked conversion of preadipocytes to adi- pocytes were not cytotoxic. The doubling time of ST 13 cells and 3T3 L1 cells was not affected by treating with 1,25(OH)2D3 at concentrations, of lop7 M or lops M. The saturation density of 1,25(OH)zD3-treated ST 13 cells was slightly lower than that of nontreated con- trols. In contrast to the ST 13 cell line, the saturation density of 1,25(OH)zD3-treated 3T3 L1 cells was higher than that of controls (Fig. 1). Inclusion of lop7 M 1,25(OH)zDS completely blocked induction of the differ- entiation phenotype, as characterized by morphology and triacylglycerol accumulation (Fig. 2). This inhibi- tion was reversible, since cell differentiation readily resumed after simple 1,25(OHI2D3 removal, without a need for cell replating (Fig. 2D).

Inhibition of preadipocyte differentiation by vitamin D3 derivatives

When preadipocytes were maintained as a confluent monolayer in a medium containing fetal bovine serum and insulin, more than 90% of them converted from fibroblast-like cells into triacylglycerol-containing adiopose-like cells within 2 weeks of cell seeding. The increased expression of type I1 CAMP-dependent pro- tein kinase during ST 13 adipose conversion (Sato et

DIHYDROXYVITAMIN D3 RECEPTOR IN PREADIPOCYTES 547

Fig. 2. Phase-contrast photomicrographs of ST 13 cells. Photomicro- graphs were taken on day 10. A: Cells were cultured in growth medium. B: Cells were cultured with induction medium. C: Cells were cultured with induction medium containing M 1,25(OH)2D3.

al., 1985) and of type I in 3T3 L1 preadipocyte differ- entiation (Liu, 1982) is an early biochemical marker for differentiation in both preadipocytes. Differentia- tion of preadipocytes was estimated by determining the content of triacylglycerol per mg of cellular protein and detection of CAMP-dependent protein kinase type I1 regulatory subunit protein in ST 13 cells. The effects of vitamin D3 derivatives on preadipocyte differentiation were examined with continuously treated cultures, with concentrations up to M in the induction medium. The triacylglycerol accumulation was esti- mated on the 14th day of cell seeding, when control cultures developed a homogeneous adipose conversion. The results are shown in the form of semilog dose- response plots in Figure 3. Differentiation was inhib- ited dose-dependently by 1,25(OH)2D3 or 1(OH)D3. Concentrations required to inhibit differentiation by 50% (IC50) are shown in Table 1.

To consider the effects of vitamin D3 metabolite and synthetic compound on the expression of CAMP- dependent protein kinase RII subunit during adipose conversion, we used a specific anti-RII IgG. Immuno-

D: Cells were cultured with induction medium containing M ~ , ~ F I ( O H ) ~ D ~ from day 1 to day 4 and then cultured with induction medium only. Bar 50 bm.

~

A. ST 13 I B.3T3 L1

JiW- 0 -10 -9 -8 -7 -6 0 -10 -9 -8 -7 (M)

Fig. 3. Effects of various concentrations of different analogues of vitamin D3 on triacylglycerol accumulation. Each derivative was added to induction medium from day 1 until day 14. Cells were harvested 14 days after cell seeding and were evaluated for triacylgly- cerol and protein content. A: ST 13 preadipose cells. B: 3T3 L1 preadipose cells. Each point represents the mean of triplicate counts. All SD values were less than 5% of each points. The analogues are: 0 , 1,25(OH)zD3; A, l(OHID3; W , 25(OH)DZ; 0, 24,25(OH)zD3.

548 SAT0 AND HIRAGUN

1 2 3 4 5 1' 2' 3' 4' 5'

Fig. 4. Detection of RII protein by immunoblot analysis. ST 13 cells were cultured for 10 days with 10-7 M vitamin D3 analogues in the induction medium. Each cell extract was subjected to electrophoresis (10 pg proteinil cm width) and electrically blotted to nitrocellulose membrane. Lanes 1-5 were reacted with anti-RII IgG (70 kgiml) and

lanes 1'-5' with preimmune IgG a t the same concentration. Lanes 1 and 1': 1,25(0H2D3-treated cells. Lanes 2 and 2': 1(OH)D3. Lanes 3 and 3': 25(OH)D3. Lanes 4 and 4': 24,25(OH)&. Lanes 5 and 5': Induction medium only.

TABLE 1. Inhibitory effect of vitamin D3 on adipose conversion'

Compound ST13 3T3 L1

1(OH)D3 7.0 x lo-' 7.0 x lo-' 1,25(OH)zD3 6.4 x lo-'' 2.0 x 10-9

25(OH)D3 5.7 x 10-7 > 10-7 24,25(OH)zD3 1.0 x 10-6 > 10-7

'1C5,,(M) is the concentration produced 50% inhibition of preadipocyte differen- tiation determined by triacylglycerol accumulation.

blot analysis of vitamin D3 compound-treated cells, as shown in Figure 4, revealed that 1,25(OH)2D3 effec- tively blocked increase of RII subunit during adipose conversion.

Detection of receptor-like molecule for 1,25(OH)zD3 in ST 13 cells and 3T3 L1 cells

Receptor-like molecules for 1,25(OH)2D3 were as- sayed by incubating the cytosols of ST 13 preadipose and adipose cells, and of 3T3 L1 preadipose and adipose cells, with increasing concentrations (-0.03-0.9 nM) of 1,25(OH)2[3H]D3 for 60 min at 25 "C. A mixing experiment using a 1:l mixture of 3T3 L1 preadipose and adipose-cell cytosol was also performed. As shown in Figure 5A, D, a saturation plot was obtained that revealed a plateau a t 0.4 nM for both ST 13 and 3T3 L1 preadipose cells. The maximal number of binding sites (Nmax) was 70 fmolimg protein for ST 13 preadipose cells and 90 fmol/mg protein for 3T3 L1 preadipose cells. Adipocytes of both ST 13 and 3T3 L1 exhibited greatly diminished binding capacity, as compared with undifferentiated cells (Fig. 5B, E). Scatchard analysis of the specific binding of 1,25(OH)2[3H]D3 for preadi-

pose and adipose ST 13 and 3T3 L1 cells is shown in Figure 5C, F. The dissociation constants (Kd value) were calculated as 170 and 37 pM for ST 13 cells and 3T3 L1 cells, respectively. A 1: l mixture of preadipose cells and adipose 3T3 L1 cytosol showed exactly the intermediate binding capacity with the same Kd value (Fig. 5D, F, dashed line). The specific binding of 1,25(OH)2[3H]D3 for ST 13 cells and 3T3 L1 cells was further characterized by sucrose gradient analysis (Fig. 6). The results showed a single peak in both preadipocytes a t approximately 3.3 S, similar to that of 1,25(OH)2D3 receptors characterized in several sys- tems. In contrast to preadipose cells, both adipocytes contained greatly diminished peaks of radioactivity a t the same migration position (Fig. 6C, D).

Figure 7 shows the time course of the levels for 1,25(OH)2D3 receptor binding capacity in ST 13 cells and 3T3 L1 cells cultured with induction medium and growth medium. When cultured in induction medium, more than 90% of the preadipocytes converted to adi- pocytes, and the number of binding sites for 1,25(OH)2D3 decreased, accompanied by differentia- tion. Of the cells cultured in growth medium, more than 90% remained morphologically undifferentiated, and the number of 1,25(OH)2D3 binding sites slightly decreased during 2 weeks of observation period. Cell growth characteristics were almost identical between cultures maintained in growth medium and induction medium, as judged by [3H] thymidine incorporation (Fig. 7). Therefore, decrease in 1,25(OH)2D3 binding sites for cells cultured with induction medium was the result of adipose conversion and was not correlated with cell growth. These results demonstrate that pre- adipocytes possess specific and high-affinity binding

DIHYDROXYVITAMIN D3 RECEPTOR IN PREADIPOCYTES 549

D

Kd=170pM k 0 '-'(nM) SO 100

Bound (pM)

r v

'0 E

0 a m

aJ al LL \

L

PE s m

nM

2 .o k F

Fig. 5. Analysis of 1,25(OH)2[3H]D3 binding in preadipocytes and adipoctyes. Dose-dependent binding of 1,25(OH)~[~HlD3 in ST 13 preadipocyte (A) and adipoctye (B), and 3T3 LI preadipocyte (D) and adipocyte (E). Specific binding ( 0 ) was total binding (0) minus nonspecific binding (A). Values are the means of triplicate determi- nation. Scatchard analysis of specific binding from ST 13 cells (C) and 3T3 L1 cell (F). Dashed line is a 1:l mixture of 3T3 L1 preadipocytes and adipocytes.

. . - 1 I

0' 5 10 15 20 2 5 30 5 10 15 20 25 30

Fraction

Fig. 6. Sucrose density gradient analysis of 1,25(OH)2[3H1D3 bind- ing in ST 13 preadipose cells (A) and adipose cells (C), 3T3 L1 preadipose cells (B), and adipose cells (D). Aliquots (0.3 ml) of KTEDM-extracted cytosol containing 1 mg protein were incubated for 1 hr at 25°C with 1 nM 1,25(OH)2[3HlD, alone (0 ) plus 200, nM unlabeled 1,24(OH)zD3 (0). As external standards, the arrows indi- cate sedimentation of chymotrypsinogen (2.5 S) and ovalbumin (3.7 S).

L

1

c 1 ;

sites for 1,25(OH)2D3 a t significant levels, whereas adipocytes possess very few receptors, if any. This observation corresponds to the results of sucrose den- sity gradient analysis.

DISCUSSION In this report, we have demonstrated that the pres-

ence of a specific binding component for 1,25(OH)& in preadipose cell lines is associated with a functional response. These preadipose cells possess specific receptor-like macromolecules for 1,25(OH)zD3 that have biochemical properties highly similar to those of other 1,25(OH)& receptors. Specific binding de- creased along with cell differentiation from preadi- pocytes into adipocytes in both ST 13 and 3T3 L1 cells. The decrease in 1,25(OH)2D3 binding associated with differentiation from preadipocytes to adipocytes was

Days after cell seeding

Fig. 7. Decrease of specific binding for 1,25(OH)2[3HlD3 accompa- nied by differentiation. (0-0) 3T3 L1 cells or (0-n) ST 13 cells were cultured with induction medium. (W-W) ST 13 cells were cultured with growth medium. Values are the means of triplicate determina- tions. 13H]thymidine incorporation was measured as described in Materials and Methods. Open bars show c3H1thymidine incorporation of ST 13 cells cultured with growth medium and hatched bars indicate ST 13 cells cultured with induction medium. Media were changed on 3, 5, 8, and 11 days after cell seeding, as indicated by arrows.

due to the decrease of binding sites and not to a change in affinity for 1,25(OH)& binding. A mixing experi- ment of cytosol from 3T3 L1 preadipocytes and adi- pocytes demonstrated the absence of factors in adi- pocytes that interfere with the receptor assay (Fig. 5). The results indicate that preadipocytes possess a sig- nificant level of specific binding sites to 1,25(OH)zD3,

550 SAT0 AND HIRAGUN

whereas adipocytes possess few specific binding sites. The small number of binding sites detected in cytosol of adipocytes might be derived from the small fraction (- 10%) of undifferentiated cells contaminating the differentiated cultures (Figs. 2B, 5, and 6).

Pike and Haussler (1983) demonstrated the presence of receptor for 1,25(OH)zD3 with an apparent dissocia- tion constant of 7.8 x M in intact Swiss 3T3 cells, and an apparent uptake constant of 6.3 x lo-'' M and Kd value of 6.1 X M in swiss 3T6 fibroblasts. The Kd value of the 3T3 L1 preadipose cell line was highly similar to that reported for the original cell line. These results indicate that fibroblast and preadipose fibroblast possess receptors with similar Kd values, but adipocytes do not possess receptors for vitamin D3.

Inhibition of preadipocyte differentiation was in- duced not only by 1,25(0HI2D3 but also by vitamin D3 metabolites 0; by a synthetic analogue. AS summarized in Table 1, the ability of three vitamin D metabolites and a synthetic analogue to inhibit differentiation in preadipocytes correlated well with their affinity for the 1,25(OH)2D3 cytoplasmic receptor. We therefore pre- sume that the action of vitamin D3 on adipocyte differ- entiation is a receptor-mediated event. The normal physiologic level of 1,25(OH)2D3 is 0.1 nM, but differ- entiation of preadipocytes to adipocytes is slightly affected a t 0.1 nM of 1,25(OH)2D3. The reason why preadipose cells required a high dose, rather than a physiologic concentration of 1 ,25(OH)2D3 to inhibit adipose conversion is unclear. Pike and Haussler (1983) reported that the Kd value of Swiss 3T6 cells was 10 times higher when assayed in intact cells than when assayed in the cytosol fraction. The presence of serum in the culture medium may cause a reduction in apparent affinity of the receptor for 1,25(OH),D3. We must therefore examine the dose requirement of 1,25(OH)zD3 for adipose conversion under serum free- conditions.

The evidence indicates that vitamin D3 can affect the expression of a genetic program involving differentia- tion of preadipocytes into adipocytes. Vitamin D3 in- hibited triacylglycerol accumulation and suppressed expression of CAMP-dependent protein kinase type I1 during adipose conversion of ST 13 cells, which forms an early biochemical marker. It is therefore likely that vitamin D3 acts at an early point in differentiation of preadipocytes to adipocytes in vitro.

The mechanism of action of vitamin D3 on the differentiation of ST 13 and 3T3 L1 preadipocytes remains unclear. Sucrose gradient analysis demon- strated binding protein to 1,25(OH)2D3 in ST 13 sedi- ments a t 3.3 S under high-salt conditions and 6 S under low-salt conditions (data not shown). This suggests that the receptor-like molecules in ST 13 preadipose

cells apparently have a normal vitamin D3 binding site and also a DNA binding site. Further characterization is needed with respect to DNA binding ability of receptor-like molecules in ST 13 preadipocyte.

ACKNOWLEDGMENTS This work was supported in part by a Grant-in-Aid

from the Ministry of Education, Science, and Culture of Japan.

LITERATURE CITED Abe, E., Miyaura, C., Sakagami, H., Takeda, M., Konno, K., Yama-

zaki, Y., Yoshiki, S., and Suda, T. (1981) Differentiation of mouse myeloid leukemia cells induced by a la25-dihydroxyvitamin D3. Proc. Natl. Acad. Sci. USA, 78t4990-4994.

Green, H., and Kehinde, 0. (1974) Sublines of mouse 3T3 cells that accumulate lipid. Cell, 1t113-116.

Hiragun, A,, Sato, M., and Mitsui, H. (1980) Establishment of a clonal cell line that differentiates into adipose cells in vitro. In Vitro, 16:685-693.

Hosomi, J., Hosoi, J., Abe, E., Suda, T., and Kuroki, T. (1983) Regulation of terminal differentiation of cultured mouse epidermal cells by a la,25-dihydroxyvitamin D3. Endocrinology, 113:1950- 1957.

Ignotz, R.A., and Massague, J . (1985) Type p transforming growth factor controls the adipogenic differentiation of 3T3 fibroblasts. Proc. Natl. Acad. Sci. USA, 8253530-8534.

Keay, S., and Grossberg, S.E. (1980) Interferon inhibits the conver- sion of 3T3-Ll mouse fibroblasts into adipocytes. Proc. Natl. Acad. Sci. USA, 77t4099-4103.

Liu, A.Y.C. (1982) Differentiation-specific increase of CAMP- dependent protein kinase in the 3T3 L1 cells. J. Biol. Chem., 257t298-306.

McDonnell, D.P., Mangelsdorf, D.J., Pike, J.W., Haussler, M.R., and OMalley, W. (1987) Molecular cloning of complementary DNA encoding the avian receptor for vitamin D. Science, 235t1214-1217.

Miyaura, C., Abe, E., Kuribayashi, T., Konno, K., Nishii, Y., and Suda, T. (1981) la25-dihydroxyvitamin D3 induces differentiation of human myeloid leukemia cells. Biochem. Biophys. Res. Com- mun., 102:937-943.

Pike, J.W., and Haussler, M.R. (1983) Association of 1,25- dihydroxyvitamin D3 with cultured 3T6 mouse fibroblasts. J. Biol. Chem., 258:8554-8560.

Sato, M., Hiragun, A., and Mitsui, H. (1980) Preadipocytes possess cellular retinoid binding proteins and their differentiation is inhib- ited by retinoids. Biochem. Biophys. Res. Commun., 95:1839-1845.

Sato, M., Hiragun, A., and Mitsui, H. (1982) The effect of prostaglan- dins on adipocyte differentiation. Tissue Culture Res. Commun., 1 :27-28.

Sato, M., Hiragun, A., and Mitsui, H. (1985) Differentiation- associated increase of CAMP-dependent type I1 protein kinase in a murine preadipose cell line (ST-13). Biochim. Biophys. Acta., 844: 296-305.

Sato M., Shudo, K., and Hiragun, A. (1988) Functional studies of newly synthesized benzoic acid derivatives: Identification of highly potent retinoid-like activity. J. Cell. Physiol., I35:1-10.

Suda, S., Enomoto, S., Abe, E., and Suda, T. (1984) Inhibition by 1 a,25-dihydroxy vitamin D3 of dimethyl sulfoxide-induced differ- entiation of Friend erythroleukemia cells. Biochem. Biophys. Res. Commun., 119:807-813.

Wecksler, W.R., and Norman, A.W. (1979) A hyroxyapatite batch assay for the quantitation of 1~25-dihydroxyvitamin D3-receptor complexes. Anal. Biochem., 92t314-323.