JOURNAL OF BONE AND MINERAL RESEARCH Volume 6, Number 10, 1991 Mary Ann Liebert, Inc., Publishers
Vitamin D Analogs with Low Affinity for the Vitamin D Binding Protein: Enhanced In Vitro and Decreased In Vivo Activity ROGER BOUILLON, KATRIEN ALLEWAERT, DA ZHEN XIANG, BIAUW KENG TAN, and HUGO VAN BAELEN
ABSTRACT The affinity of la,25-dihydroxyvitamin D, [ la,25-(0H),D3] and analogs with side-chain modifications [MC 903 or calcipotriol, MC 1147 or 24,24-dihomo-1a,25-(OH),D3and 1,25-(OH),-16ene-23yne-D3] for the vitamin D receptor and the serum vitamin D binding protein (DBP) were compared. The affinity of MC 903 for the receptor from chick and rat duodenum or from human peripheral blood mononuclear cells or HL-60 cells varied between 60 and 100% relative to the affinity of 1,25-(OH),D3. The relative affinity of 1,25(OH)2-16ene-23yne-D, and MC 1147 varied for the same receptors between 45-70 and 3.5-25'4'0, respectively. The relative affinity of MC 903 for human DBP was 30-fold decreased, whereas the two other analogs did not bind to DBP at all even in more than 1000-fold excess. The in vitro biologic activity of la,25(OH),D, on phytohemagglutinin-stimulated normal human lymphocyte proliferation was markedly inhibited by the addition of physiologic amounts of DBP to the cell culture medium. No such inhibition was observed when MC 903 or 1147 was evaluated similarly. DBP therefore reversed the rank order of the in vitro potency of these analogs. Intramuscular injections for 10 consecutive days to vitamin D-deficient chicks demonstrated a 1100-fold lower biologic activity of MC 903, MC 1147, and 1,25-(OH),-16ene-23yne-D3compared to that of la,25-(OH),D3 as evaluated by serum calcium and osteocalcin concentrations, as well as by duodenal calbindin D28K and bone calcium content. We conclude that the biologic activity of vitamin D metabolites and analogs depends on their affinity for the vitamin D receptor as well as their affinity for DBP. Analogs with a low DBP but good receptor binding properties display low in vivo biologic activity on calcium and bone homeostasis, at least partly due to altered pharmacokinetics.
INTRODUCTION D HORMONE, 1,25-dihydroxyvitamin D [ 1,25-(OH),D,], not only regulates extracellular calcium metabolism but also modifies the secretion of several hormones and has profound influences on the immune system and cell differentiation.(1-5)Several analogs of 1,25(OH),D, with modification in the side chain(6-*1) have potent in vitro activity on normal or tumoral lymphocytes, whereas their in vivo activity on calcium metabolism is markedly decreased. The reason for this discrepancy is not fully understood, and differences in receptor binding, cellular access, and metabolism or differences in pharmacokinetics have all been suggested.
T
HE ACTIVE VITAMIN
Since the binding of vitamin D metabolites and analogs to the serum vitamin D binding protein (DBP) has a marked influence on their intracellular access to the vitamin D receptor and, therefore, biologic activity,('z-'4)we evaluated the in vitro and in vivo activity of analogs with side modifications and conclude that 1,25-(OH),D3 analogs with low DBP and good receptor binding properties display relatively high activity on lymphocyte proliferation in vitro but low biologic activity on calcium metabolism in vivo.
MATERIALS AND METHODS
Materials MC 903, calcipotriol, or 20(R)-[3'(S)-CycIopropyl-3'-hy-
Laboratorium voor Experimentele Geneeskunde en Endocrinologie, Katholieke Universiteit Leuven, Leuven, Belgium.
105 1
BOUILLON ET AL.
1052
droxyprop- l'(E)-enyll- 1 (S),3(R)-dihydroxy-9,lO-secopregna-S(Z),7(E), 10(19)-triene, and MC 1147, or 24,24-dihomo-I ,25-(OH),D3, were gifts from L. Binderup, Leo Pharmaceutical Products (Ballerup, Denmark); 1,25(OH),D, and 1,25-(OH),-16ene-23yne-D3were gifts from M. Uskokovic, Hoffman-La Roche (Nutley, NJ). [,HI1,25-(OH),D, (specific activity of 170 Ci/mmol) and [methyl-'Hlthymidine (2 Ci/mmol) were purchased from Amersham (Buckinghamshire, England). Aprotinine was obtained from Novo Industri (Bagsvaerd, Denmark). Human DBP and DBP-free serum were prepared by affinity chromatography as described previously. ('zl')
Methods Affinity for Vitamin D Receptor: Duodenal mucosa from normal Wistar rats or young chicks was sonicated in 4 vol of buffer (0.05 M Tris-HC1, 0.5 M KCI, 5 mM dithiotreitol, 10 mM Na,MoO,, and 1.5 mM EDTA, pH 7.5). Cytosol from HL-60 cells and phytohemagglutinin (PHA)-stimulated peripheral blood mononuclear cells (PBMC) were prepared similarly. The homogenate was centrifuged at 6OOO x g for 10 minutes and 30,000 x g for 1 h; the cytosol was stirred with ammonium sulfate (18% wt/vol) for 1 h and centrifuged; the pellet was then redissolved in the original buffer and used for binding studies. [,HI 1,25-(OH),D, (0.2 nM) and increasing concentrations of 1.25-(OH),D3 or its analogs were incubated with the duodenal cytosol in a final volume of 0.3 ml for 20 h at 4°C. Phase separation was then obtained by the addition of cold dextran-coated charcoal. Affinity for Human DBP: Binding of vitamin D metabolites and analogs to DBP was performed at 4°C essentially as described previously."') [)HI 1,25-(OH),D, and 1,25-(OH),D, or its analogs were added in 5 pl ethanol into glass tubes and incubated with DBP (0.18 pM) in a final volume of 1 ml (0.01 M Tris-HCI buffer and 0.154 M NaCI, pH 7.4) for 3 h at 4°C. Phase separation was then obtained by the addition of cold dextran-coated charcoal. Radioimmunoassay (RIA) for Chick Osteocalcin and Calbindin D28fi. Osteocalcin was purified from chick long bones essentially as described('s1and specific antibodies were raised in a rabbit. The RIA for chick osteocalcin used this rabbit antiserum (1:300,000 in 0.05 M phosphate buffer, pH 7.4, 1% BSA, 25 mM EDTA, and 0.125 M NaCI), [ l ~ S I ] ~ ~ t e ~ c a(iodination lcin by chloramine-T method, 15,000 cpm), and 100 p1 serum (diluted 200-fold) or standard in a total volume of 0.8 ml for 24 h at 4°C. Phase separation was then obtained by second antibody precipitation. Calbindin D28K was measured by RIA using a similar procedure with an antiserum against chick calbindin D28K (kindly provided by Norman) raised in a guinea pig (diluted 1:400,000) and ['ZSI]calbindinD28K, ionidated by the chloramine-T method. Homogenates of chick intestinal mucosa, obtained from 1,25-(OH),D,-treated normal chicks and calibrated versus purified calbindin D28K, were used as internal standard. The between-assay coefficient of variation (CV) was below 7%.
Effect of Vitamin D Metabolites and Analogs on Lymphocyte Proliferation: The human promyeolocytic leukemia cell line (HL-60) was obtained from the American Type Culture Collection (Rockville, MD). HL-60 cells were seeded at 1.2 x los cells per ml, and 1.25-(OH),D3 or its analogs were added in ethanol (final concentration below 0.2%) in RPMI-1640 medium supplemented with 10% heat-inactivated fetal calf serum (GIBCO, Roskilde, Denmark), 100 U/ml of penicillin. and 100 p g h l of streptomycin (Boehringer, Mannheim, Germany). After 4 days of culture in a humidified atmosphere of 5% CO, in air at 37"C, the dishes were shaken to loosen any adherent cells, and all cells were then assayed for maturation by nitroblue tetrazolium (NBT) reduction assay.(16)HL-60 cells at 1.5 x lo6 cells per ml were therefore mixed with an equal volume of a freshly prepared solution of phorbol 12-myristate 13-acetate (200 ng/ml; Sigma, St. Louis, MO) and NBT (2 mg/ml; Sigma) and incubated for 30 minutes at 37°C. The percentage of cells containing black formazan deposits was determined using a hemacytometer. Peripheral blood mixed mononuclear cells (PBMC) were prepared from buffy coats of human blood (obtained from the Belgian Red Cross) by centrifugation on Ficoll-Hypaque (Pharmacia, Uppsala, Sweden). The culture conditions were except that the identical to those previously de~cribed,"~) culture medium in the present study was RPMI-1640. Vitamin D-Deficient Chicks: Vitamin D-deficient chicks were reared on a vitamin D-deficient diet (1 Qo calcium and 1 Yo phosphorus; Hope Farms, Woerden, The Netherlands) and housed in a windowless room. After 3 weeks they received a daily intramuscular injection for 10 days with la,25-(OH),D3 or its analogs (solubilized in 10% ethanol, 10% Tween 80, and 80% NaCI, 0.9% in water). Chicks were then weighed and serum collected for measurements of calcium (by atomic absorptiometry) and osteocalcin (by RIA). The right tibia was removed and prepared free from adherent tissue, dried at 100°C for 24 h, and ashed at 600°C for 24 h in a muffle furnace. The ash of each bone was then weighed and dissolved in HCI (12 N) for measurement of calcium content. The duodenum was immediately excised, and the mucosa was scraped from the underlying muscle layer at 0°C. A homogenate was prepared in Tris Buffer (14 mM Tris HCI, pH 7.4, 0.125 M NaCI, 3 mM KCI, and 0.01 U/ml of aprotinine), centrifuged at 4°C at 10,000 x g for 1 h, and then assayed for calbindin D28K. Statistical analysis was performed using Student's t-tests.
RESULTS
Binding studies The affinity of 1,25-(OH),D, for the vitamin D receptor from rat duodenal mucosa (3.2 f 1.2 x lo9 M-', mean f SD; n = 6) was significantly (p < 0.01) lower than for chick duodenal mucosa (1.2 f 0.3 x I O ' O M - l ) , HL-60 cells (1.9 + 0.1 x 10" M-I), or PBMC (9.1 f 3.2 x lo9 M-I). The affinity of MC 903 was nearly identical to that of 1,25(OH)zD3 in HL-60 cells and chick duodenum and only slightly lower in rat duodenum or PBMC (Fig. 1). The relative affinity of MC 1147 varied between 3.5% (rat duode-
1053
VITAMIN D ANALOGS WITH LOW DBP AFFINITY
B
\
q.
~~
I -
1
1 L
1o - ~ 10-8 1 25- (OH12D3 OR ANALOG CONCENTRATION
10.’
ro-’O
(MI
I
10-’0
.
I
10-Q
10-0
1 25- (OH) 2D3 DA ANALOG CONCENTRATION
10
’
(MI
FIG. 1. Binding of [’H]1,25-(OH),D, in the presence or absence of nonradioactive 1,25-(OH),D, or its synthetic analogs to the vitamin D receptor from (A) rat duodenal mucosa; (B) chick duodenal mucosa; (C) HL-60 cells; and (D) phytohemagglutinin-stimulated human peripheral blood mononuclear cells. All values given as mean f standard deviation (SD; n = 6 in three experiments). ( 0 ) 1,25-(OH),D,; (H) MC 903 or calcipotriol; (A) MC 1147 or 24,24-dihomo-1,251 ,25-(0H),-26ene-23yne-D3. (OH),D,; (0)
num) and 25% (PBMC) compared to the affinity of 1,25OH),D,, whereas the relative affinity of 1,25-(OH),-16ene23yne-D3 varied between 45 and 70% (Fig. 1). The affinity of 1,25-(OH),D, for purified human DBP was 1.5 f 0.2 x 10’ M-’ (pH 7.4, 4°C; n = 6). The binding of the analogs with side-chain modifications was decreased: to obtain a 50% displacement a 30-fold molar excess of MC 903 was needed in comparison with 1,25(OH)& whereas MC 1 147 and 1,25-(0H),-16ene-23yneD, did not displace radioactive 1,25-(OH),D, even at a 1000-fold molar excess (Fig. 2).
Effect of vitamin D analogs on lymphocyte function in vitro The effect of vitamin D analogs on the proliferation of normal human lymphocytes stimulated by P H A was dependent on the presence of DBP during the culture period. A physiologic concentration of DBP (300 pg/ml or 5.8 pM) increased the concentration of 1,25-(OH),D, necessary for 50% of proliferation inhibition from 1 x to
5 x M (Fig. 3). In the absence of DBP, MC 903 was nearly equally active as 1,25-(OH),D, whereas MC 1147 was about 10-fold less potent. The addition of DBP did not markedly influence the activity of MC 903 and MC 1147 but decreased the effects of the natural hormone. In the presence of normal DBP concentrations, MC 903 therefore became the most active metabolite, followed by MC 1147, 1,25-(OH),D, being the least active (Fig. 3). The differentiating effects of the vitamin D hormone and its analogs were evaluated by NBT reduction in HL-60 cells in the presence of fetal calf serum (10%). MC 1147 was slightly more effective than 1,25-(OH),D, or MC 903, whereas 1 ,25-(OH),-16ene-23yne-D3 was markedly more effective (Fig. 4).
In vivo effects on calcium metabolism The daily intramuscular injection of 1,25-(OH),D, and its analogs increased but did not normalize the body weight of vitamin D-deficient chicks (data not shown) compared to animals raised on a D-replete diet. All vita-
BOUILLON ET AL.
1054
The duodenal calbindin D28K concentration was nearly undetectable in D-deficient chicks and increased to supraphysiologic amounts after a daily dose of even 20 ng of 1,25-(OH),D3. Compared to 1,25-(OH),D,, a more than 100-fold higher amount of MC 903 or MC 1147 was needed to increase the duodenal calbindin D28K concentration (Fig. 5 ) . The serum concentration of osteocalcin, measured by specific RIA, also increased above the concentration in Drepleted chicks after injection of 1,25-(OH),D, in an amount that still did not normalize serum calcium. Intramuscular injections of MC 903 and MC 1147 were again much less active (at least 40-fold) than 1,25-(OH),D, on osteoblastic osteocalcin secretion. Serum osteocalcin and duodenal calbindin D Z ~ concentrations K hardly increased after a 10 day in vivo administration of 1,25-(OH),-16ene23yne-D,, even when given in 100-fold excess compared to 1,25-(OH),D, (Fig. 5 ) .
150,
T 5
T
tK
Y
\ Y
1 25
I
1
10
10
(OH)Z03
-
10
10-6
O R A N A L O G C O N C E N T R A T I O N (MI
FIG. 2. Binding of [3H]1,25-(OH),D3 to purified human vitamin D binding protein in the presence or absence of nonradioactive 1,25-(0H),D3 or analogs. Same symbols as in Fig. 1.
-. 1
~~
10
10
10
lo
10
10
8
10
’
VITAMIN D H E l A B O L l l E OR A N A L O G IM1
FIG. 3. Effect of 1,25-(OH),D, or its analogs on the proliferation of PHA-stimulated mixed peripheral human blood lymphocytes as measured by [,H]thymidine incorporation. All measurements were performed in 6-fold microcultures in human DBP-free serum reconstituted (5.8 pM DBP, open symbols) or not (closed symbols) with human 1,25-(OH),D, (with and without DBP, reDBP. (O,.) MC 903 (with and without DBP, respectively); (El,.) spectively); (&A) MC 1147 (with and without DBP, respectively).
min D analogs were markedly less potent than the natural hormone. The serum calcium concentration and bone calcium content increased markedly by vitamin D therapy, but 1,25-(OH),D, was 30- to 100-fold more efficient than MC 903 and MC 1147 (Fig. 5 ) .
DISCUSSION The active vitamin D hormone has an important function not only in calcium metabolism but also in lymphocyte proliferation and differentiation. Several vitamin D analogs are more potent when tested in vitro in cell proliferation or differentiation than in systemic mineral homeO S ~ ~ S ~ S . ( ~ .Moreover, ~ . ~ . ~ ~some . ~ ~ of , ~these ~ ~ analogs were more active in cell differentiation in vitro than could be expected from their relative affinity for the vitamin D receptor.‘16.17) Such an effect may be due to target-specific differences in intrinsic receptor binding or activation. The biologic activity of vitamin D analogs, however, also depends on cellular accessibility (influenced by affinity for DBP) and metabolic clearance (probably also modified by affinity for DBP) or intracellular activation or inactivation in target cells. Differences in binding to the serum vitamin D binding protein could therefore at least partly explain differences in pharmacokinetics. Indeed, we previously demonstrated by in vitro and in vivo experiments that the free or unbound vitamin D metabolites represent the active hormone concentration. ( 1 * , 1 4 , ’ 8 ) Moreover, the addition of physiologic amounts of DBP decreased the biologic efficacy of natural vitamin D metabolites on human lymphocyte proliferation. ( I 3 ) The present data demonstrate that several vitamin D analogs with modifications in the side chain have markedly (30- to more than 200-fold) decreased DBP binding properties whereas the receptor affinity either remained largely unchanged (MC 903) or was mildly decreased [MC 1147 and 1,25-(OH),-16ene-23yne-D3], compared to the decrease in their affinity for DBP (Fig. 2). Other modifications in the side chain also decreased the affinity for DBP as described for 22-oxa-1,25-(OH),D3,‘19)26-homo-1,25(OH),D, ( p e r s o n a l o b s e r v a t i o n ) , a n d 1,25(OH),-26,27-dimethyl-D,.’” Recent data presented after initial submission of this manuscript using diluted serum and [’H]25-OHD3 instead of purified DBP and [,H]1,25(OH)2D, largely confirmed our data.(20)The affinity of MC 1147 for rat duodenal receptor and of 1,25-(OH),-
1055
VITAMIN D ANALOGS WITH LOW DBP AFFINITY
1011,
16ene-23yne-D, for the chick and human receptor agreed well with previous In experiments without extracellular DBP (serum-free or DBP-free cultures of isolated cells of tissues), MC 903 and MC 1147 displayed an activity closely resembling their affinity (Fig. 3). Previous experiments with other cell types also demonstrated that MC 903 was nearly equipotent to 1,25-(OH),D, when tested in medium with little or no DBP or s e ~ u m . ( ~ . In ~ ' -the ~ ~presence ) of DBP, however, 1,25(OH),D, became markedly less active (30-fold) as a result of decreased free hormone concentration. Because of low DBP binding the analogs remained totally accessible for cellular entry and activation. Addition of DBP therefore did not decrease the biologic activity of MC 903 or MC 1147 on human lymphocyte proliferation (Fig. 3). Similarly, addition of DBP to DBP-free fetal calf serum did not influence the biologic activity of 1,25-(OH),-16ene-23yneD, on HL-60 cells in vitro (unpublished personal results). The low in vivo effects of the vitamin D analogs with low DBP binding can also at least partly be explained by
? T
FIG. 4. Differentiating effect of 1,25-(OH),D, and its analogs on HL-60 cells, grown in RPMI-1640 medium supplemented with 10% FCS, as evaluated by NBT reduction. Same symbols as in Fig. 1.
r
I
I
-
100 W
0
Z m
--a -
0
.-
50 $
P r0
H
q
- e 3.
m w
I
10-11
10-l0
o-~
10-l1
1
la. 25- (OH) 203 OR ANALOG (rnol/day)
Io-g
10-lO
la, 25- (OH)2D3 OR ANALOG (rnol/day)
X
T
a
iF---++ 90-11 la,
10-l0
25- (OH) D, OR ANALOG (mol /day)
1o-g la,
10-11 25- (OH),D,
t0-l0 1 OR ANALOG (mol/day)
FIG. 5. Effects of 1,25-(OH),D, and its analogs on serum calcium concentration (top left), tibia1 calcium content (top right), serum osteocalcin (bottom right), and duodenal calbindin DZsKconcentration (bottom left) after 10 days of IM (intramuscular) injection of l,25-(OH),D3 or its analogs. All data are given as mean 1 SEM (standard error of the mean) using between 6 and 10 animals per group. ( 0 )1,25-(OH),D,; (m) MC 903; (A) MC 1147 or 24,24-dihomo-1,25(OH),D,; (0) 1 ,25-(OH),-16ene-23yne-D,; (*) data obtained in vitamin D,-replete chicks.
*
BOUILLON ET AL.
1056
their differences in affinity for DBP and the vitamin D receptor. The vitamin D analogs with low DBP binding remain mainly in the free form after administration and are therefore subject to more rapid degradation. Indeed, the half-life of MC 903 was much shorter than that of 1,25(OH),D, in both Sprague-Dawley rats (12 minutes) and minipigs [60 minutes after IV (intravenous) injection of MC 903; H. Sorensen, Leo Pharmaceutical Products, personal communication]. This rapid degradation can largely explain why serum and bone calcium concentration or content increased in vitamin D-deficient chick only after a nearly 100-fold higher dose of analogs compared to the natural hormone. Similar observations were made for more specific vitamin D-dependent proteins (duodenal calbindin D Z ~and K serum osteocalcin concentration). An even lower in vivo calcemic effect was previously in rachitic observed for 24,24-dihomo-1,25-(OH),D3 rats,(111whereas no bone-resorbing activity could be demonstrated in rat bone cultures in v i t r ~ . ( ’ ~ ”In~ Perlman’s study, however, the rat intestinal calcium absorption could with an also be stimulated by 24,24-dihomo-1,25-(OH),D3, approximately 10-fold lower potency than 1,25-(OH),D,. In contrast to the in vitro data with rat MC 1147 was equipotent on bone (serum osteocalcin) compared to the duodenum (calbindin D28~)in rachitic chicks. The discrepancy may be due to species differences in the relative affinity of MC 1147 for the chick duodenal receptor [15% of that of 1,25-(OH),D,] was greater than observed in the rat [3.5% of the affinity of 1,25-(OH),D3] (Fig. 1). Moreover, our in vivo data used the measurement of two vitamin D-dependent proteins that allowed the detection of a much greater difference between D-deficient and vitamin D-treated animals than simple serum calcium or calcium absorption tests (Fig. 5 ) . The low in vivo calcemic effects of l ,25-(OH),-16ene-23yne-D, confirm previous data after its acute administration in rachitic chicks(9) or Other vitamin D analogs with modifications in the C-ring (C1J behave quite differently as such analogs display increased DBP affinity with maintained or decreased receptor affinity. The addition of DBP therefore decreases the biologic availability and activity of such analogs in vitro.(zs) In conclusion, several vitamin D analogs with side-chain modification have a low affinity for DBP but (nearly) normal affinity for the vitamin D receptor. In the presence of DBP in vitro such analogs have therefore better access and increased activity but, when given in vivo, have low potency on bone and duodenum as a result of rapid degradation. Such differences in pharmacokinetics may find interesting applications, however, because short-term exposure to cells with different cell cycles may induce a totally different activity pattern.
No. 3.0044.89) is kindly acknowledged. Presented in part at the joint meeting of the ASBMR and ICCRH, Montreal 1989, Abstract 734.
REFERENCES 1. Manolagas SC, Provvedini DM, Tsoukas CD 1985 Interac-
2.
3. 4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
ACKNOWLEDGMENTS
15.
The financial support of the Belgian Nationaal Fonds voor Geneeskundig Wetenschappelij k Onderzoek (Grant
16.
tions of 1.25-dihydroxyvitamin D, and the immune system. Mol Cell Endocrinol 43:113-122. Suda T 1989 The role of la,25-dihydroxyvitarnin D, in the myeloid cell differentiation. Proc SOCExp Biol Med 192:214220. Bouillon R, Van Baelen H 1989 The biological roles of vitamin D. Saudi Med J 10:260-266. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980-991. DeLuca HF 1989 The vitamin D story: A collaborative effort of basic science and clinical medicine. Proc SOCExp Biol Med 192:224-236. Binderup L, Bramm E 1988 Effects of a novel vitamin D analogue MC 903 on cell proliferation and differentiation in vitro and on calcium metabolism in vivo. Biochem PharmaCOI 37~889-895. Paulson SK, Perlamn K, DeLuca HF, Stern PH 1990 24- and 26-homo-1,25-dihydroxyvitaminD, analogs: Potencies on in vitro bone resorption differ from those reported for cell differentiation. J Bone Miner Res 5201-206. Brown AJ, Ritter CR, Finch JL, Morrissey J, Martin KJ, Murayama E, Nishii Y, Slatopolsky E 1989 The noncalcemic analogue of vitamin D, 22-oxacalcitriol, suppresses parathyroid hormone synthesis and secretion. J Clin Invest 84:728732. Zhou J-Y, Norman AW, Lubbert M, Collins ED, Uskokovic MR, Koeffler HP 1989 Novel vitamin D analogs that modulate leukemic cell growth and differentiation with little effect on either intestinal calcium absorption or bone calcium mobilization., Blood 74:82-93. Gill HS, Londowski JM, Corradino RA, Zinsmeister AR, Kumar R 1988 The synthesis and biological activity of 25-hydroxy-26,27-dimethylvitaminD, and 1,25-dihydroxy-26,27dimethylvitamin D,: Highly potent novel analogs of vitamin D,. J Steroid Biochem 31:147-160. Perlman K, Kutner A, Prahl J, Smith C, Inaba M, Schnoes HK, DeLuca HF 1990 24-Homologated 1,25-dihydroxyvitamin D, compounds: Separation of calcium and cell differentiation activities. Biochemistry 29:190-196. Bouillon R, Nyomba B, Decleer W, Van Baelen H 1985 Vitamin D-binding protein and free l ,25-dihydroxyvitamin D in animals and clinical medicine. In: Norman AW, Schaefer K, Grigoleit HG, von Herrath D (eds.) Vitamin D, Chemical, Biochemical and Clinical Update. Walter de Gruyter, Berlin, pp. 669-672. Vanham G, Ban Baelen H, Tan BK, Bouillon R 1988 The effect of vitamin D analogs and of vitamin D-binding protein on lymphocyte proliferation. J Steroid Biochem 29:381-386. Bikle DD, Gee E 1989 Free, and not total, 1.25-dihydroxyvitamin D regulates 25-hydroxyvitamin D metabolism by keratinocytes. Endocrinology 124:649-654. Gundberg CM, Hauschka PV, Lian JB, Gallop PM 1984 Osteocalcin: Isolation, characterization and detection. Methods Enzymol 107516-544. Ostrem VK, Tanaka Y, Prahl J, DeLuca HF, Ikekawa N
VITAMIN D ANALOGS WITH LOW DBP AFFINITY
17.
18.
19.
20.
21.
22.
23.
1987 24- And 26-homo-I ,25-dihydroxyvitamin D,: Preferential activity in inducing differentiation of human leukemia cells HL-60 in vitro. Proc Natl Acad Sci USA 84:2610-2614. Abe J , Morikawa M, Miyamoto K, Kaiho S-I, Fukushima M, Miyaura C, Abe E, Suda T, Nishii Y 1987 Synthetic analogues of vitamin D, with an oxygen atom in the side chain skeleton. FEBS Lett 22658-62. Vargas S, Bouillon R, Van Baelen H, Raisz LG 1990 Effects of vitamin D-binding protein on bone resorption, stimulated by 1,25 dihydroxyvitamin D,. Calcif Tissue Int 47:164168. Okano T, Tsugawa N, Masuda S, Takeuchi A, Kobayashi T, Nishii Y 1989 Protein-binding properties of 22-oxa-la,25dihydroxyvitamin D,, a synthetic analogue of Ia,25-dihydroxyvitamin D,. J Nutr Sci Vitaminol 35529-533. Dusso AS, Lopez-Hilker S, Finch J , Mori T , Nishii Y , Slatopolsky E, Brown AJ 1990 On the mechanisms for the selective action of vitamin D analogs (abstract). J Bone Miner Res J(Suppl. 2):S202. Trydal T , Lillehaug JR, Aksnes L, Aarskog D 1989 The effect of the 1,25-(OH),D, analog, MC 903, o n non-transformed and transformed mouse embryo fibroblast cells (abstract). J Bone Miner Res 4(Suppl. 1):943. Evans DB, Thavarajah M, Binderup L, Kanis J A 1989 Comparison of the actions of MC 903 and 1,25-(OH),D, on human osteoblast-like cells in vitro (abstract). J Bone Miner Res 4(Suppl. 1):868. Valaja T, Mahonen A, Priskanen A, Maenpaa P H 1989 Af-
1057 finity of a novel vitamin D analog MC 903 for vitamin D receptor and effects of the analog on the synthesis of osteocalcin by human osteosarcoma cells (abstract). J Bone Miner Res 4(Suppl. 1):732. 24. Holick MF, Uskokovic M, Persons K, Horst R, Bagglioni EG, Truitt G A 1988 An evaluation of analogs of 1,25-dihydroxyvitamin D3 on the inhibition of proliferation and induction of terminal differentiation in cultured human normal keratinocytes. In: Norman AW, Schaefer K, Grigoliet HG, von Herrath D (eds.) Vitamin D, Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 363-371. 25. Bouillon R, Allewaert K, Xiang DZ, Vandewalle M. De Clercq P 1989 Structure-function analysis of vitamin D analogs: Difference in binding to the vitamin D receptor and vitamin D-binding serum protein (abstract). J Bone Miner Res 4(Suppl. 1):734.
Address reprint requests to: Dr. R . Bouillon Legendo Onderwijs en Navorsing Gas(huisberg B-3000 Leuven, Belgium Received for publication August 13, 1990; in revised form April 15, 1991; accepted April 17, 1991.
Vitamin D Analogs with Low Affinity for the Vitamin D Binding Protein: Enhanced In Vitro and Decreased In Vivo Activity ROGER BOUILLON, KATRIEN ALLEWAERT, DA ZHEN XIANG, BIAUW KENG TAN, and HUGO VAN BAELEN
ABSTRACT The affinity of la,25-dihydroxyvitamin D, [ la,25-(0H),D3] and analogs with side-chain modifications [MC 903 or calcipotriol, MC 1147 or 24,24-dihomo-1a,25-(OH),D3and 1,25-(OH),-16ene-23yne-D3] for the vitamin D receptor and the serum vitamin D binding protein (DBP) were compared. The affinity of MC 903 for the receptor from chick and rat duodenum or from human peripheral blood mononuclear cells or HL-60 cells varied between 60 and 100% relative to the affinity of 1,25-(OH),D3. The relative affinity of 1,25(OH)2-16ene-23yne-D, and MC 1147 varied for the same receptors between 45-70 and 3.5-25'4'0, respectively. The relative affinity of MC 903 for human DBP was 30-fold decreased, whereas the two other analogs did not bind to DBP at all even in more than 1000-fold excess. The in vitro biologic activity of la,25(OH),D, on phytohemagglutinin-stimulated normal human lymphocyte proliferation was markedly inhibited by the addition of physiologic amounts of DBP to the cell culture medium. No such inhibition was observed when MC 903 or 1147 was evaluated similarly. DBP therefore reversed the rank order of the in vitro potency of these analogs. Intramuscular injections for 10 consecutive days to vitamin D-deficient chicks demonstrated a 1100-fold lower biologic activity of MC 903, MC 1147, and 1,25-(OH),-16ene-23yne-D3compared to that of la,25-(OH),D3 as evaluated by serum calcium and osteocalcin concentrations, as well as by duodenal calbindin D28K and bone calcium content. We conclude that the biologic activity of vitamin D metabolites and analogs depends on their affinity for the vitamin D receptor as well as their affinity for DBP. Analogs with a low DBP but good receptor binding properties display low in vivo biologic activity on calcium and bone homeostasis, at least partly due to altered pharmacokinetics.
INTRODUCTION D HORMONE, 1,25-dihydroxyvitamin D [ 1,25-(OH),D,], not only regulates extracellular calcium metabolism but also modifies the secretion of several hormones and has profound influences on the immune system and cell differentiation.(1-5)Several analogs of 1,25(OH),D, with modification in the side chain(6-*1) have potent in vitro activity on normal or tumoral lymphocytes, whereas their in vivo activity on calcium metabolism is markedly decreased. The reason for this discrepancy is not fully understood, and differences in receptor binding, cellular access, and metabolism or differences in pharmacokinetics have all been suggested.
T
HE ACTIVE VITAMIN
Since the binding of vitamin D metabolites and analogs to the serum vitamin D binding protein (DBP) has a marked influence on their intracellular access to the vitamin D receptor and, therefore, biologic activity,('z-'4)we evaluated the in vitro and in vivo activity of analogs with side modifications and conclude that 1,25-(OH),D3 analogs with low DBP and good receptor binding properties display relatively high activity on lymphocyte proliferation in vitro but low biologic activity on calcium metabolism in vivo.
MATERIALS AND METHODS
Materials MC 903, calcipotriol, or 20(R)-[3'(S)-CycIopropyl-3'-hy-
Laboratorium voor Experimentele Geneeskunde en Endocrinologie, Katholieke Universiteit Leuven, Leuven, Belgium.
105 1
BOUILLON ET AL.
1052
droxyprop- l'(E)-enyll- 1 (S),3(R)-dihydroxy-9,lO-secopregna-S(Z),7(E), 10(19)-triene, and MC 1147, or 24,24-dihomo-I ,25-(OH),D3, were gifts from L. Binderup, Leo Pharmaceutical Products (Ballerup, Denmark); 1,25(OH),D, and 1,25-(OH),-16ene-23yne-D3were gifts from M. Uskokovic, Hoffman-La Roche (Nutley, NJ). [,HI1,25-(OH),D, (specific activity of 170 Ci/mmol) and [methyl-'Hlthymidine (2 Ci/mmol) were purchased from Amersham (Buckinghamshire, England). Aprotinine was obtained from Novo Industri (Bagsvaerd, Denmark). Human DBP and DBP-free serum were prepared by affinity chromatography as described previously. ('zl')
Methods Affinity for Vitamin D Receptor: Duodenal mucosa from normal Wistar rats or young chicks was sonicated in 4 vol of buffer (0.05 M Tris-HC1, 0.5 M KCI, 5 mM dithiotreitol, 10 mM Na,MoO,, and 1.5 mM EDTA, pH 7.5). Cytosol from HL-60 cells and phytohemagglutinin (PHA)-stimulated peripheral blood mononuclear cells (PBMC) were prepared similarly. The homogenate was centrifuged at 6OOO x g for 10 minutes and 30,000 x g for 1 h; the cytosol was stirred with ammonium sulfate (18% wt/vol) for 1 h and centrifuged; the pellet was then redissolved in the original buffer and used for binding studies. [,HI 1,25-(OH),D, (0.2 nM) and increasing concentrations of 1.25-(OH),D3 or its analogs were incubated with the duodenal cytosol in a final volume of 0.3 ml for 20 h at 4°C. Phase separation was then obtained by the addition of cold dextran-coated charcoal. Affinity for Human DBP: Binding of vitamin D metabolites and analogs to DBP was performed at 4°C essentially as described previously."') [)HI 1,25-(OH),D, and 1,25-(OH),D, or its analogs were added in 5 pl ethanol into glass tubes and incubated with DBP (0.18 pM) in a final volume of 1 ml (0.01 M Tris-HCI buffer and 0.154 M NaCI, pH 7.4) for 3 h at 4°C. Phase separation was then obtained by the addition of cold dextran-coated charcoal. Radioimmunoassay (RIA) for Chick Osteocalcin and Calbindin D28fi. Osteocalcin was purified from chick long bones essentially as described('s1and specific antibodies were raised in a rabbit. The RIA for chick osteocalcin used this rabbit antiserum (1:300,000 in 0.05 M phosphate buffer, pH 7.4, 1% BSA, 25 mM EDTA, and 0.125 M NaCI), [ l ~ S I ] ~ ~ t e ~ c a(iodination lcin by chloramine-T method, 15,000 cpm), and 100 p1 serum (diluted 200-fold) or standard in a total volume of 0.8 ml for 24 h at 4°C. Phase separation was then obtained by second antibody precipitation. Calbindin D28K was measured by RIA using a similar procedure with an antiserum against chick calbindin D28K (kindly provided by Norman) raised in a guinea pig (diluted 1:400,000) and ['ZSI]calbindinD28K, ionidated by the chloramine-T method. Homogenates of chick intestinal mucosa, obtained from 1,25-(OH),D,-treated normal chicks and calibrated versus purified calbindin D28K, were used as internal standard. The between-assay coefficient of variation (CV) was below 7%.
Effect of Vitamin D Metabolites and Analogs on Lymphocyte Proliferation: The human promyeolocytic leukemia cell line (HL-60) was obtained from the American Type Culture Collection (Rockville, MD). HL-60 cells were seeded at 1.2 x los cells per ml, and 1.25-(OH),D3 or its analogs were added in ethanol (final concentration below 0.2%) in RPMI-1640 medium supplemented with 10% heat-inactivated fetal calf serum (GIBCO, Roskilde, Denmark), 100 U/ml of penicillin. and 100 p g h l of streptomycin (Boehringer, Mannheim, Germany). After 4 days of culture in a humidified atmosphere of 5% CO, in air at 37"C, the dishes were shaken to loosen any adherent cells, and all cells were then assayed for maturation by nitroblue tetrazolium (NBT) reduction assay.(16)HL-60 cells at 1.5 x lo6 cells per ml were therefore mixed with an equal volume of a freshly prepared solution of phorbol 12-myristate 13-acetate (200 ng/ml; Sigma, St. Louis, MO) and NBT (2 mg/ml; Sigma) and incubated for 30 minutes at 37°C. The percentage of cells containing black formazan deposits was determined using a hemacytometer. Peripheral blood mixed mononuclear cells (PBMC) were prepared from buffy coats of human blood (obtained from the Belgian Red Cross) by centrifugation on Ficoll-Hypaque (Pharmacia, Uppsala, Sweden). The culture conditions were except that the identical to those previously de~cribed,"~) culture medium in the present study was RPMI-1640. Vitamin D-Deficient Chicks: Vitamin D-deficient chicks were reared on a vitamin D-deficient diet (1 Qo calcium and 1 Yo phosphorus; Hope Farms, Woerden, The Netherlands) and housed in a windowless room. After 3 weeks they received a daily intramuscular injection for 10 days with la,25-(OH),D3 or its analogs (solubilized in 10% ethanol, 10% Tween 80, and 80% NaCI, 0.9% in water). Chicks were then weighed and serum collected for measurements of calcium (by atomic absorptiometry) and osteocalcin (by RIA). The right tibia was removed and prepared free from adherent tissue, dried at 100°C for 24 h, and ashed at 600°C for 24 h in a muffle furnace. The ash of each bone was then weighed and dissolved in HCI (12 N) for measurement of calcium content. The duodenum was immediately excised, and the mucosa was scraped from the underlying muscle layer at 0°C. A homogenate was prepared in Tris Buffer (14 mM Tris HCI, pH 7.4, 0.125 M NaCI, 3 mM KCI, and 0.01 U/ml of aprotinine), centrifuged at 4°C at 10,000 x g for 1 h, and then assayed for calbindin D28K. Statistical analysis was performed using Student's t-tests.
RESULTS
Binding studies The affinity of 1,25-(OH),D, for the vitamin D receptor from rat duodenal mucosa (3.2 f 1.2 x lo9 M-', mean f SD; n = 6) was significantly (p < 0.01) lower than for chick duodenal mucosa (1.2 f 0.3 x I O ' O M - l ) , HL-60 cells (1.9 + 0.1 x 10" M-I), or PBMC (9.1 f 3.2 x lo9 M-I). The affinity of MC 903 was nearly identical to that of 1,25(OH)zD3 in HL-60 cells and chick duodenum and only slightly lower in rat duodenum or PBMC (Fig. 1). The relative affinity of MC 1147 varied between 3.5% (rat duode-
1053
VITAMIN D ANALOGS WITH LOW DBP AFFINITY
B
\
q.
~~
I -
1
1 L
1o - ~ 10-8 1 25- (OH12D3 OR ANALOG CONCENTRATION
10.’
ro-’O
(MI
I
10-’0
.
I
10-Q
10-0
1 25- (OH) 2D3 DA ANALOG CONCENTRATION
10
’
(MI
FIG. 1. Binding of [’H]1,25-(OH),D, in the presence or absence of nonradioactive 1,25-(OH),D, or its synthetic analogs to the vitamin D receptor from (A) rat duodenal mucosa; (B) chick duodenal mucosa; (C) HL-60 cells; and (D) phytohemagglutinin-stimulated human peripheral blood mononuclear cells. All values given as mean f standard deviation (SD; n = 6 in three experiments). ( 0 ) 1,25-(OH),D,; (H) MC 903 or calcipotriol; (A) MC 1147 or 24,24-dihomo-1,251 ,25-(0H),-26ene-23yne-D3. (OH),D,; (0)
num) and 25% (PBMC) compared to the affinity of 1,25OH),D,, whereas the relative affinity of 1,25-(OH),-16ene23yne-D3 varied between 45 and 70% (Fig. 1). The affinity of 1,25-(OH),D, for purified human DBP was 1.5 f 0.2 x 10’ M-’ (pH 7.4, 4°C; n = 6). The binding of the analogs with side-chain modifications was decreased: to obtain a 50% displacement a 30-fold molar excess of MC 903 was needed in comparison with 1,25(OH)& whereas MC 1 147 and 1,25-(0H),-16ene-23yneD, did not displace radioactive 1,25-(OH),D, even at a 1000-fold molar excess (Fig. 2).
Effect of vitamin D analogs on lymphocyte function in vitro The effect of vitamin D analogs on the proliferation of normal human lymphocytes stimulated by P H A was dependent on the presence of DBP during the culture period. A physiologic concentration of DBP (300 pg/ml or 5.8 pM) increased the concentration of 1,25-(OH),D, necessary for 50% of proliferation inhibition from 1 x to
5 x M (Fig. 3). In the absence of DBP, MC 903 was nearly equally active as 1,25-(OH),D, whereas MC 1147 was about 10-fold less potent. The addition of DBP did not markedly influence the activity of MC 903 and MC 1147 but decreased the effects of the natural hormone. In the presence of normal DBP concentrations, MC 903 therefore became the most active metabolite, followed by MC 1147, 1,25-(OH),D, being the least active (Fig. 3). The differentiating effects of the vitamin D hormone and its analogs were evaluated by NBT reduction in HL-60 cells in the presence of fetal calf serum (10%). MC 1147 was slightly more effective than 1,25-(OH),D, or MC 903, whereas 1 ,25-(OH),-16ene-23yne-D3 was markedly more effective (Fig. 4).
In vivo effects on calcium metabolism The daily intramuscular injection of 1,25-(OH),D, and its analogs increased but did not normalize the body weight of vitamin D-deficient chicks (data not shown) compared to animals raised on a D-replete diet. All vita-
BOUILLON ET AL.
1054
The duodenal calbindin D28K concentration was nearly undetectable in D-deficient chicks and increased to supraphysiologic amounts after a daily dose of even 20 ng of 1,25-(OH),D3. Compared to 1,25-(OH),D,, a more than 100-fold higher amount of MC 903 or MC 1147 was needed to increase the duodenal calbindin D28K concentration (Fig. 5 ) . The serum concentration of osteocalcin, measured by specific RIA, also increased above the concentration in Drepleted chicks after injection of 1,25-(OH),D, in an amount that still did not normalize serum calcium. Intramuscular injections of MC 903 and MC 1147 were again much less active (at least 40-fold) than 1,25-(OH),D, on osteoblastic osteocalcin secretion. Serum osteocalcin and duodenal calbindin D Z ~ concentrations K hardly increased after a 10 day in vivo administration of 1,25-(OH),-16ene23yne-D,, even when given in 100-fold excess compared to 1,25-(OH),D, (Fig. 5 ) .
150,
T 5
T
tK
Y
\ Y
1 25
I
1
10
10
(OH)Z03
-
10
10-6
O R A N A L O G C O N C E N T R A T I O N (MI
FIG. 2. Binding of [3H]1,25-(OH),D3 to purified human vitamin D binding protein in the presence or absence of nonradioactive 1,25-(0H),D3 or analogs. Same symbols as in Fig. 1.
-. 1
~~
10
10
10
lo
10
10
8
10
’
VITAMIN D H E l A B O L l l E OR A N A L O G IM1
FIG. 3. Effect of 1,25-(OH),D, or its analogs on the proliferation of PHA-stimulated mixed peripheral human blood lymphocytes as measured by [,H]thymidine incorporation. All measurements were performed in 6-fold microcultures in human DBP-free serum reconstituted (5.8 pM DBP, open symbols) or not (closed symbols) with human 1,25-(OH),D, (with and without DBP, reDBP. (O,.) MC 903 (with and without DBP, respectively); (El,.) spectively); (&A) MC 1147 (with and without DBP, respectively).
min D analogs were markedly less potent than the natural hormone. The serum calcium concentration and bone calcium content increased markedly by vitamin D therapy, but 1,25-(OH),D, was 30- to 100-fold more efficient than MC 903 and MC 1147 (Fig. 5 ) .
DISCUSSION The active vitamin D hormone has an important function not only in calcium metabolism but also in lymphocyte proliferation and differentiation. Several vitamin D analogs are more potent when tested in vitro in cell proliferation or differentiation than in systemic mineral homeO S ~ ~ S ~ S . ( ~ .Moreover, ~ . ~ . ~ ~some . ~ ~ of , ~these ~ ~ analogs were more active in cell differentiation in vitro than could be expected from their relative affinity for the vitamin D receptor.‘16.17) Such an effect may be due to target-specific differences in intrinsic receptor binding or activation. The biologic activity of vitamin D analogs, however, also depends on cellular accessibility (influenced by affinity for DBP) and metabolic clearance (probably also modified by affinity for DBP) or intracellular activation or inactivation in target cells. Differences in binding to the serum vitamin D binding protein could therefore at least partly explain differences in pharmacokinetics. Indeed, we previously demonstrated by in vitro and in vivo experiments that the free or unbound vitamin D metabolites represent the active hormone concentration. ( 1 * , 1 4 , ’ 8 ) Moreover, the addition of physiologic amounts of DBP decreased the biologic efficacy of natural vitamin D metabolites on human lymphocyte proliferation. ( I 3 ) The present data demonstrate that several vitamin D analogs with modifications in the side chain have markedly (30- to more than 200-fold) decreased DBP binding properties whereas the receptor affinity either remained largely unchanged (MC 903) or was mildly decreased [MC 1147 and 1,25-(OH),-16ene-23yne-D3], compared to the decrease in their affinity for DBP (Fig. 2). Other modifications in the side chain also decreased the affinity for DBP as described for 22-oxa-1,25-(OH),D3,‘19)26-homo-1,25(OH),D, ( p e r s o n a l o b s e r v a t i o n ) , a n d 1,25(OH),-26,27-dimethyl-D,.’” Recent data presented after initial submission of this manuscript using diluted serum and [’H]25-OHD3 instead of purified DBP and [,H]1,25(OH)2D, largely confirmed our data.(20)The affinity of MC 1147 for rat duodenal receptor and of 1,25-(OH),-
1055
VITAMIN D ANALOGS WITH LOW DBP AFFINITY
1011,
16ene-23yne-D, for the chick and human receptor agreed well with previous In experiments without extracellular DBP (serum-free or DBP-free cultures of isolated cells of tissues), MC 903 and MC 1147 displayed an activity closely resembling their affinity (Fig. 3). Previous experiments with other cell types also demonstrated that MC 903 was nearly equipotent to 1,25-(OH),D, when tested in medium with little or no DBP or s e ~ u m . ( ~ . In ~ ' -the ~ ~presence ) of DBP, however, 1,25(OH),D, became markedly less active (30-fold) as a result of decreased free hormone concentration. Because of low DBP binding the analogs remained totally accessible for cellular entry and activation. Addition of DBP therefore did not decrease the biologic activity of MC 903 or MC 1147 on human lymphocyte proliferation (Fig. 3). Similarly, addition of DBP to DBP-free fetal calf serum did not influence the biologic activity of 1,25-(OH),-16ene-23yneD, on HL-60 cells in vitro (unpublished personal results). The low in vivo effects of the vitamin D analogs with low DBP binding can also at least partly be explained by
? T
FIG. 4. Differentiating effect of 1,25-(OH),D, and its analogs on HL-60 cells, grown in RPMI-1640 medium supplemented with 10% FCS, as evaluated by NBT reduction. Same symbols as in Fig. 1.
r
I
I
-
100 W
0
Z m
--a -
0
.-
50 $
P r0
H
q
- e 3.
m w
I
10-11
10-l0
o-~
10-l1
1
la. 25- (OH) 203 OR ANALOG (rnol/day)
Io-g
10-lO
la, 25- (OH)2D3 OR ANALOG (rnol/day)
X
T
a
iF---++ 90-11 la,
10-l0
25- (OH) D, OR ANALOG (mol /day)
1o-g la,
10-11 25- (OH),D,
t0-l0 1 OR ANALOG (mol/day)
FIG. 5. Effects of 1,25-(OH),D, and its analogs on serum calcium concentration (top left), tibia1 calcium content (top right), serum osteocalcin (bottom right), and duodenal calbindin DZsKconcentration (bottom left) after 10 days of IM (intramuscular) injection of l,25-(OH),D3 or its analogs. All data are given as mean 1 SEM (standard error of the mean) using between 6 and 10 animals per group. ( 0 )1,25-(OH),D,; (m) MC 903; (A) MC 1147 or 24,24-dihomo-1,25(OH),D,; (0) 1 ,25-(OH),-16ene-23yne-D,; (*) data obtained in vitamin D,-replete chicks.
*
BOUILLON ET AL.
1056
their differences in affinity for DBP and the vitamin D receptor. The vitamin D analogs with low DBP binding remain mainly in the free form after administration and are therefore subject to more rapid degradation. Indeed, the half-life of MC 903 was much shorter than that of 1,25(OH),D, in both Sprague-Dawley rats (12 minutes) and minipigs [60 minutes after IV (intravenous) injection of MC 903; H. Sorensen, Leo Pharmaceutical Products, personal communication]. This rapid degradation can largely explain why serum and bone calcium concentration or content increased in vitamin D-deficient chick only after a nearly 100-fold higher dose of analogs compared to the natural hormone. Similar observations were made for more specific vitamin D-dependent proteins (duodenal calbindin D Z ~and K serum osteocalcin concentration). An even lower in vivo calcemic effect was previously in rachitic observed for 24,24-dihomo-1,25-(OH),D3 rats,(111whereas no bone-resorbing activity could be demonstrated in rat bone cultures in v i t r ~ . ( ’ ~ ”In~ Perlman’s study, however, the rat intestinal calcium absorption could with an also be stimulated by 24,24-dihomo-1,25-(OH),D3, approximately 10-fold lower potency than 1,25-(OH),D,. In contrast to the in vitro data with rat MC 1147 was equipotent on bone (serum osteocalcin) compared to the duodenum (calbindin D28~)in rachitic chicks. The discrepancy may be due to species differences in the relative affinity of MC 1147 for the chick duodenal receptor [15% of that of 1,25-(OH),D,] was greater than observed in the rat [3.5% of the affinity of 1,25-(OH),D3] (Fig. 1). Moreover, our in vivo data used the measurement of two vitamin D-dependent proteins that allowed the detection of a much greater difference between D-deficient and vitamin D-treated animals than simple serum calcium or calcium absorption tests (Fig. 5 ) . The low in vivo calcemic effects of l ,25-(OH),-16ene-23yne-D, confirm previous data after its acute administration in rachitic chicks(9) or Other vitamin D analogs with modifications in the C-ring (C1J behave quite differently as such analogs display increased DBP affinity with maintained or decreased receptor affinity. The addition of DBP therefore decreases the biologic availability and activity of such analogs in vitro.(zs) In conclusion, several vitamin D analogs with side-chain modification have a low affinity for DBP but (nearly) normal affinity for the vitamin D receptor. In the presence of DBP in vitro such analogs have therefore better access and increased activity but, when given in vivo, have low potency on bone and duodenum as a result of rapid degradation. Such differences in pharmacokinetics may find interesting applications, however, because short-term exposure to cells with different cell cycles may induce a totally different activity pattern.
No. 3.0044.89) is kindly acknowledged. Presented in part at the joint meeting of the ASBMR and ICCRH, Montreal 1989, Abstract 734.
REFERENCES 1. Manolagas SC, Provvedini DM, Tsoukas CD 1985 Interac-
2.
3. 4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
ACKNOWLEDGMENTS
15.
The financial support of the Belgian Nationaal Fonds voor Geneeskundig Wetenschappelij k Onderzoek (Grant
16.
tions of 1.25-dihydroxyvitamin D, and the immune system. Mol Cell Endocrinol 43:113-122. Suda T 1989 The role of la,25-dihydroxyvitarnin D, in the myeloid cell differentiation. Proc SOCExp Biol Med 192:214220. Bouillon R, Van Baelen H 1989 The biological roles of vitamin D. Saudi Med J 10:260-266. Reichel H, Koeffler HP, Norman AW 1989 The role of the vitamin D endocrine system in health and disease. N Engl J Med 320:980-991. DeLuca HF 1989 The vitamin D story: A collaborative effort of basic science and clinical medicine. Proc SOCExp Biol Med 192:224-236. Binderup L, Bramm E 1988 Effects of a novel vitamin D analogue MC 903 on cell proliferation and differentiation in vitro and on calcium metabolism in vivo. Biochem PharmaCOI 37~889-895. Paulson SK, Perlamn K, DeLuca HF, Stern PH 1990 24- and 26-homo-1,25-dihydroxyvitaminD, analogs: Potencies on in vitro bone resorption differ from those reported for cell differentiation. J Bone Miner Res 5201-206. Brown AJ, Ritter CR, Finch JL, Morrissey J, Martin KJ, Murayama E, Nishii Y, Slatopolsky E 1989 The noncalcemic analogue of vitamin D, 22-oxacalcitriol, suppresses parathyroid hormone synthesis and secretion. J Clin Invest 84:728732. Zhou J-Y, Norman AW, Lubbert M, Collins ED, Uskokovic MR, Koeffler HP 1989 Novel vitamin D analogs that modulate leukemic cell growth and differentiation with little effect on either intestinal calcium absorption or bone calcium mobilization., Blood 74:82-93. Gill HS, Londowski JM, Corradino RA, Zinsmeister AR, Kumar R 1988 The synthesis and biological activity of 25-hydroxy-26,27-dimethylvitaminD, and 1,25-dihydroxy-26,27dimethylvitamin D,: Highly potent novel analogs of vitamin D,. J Steroid Biochem 31:147-160. Perlman K, Kutner A, Prahl J, Smith C, Inaba M, Schnoes HK, DeLuca HF 1990 24-Homologated 1,25-dihydroxyvitamin D, compounds: Separation of calcium and cell differentiation activities. Biochemistry 29:190-196. Bouillon R, Nyomba B, Decleer W, Van Baelen H 1985 Vitamin D-binding protein and free l ,25-dihydroxyvitamin D in animals and clinical medicine. In: Norman AW, Schaefer K, Grigoleit HG, von Herrath D (eds.) Vitamin D, Chemical, Biochemical and Clinical Update. Walter de Gruyter, Berlin, pp. 669-672. Vanham G, Ban Baelen H, Tan BK, Bouillon R 1988 The effect of vitamin D analogs and of vitamin D-binding protein on lymphocyte proliferation. J Steroid Biochem 29:381-386. Bikle DD, Gee E 1989 Free, and not total, 1.25-dihydroxyvitamin D regulates 25-hydroxyvitamin D metabolism by keratinocytes. Endocrinology 124:649-654. Gundberg CM, Hauschka PV, Lian JB, Gallop PM 1984 Osteocalcin: Isolation, characterization and detection. Methods Enzymol 107516-544. Ostrem VK, Tanaka Y, Prahl J, DeLuca HF, Ikekawa N
VITAMIN D ANALOGS WITH LOW DBP AFFINITY
17.
18.
19.
20.
21.
22.
23.
1987 24- And 26-homo-I ,25-dihydroxyvitamin D,: Preferential activity in inducing differentiation of human leukemia cells HL-60 in vitro. Proc Natl Acad Sci USA 84:2610-2614. Abe J , Morikawa M, Miyamoto K, Kaiho S-I, Fukushima M, Miyaura C, Abe E, Suda T, Nishii Y 1987 Synthetic analogues of vitamin D, with an oxygen atom in the side chain skeleton. FEBS Lett 22658-62. Vargas S, Bouillon R, Van Baelen H, Raisz LG 1990 Effects of vitamin D-binding protein on bone resorption, stimulated by 1,25 dihydroxyvitamin D,. Calcif Tissue Int 47:164168. Okano T, Tsugawa N, Masuda S, Takeuchi A, Kobayashi T, Nishii Y 1989 Protein-binding properties of 22-oxa-la,25dihydroxyvitamin D,, a synthetic analogue of Ia,25-dihydroxyvitamin D,. J Nutr Sci Vitaminol 35529-533. Dusso AS, Lopez-Hilker S, Finch J , Mori T , Nishii Y , Slatopolsky E, Brown AJ 1990 On the mechanisms for the selective action of vitamin D analogs (abstract). J Bone Miner Res J(Suppl. 2):S202. Trydal T , Lillehaug JR, Aksnes L, Aarskog D 1989 The effect of the 1,25-(OH),D, analog, MC 903, o n non-transformed and transformed mouse embryo fibroblast cells (abstract). J Bone Miner Res 4(Suppl. 1):943. Evans DB, Thavarajah M, Binderup L, Kanis J A 1989 Comparison of the actions of MC 903 and 1,25-(OH),D, on human osteoblast-like cells in vitro (abstract). J Bone Miner Res 4(Suppl. 1):868. Valaja T, Mahonen A, Priskanen A, Maenpaa P H 1989 Af-
1057 finity of a novel vitamin D analog MC 903 for vitamin D receptor and effects of the analog on the synthesis of osteocalcin by human osteosarcoma cells (abstract). J Bone Miner Res 4(Suppl. 1):732. 24. Holick MF, Uskokovic M, Persons K, Horst R, Bagglioni EG, Truitt G A 1988 An evaluation of analogs of 1,25-dihydroxyvitamin D3 on the inhibition of proliferation and induction of terminal differentiation in cultured human normal keratinocytes. In: Norman AW, Schaefer K, Grigoliet HG, von Herrath D (eds.) Vitamin D, Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 363-371. 25. Bouillon R, Allewaert K, Xiang DZ, Vandewalle M. De Clercq P 1989 Structure-function analysis of vitamin D analogs: Difference in binding to the vitamin D receptor and vitamin D-binding serum protein (abstract). J Bone Miner Res 4(Suppl. 1):734.
Address reprint requests to: Dr. R . Bouillon Legendo Onderwijs en Navorsing Gas(huisberg B-3000 Leuven, Belgium Received for publication August 13, 1990; in revised form April 15, 1991; accepted April 17, 1991.
