- Email: [email protected]
Synthesis of 25-hydroxy-[26,27-3H]vitamin D2, 1,25-dihydroxy-[26,27-3H]vitamin D2 and their (24R)-epimers
ANALYTICALBIOCHEMISTRY
161, 96-102
(1987)
Synthesis of 25-Hydroxy-[26,27-3H]vitamin D2, 1 ,25-Dihydroxy-[26,27-3H]vitamin D2 and Their (24R)-Epimers’ RAFAL R. SICINSKJ~YOKOTANAKA,~
MARYPHELPS,HEINRICHK.SCHNOES, AND HECTOR F. DELUCA
Department of Biochemistry, University of Wisconsin, College ofAgricultural 420 Henry Mall, Madison, Wisconsin 53706
and L&e Sciences,
Received August 11, 1986 Synthesis of a C-24-epimeric mixture of 25-hydroxy-[26,27-3H]vitamin D2 and a C-24-epimeric mixture of 1,25-dihydroxy-[26,27-‘HIvitamin D2 by the Grignard reaction of the corresponding 25-keto-27-nor-vitamin D2 and lo-acetoxy-25keto-27-nor-vitamin D3 with tritiated methyl magnesium bromide is described. Separation of epimers by high-performance liquid chromatography afforded pure radiolabeled vitamins of high specific activity (80 Ci/mmol). The identities and radiochemical purities of 25-hydroxy-[26,27-3H[vitamin D2 and 1,25-dihydroxy-[26,27-3H]vitamin D2 were established by cochromatography with synthetic 25hydroxyvitamin Dz or 1,25-dihydroxyvitamin D 2. Biological activity of 25-hydroxy[26,27-3H]vitamin D2 was demonstrated by its binding to the rat plasma binding protein for vitamin D compounds, and by its in vitro conversion to l,25-dihydroxy-[26,27-3H]vitamin D2 by kidney homogenate prepared from vitamin D-deficient chickens. The biological activity of 1,25-dihydroxy-[26,27-3H]vitamin D2 was demonstrated by its binding to the chick intestinal receptor for 1,25-dihydroxyvitamin D3. 0 1987 Academic FWSS. hc. KEY WORDS: vitamin D; calcium metabolism; intestine; bone; calcium regulating hormones; phosphorus metabolism.
Metabolism of vitamin D3 has received considerable attention during the past two decades. Substantial advances in our understanding of vitamin D3 metabolic pathways and its mode of action have been closely connected with the development of synthetic methods providing radiolabeled vitamin D3 derivatives ( 1,2). The availability of these radioactive compounds of still higher specific activity is currently of considerable interest because they serve as convenient tools for a
number of detailed biomedical experiments. During the extensive investigations on the vitamin D3 compounds, parallel work on vitamin D2 metabolism was restricted because of a lack of available labeled compounds in spite of the common clinical use of vitamin Dz . The data available so far indicate that the main metabolic activation of vitamins D2 and D3 in mammals and birds is identical (1,2). Although 25-hydroxyvitamin D2 (25 OH-D2)4 and 1,25-dihydroxyvitamin D2 ( 1,25-(OH)2D2) were isolated and chemically identified in 1969 and 1974, respectively (3,4), they have not been tested extensively and their biological significance remains unclear. Jones et al. (5) reported the isolation
’ This work was supported by Grant AM-32701 and Program Project Grant AM-14881 from the National Institutes of Health, and by the Harry Steenbock Research Fund of the Wisconsin Alumni Research Foundation. ’ Present address: Department of Chemistry, University of Warsaw, u 1, Pasteura 1,02-093 Warsaw, Poland. 3 Present address: VA Medical Center, 113 Holland Avenue, Albany, NY 12208. 0003-2697187
$3.00
Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.
4 Abbreviations used: 1,25-(OHhD3, 1,25-dihydroxyvitamin D3; 25-OH-Dr , 25-hydroxyvitamin D3 ; CT3MgBr, tritiated methyl magnesium bromide. 96
SYNTHESIS
OF
VITAMIN
of 24-OH-D2 and (24R),25-(OH)2D2, but no information on their activity is available so far. Random labeled vitamin [3H]D2 (8.6 mCi/mmol) was obtained by Suda et al, (3) by exposing vitamin D2 to tritiated acetic acid, whereas vitamin [3a-3H]D2 (1.2 Ci/ mmol) was synthesized by Jones et al. (4) using sodium borotritide reduction of 3/3acetoxyergosta-3,5,7,22-tetraene. The corresponding ~~-OH-[~CX-~H]D~ was generated in vivo in small amounts in rats by the administration of vitamin [3a-3H]D2. We, therefore, turned our attention to a chemical synthesis of radioactive 25-OH-D2 of high specific activity. The first unpublished synthesis of 25OH-D2 has been achieved at the Upjohn Co., but this method practically excluded the possibility of radiolabeling at the side chain because its construction was performed at a very early step in the process (1). A very recent synthesis of 25-OH-D2 and its 24epimer, which has been carried out in our laboratory by Morzycki et al. (6), provided a direct precursor for both unlabeled and labeled compounds. Similarly, a convenient synthesis of 1,25-(OH)2D2 has been devised that also provides a direct precursor (7). Using these approaches, we obtained (24S)and (24R)-25-OH vitamin [26,27-3H]D2 (compounds 2a and 3a, Fig. 1) and (24S)and (24R)- 1,25-(OH)~-[26,27-~H]D~ (compounds 2b and 3b) of high specific activity (80 Ci/mmol). MATERIALS
AND METHODS
Radioactive determinations were performed with a Model 3255 Packard Tri-Carb liquid scintillation counter fitted with an automatic external standardization system (Packard Instruments, Downers Grove, IL). Radioactive samples were counted in a 3a70B counting solution (Research Products Int. Corp., Elk Grove, IL). Ultraviolet spectra were recorded in ethanolic solutions with a Beckman DB-G recording spectrophotometer (Beckman Instruments, Lincolnwood, IL).
D2 COMPOUNDS
97
Vitamin D compounds. 25-OH-D3 was a gift from the Upjohn Company (Kalamazoo, MI), while 1,25-(OH)2D3 was a gift from the Hoffmann-LaRoche Company (Nutley, NJ). 25OH-D2 and its 24-epimer were synthesized using the method of Morzycki et al. (6) while 1,25-(OH)*D* and its epimer were synthesized using the method of Sicinski et al. (7). 25-OH-[26,27-3H]D3 and 1,25-(OH)2-[26,27-3H]D3 were synthesized using previously described methods (8,9). Animals. One-day-old white Leghorn chicks were purchased from Northern Hatcheries (Beaver Dam, WI) and fed a vitamin D-deficient diet (10) for 4 weeks for a source of renal 25-OH-D3- 1-hydroxylase and chick intestinal 1,25-(OHhD3 receptor. In vitro production of 1,25-(OH)2-[26,27‘HID* and 24-epi-I,25(OH)2-[26,27-‘HID*. Kidneys from vitamin D-deficient chickens were pooled and a 20% (w/v) homogenate was prepared in an ice-cold 15 mM Tris-acetate buffer (pH 7.4 at room temperature) containing 0.19 M sucrose and 1.9 mM magnesium acetate. The incubation was carried out in a 125-ml Erlenmeyer flask containing 1.2 g of kidney tissue, 0.19 M sucrose, 15 mM Tris-acetate (pH 7.4 at room temperature), 1.9 mM magnesium acetate, and 25 mM succinate in a 9-ml final volume as described by Tanaka and DeLuca (11). The reaction was initiated by addition of 120 PCi ( 1.5 nmol) of either 25-OH-[26,27-3H]D2 or 24-epi-25OH-[26,27-3H]D2 dissolved in 60 ~1 of 95% ethanol. The reaction mixtures were incubated at 37°C with shaking at 100 oscillations/min. The reaction was terminated by the addition of 18 ml of MeOH and 9 ml of dichloromethane. Extraction was carried out by the method described by Tanaka and DeLuca (11). The extract was prepurified on a small column of Sephadex LH-20 as described earlier ( 11). The purified extract from a Sephadex LH-20 column was injected into a Waters Associates Model ALC/GPC 204 HPLC system (Milford, MA) equipped with a DuPont Zorbax-Sil column (4.6 mm X 25 cm) (Wilmington, DE) developed with a solvent system of 8% 2-propanol in hexane at a
98
SICINSKI ET AL.
HO . . . . c la, R=ti lb, R=OH
FIG.
HO . . . . p
2a, 2b,
1. Synthesis of 25-OH-[26,27-3H]D2,
R=H R=OH
3a, R=H 3b,
1,25-(OH)2-[26,27-3H]Dz,
pressure of 800 psi. A 254~nm ultraviolet monitor was used to detect authentic compounds or compounds produced in vitro, while fractions of 0.8 ml each were collected, and an aliquot of each fraction was used for detection of tritium. Displacement of 2.5-OH-[26,27-3H]D2 for rat plasma binding protein for vitamin D compounds by 25-OH-D* or 25-OH-Dj. Diluted rat plasma ( 1:5000) was incubated with 25-OH-[26,27-3H]D2 and graded amounts of unlabeled 25OH-D* or 25-OH-D3 dissolved in 50 ~1 of 95% ethanol at 4’C for 16 h, as described by Shepard et al. (12). Duplicate determinations for each concentration of either compound were carried out. Displacement of 1,25-(0H)~-[26,27-~H]D~ from chick intestinal cytosol receptor protein for 1,25-(OH)*D3. Chick intestinal cytosol was incubated with 1,25-(0H)2-[26,273H]D2 and various concentrations of unlabeled 1,25-(OH)*DZ or 1,25-(OH)2D3 dissolved in 50 ~1 of 95% ethanol at 4°C for 16 h as described by Shepard et al. ( 12). Triplicate determinations for each concentration were carried out. Synthesis of 25-OH-[26,27-‘HID2 compounds (2a and 3a). 25-Keto-27-nor-vitamin Dz (la) (Fig. 1) was obtained by the method of Morzycki et al. (6) as a mixture of 24epimers. A solution of la (1 mg) in 2 ml of dry ether was added dropwise with a syringe through a septum into a solution of 0.5 mmol of tritiated methyl magnesium bromide5 (CT3MgBr) in 0.5 ml of dry ether. The 5 The methylation step was carried out in the laboratories of New England Nuclear (Boston, MA). Workup of the reaction and subsequent experiments were performed in our laboratory.
R=OH
and their C-24-epimers.
mixture was maintained at 0°C for the period of addition, then stirred at room temperature for 2 h. The solvent was evaporated under vacuum, and labile tritium was removed by three successive evaporations under vacuum of 2 ml portions of methylene chloride-methanol (1: 1). The mixture was then frozen in a liquid nitrogen bath. After addition of 1 ml of saturated aqueous ammonium chloride, the mixture was allowed to warm to room temperature. It was then dried and labile tritium was removed as before. The crude mixture was stored temporarily in 10 ml of ethanol. It was partitioned between 60 ml of ether (previously washed with water) and 3 X 20 ml of water. The aqueous layers were combined and reextracted with ether. An assay showed that 95% of the radioactivity was in the main extract. The crude reaction product was redissolved in 1 ml of hexane:methanol:chloroform (9: 1: 1) and applied to a 2X 30-cm Sephadex LH-20 column, Chromatography of the radioactive compounds was carried out with the same solvent mixture at a flow rate of 1 ml/min and 9-ml fractions were collected. The fractions in the 25-OH-D2 region (fractions 1 1- 18) were pooled, and the solvent was removed under vacuum. This material was further purified by high-performance liquid chromatography on a DuPont 830 LC (DuPont Instruments, Wilmington, DE) equipped with a DuPont Zorbax-Sil column (4.6 mm X 25 cm) using 1% 2-propanol in hexane as the solvent (flow rate 2 ml/min, 1100 psi); the eluted compounds were monitored at 254 nm. The peaks of radioactive 25-hydroxyvitamin D compounds overlapped; the (24S)-epimer was collected at
SYNTHESIS
OF VITAMIN
Dz COMPOUNDS
99
Both C-24-epimers of 25-OH-[26,27‘HID2 with specific activities of 80 Ci/mmol were obtained by the Grignard reaction of 25-keto-27-nor-vitamin D2 with CT3MgBr.6 Separation of the epimers was achieved by
HPLC. Radioactive purity and identification of 25-OH-[26,27-3H]D2 and its C-2bepimer were provided by cochromatography of the radiolabeled compound and synthetic isomers of unlabeled 25-OH-D2. One microcurie of 25-OH-[26,27-3H]D2 was mixed with 50 pg of the unlabeled (24R)-epimer of 25-OH-D2 and 25 pg of unlabeled 25-OH-D* to give a final specific activity of 0.04 wCi/bg. The mixture was subjected to an HPLC on a Zorbax-Sil semipreparative column. As shown in Fig. 2, a baseline separation of the C-24-epimers of 25-OH-D2 was achieved when fractions of each isomer were recycled three times by placing the HPLC system in a recycle mode. The recovered 25-OH-D2 had not changed in specific activity and no trace of radioactive (24R)-epimer was found. The biological identity of 25-OH-[26,273H]D2 was established by its competitive binding activity for rat plasma vitamin D transport protein. It has been shown that the binding abilities of 25-OH-D2 and 25-OH-D3 to the rat plasma protein are identical (Tanaka and DeLuca, unpublished results). As shown in Fig. 3, displacement curves were obtained by competition with increasing concentration with unlabeled 25-OH-D2 for binding sites previously saturated with a fixed amount of 25-OH-[26,27-3H]D2. Displacement curves obtained by competition of unlabeled 25-OH-D2 or 25-OH-D, for binding sites saturated with 25-OH-[26,273H]D3 were identical to those in Fig. 3 (data not shown). Evidence for biological activity of 25-OH[26,27-3H]D2 was further established by its conversion to 1,25-(OH)2-[26,27-3H]D2 in vitro by vitamin D-deficient chick kidney homogenate. It has been demonstrated by Jones et al. (13) that an in vitro chick kidney mitochondrial system converts 25-OH-D2 to 1,25-(OH)2D2. As shown in Fig. 4A, a baseline separation of chemically synthesized isomers of 1,25-(OH)2D2 was achieved by HPLC with a Zorbax-Sil column using a sol-
6 The introduction of label by the Grignard reaction produces an asymmetric center on C-25 with the label entirely on one carbon. Since the introduction is not
stereospecific, the label will be found on both C-26 and C-27 and is so designated.
68.5-71.5 ml and the (24R)-compound was collected at 73.5-79 ml. Both fractions (about 90% purity) were rechromatographed separately by HPLC in the same system yielding pure stereoisomers. Both compounds 2a and 3a exhibited the expected ultraviolet absorption (X,,, 264.5 nm) and identical constant specific activity of 80 Ci/ mmol. The identities and radiochemical purities of 25-OH-[26,27-3H]D2 and its 24epimer were also confirmed by HPLC cochromatography with synthetic 25-OH-D2 epimers. Synthesis of 1,25-(OH)2-[26,27-‘H]DJ compounds. The syntheses of 1,25-(OH)2[26,27-3H]D2 and its 24-epimer were carried out as described for 25-OH-[26,27-3H]D3 above except that la-hydroxy-25-keto-27nor-vitamin D2 (lb) as l-acetate (7) was used as the starting material. The radiolabeled crude 1,25-(OH)2-[26,27-3H]D2-epimeric mixture was chromatographed on a precalibrated 1 X 60-cm Sephadex LH-20 column packed and developed in CHC13:hexane (65:35). Fractions containing a mixture of the two isomers were collected and further purified by HPLC using a 4.6 mm X 25 cm Zorbax-Sil column (DuPont, Wilmington, DE), with 9% 2-propanol in hexane as the solvent system. The combined peaks were collected and chromatographed again on Zorbax-Sil with a solvent of 7% 2-propanol in hexane. This provided baseline separation of the epimers which were collected separately (2b, 3b). On rechromatography, each gave a single symmetrical peak (not shown). Both the 24-epi-1,25-(OH)2-[26,27-3H]D2 and 1,25-(OH)2-[26,27-3H]D2 had specific activities of 78 Ci/mmol. RESULTS
100
SICINSKI ET AL.
I.0 24-epi-25-OH-D2 f 0.6 -
E c N s
.
0.6
”
25-OH-D, --
I’
I
0.4.
a-
s 8 0.2 -
D
Jc
; 50
/ 100
Elution
150
volume
200
250
300
(ml)
FIG. 2. Chromatography of 25-OH-[26,27-3H]D2 with authentic C-2Cepimers of 25-OH-D2 on HPLC. Chromatography was performed on a Zorbax-Sil semipreparative column (6.2 mm X 25 cm) developed with 2% 2-propanol in hexane at a pressure of 1500 psi and a flow rate of 4 ml/min. The solid line represents optical density at 254 nm. The effluent was recycled three times. Fractions of 1.6 ml each were collected between 270 and 320 ml, and an aliquot of each fraction was counted as shown by the broken line (inset). The final elution volume of unlabeled 25-OH-D2 was 303 ml, while that of unlabeled (24S)-25-OH-D2 was 305 ml, due to an isotope effect. Fractions containing 25-OH-D* were combined to measure the vitamin D triene by means of a uv spcctrophotometer to calculate specific activity.
vent system of 8% 2-propanol in hexane. A compound biologically generated from 25-OH-[26,27-3H]D2 was detected by monitoring radioactivity and uv absorption, and was eluted at the position of authentic
unlobelled
25-OH-D,
or 25-OH-D,
(pmol/tube)
FIG. 3. Displacement of25-OH-[26,27-‘HID2 from rat plasma vitamin D binding protein by unlabeled 25-OHD2 (0) or 25-OH-D, (0). Each point represents the mean value of triplicate determinations.
1,25-(OH)2D2 as shown in Figure 4B. The area of the peak detected by uv absorption was measured and the amount of the compound produced in vitro was calcuiated from a standard curve. This curve was constructed by plotting the area under the uv absorption peak versus the amount of authentic 1,25-(OH)*D* injected into HPLC as described by Tanaka et al. (11). It was found that 41 &i/220 ng of 1,25-(OH)zD2 was produced in vitro, giving a specific activity of 80 Ci/mmol. Similarly, the (24Qepimer of 1,25-(OH)zDz was produced from 24-epi-25OH-[26,27-3H]D2 as shown in Fig. 4C. Its specific activity was found to be identical to that of its substrate or 48 &i/225 ng. Direct chemical syntheses of 1,25-(OH)*[26,27-3H]D2 and its 24-epimer were also accomplished (7). Figure 5 illustrates the final separation of the two synthetic epimers. The synthetic 1,25-(OH)2-[26,27-3H]D2 and the enzymatically produced 1,25-(OH)2-[26,27‘HID2 exactly corn&ate on HPLC (results not shown). The identity of 1,25-(OH)*D* produced in
SYNTHESIS
OF VITAMIN
101
Dz COMPOUNDS
equally displaced by various concentrations of unlabeled 1,25-(OH)2D2 or 1,25-(OH)zD3. Displacement curves for 1,25-(OHh-[26,273H]D3 from the receptor protein by 1,25-(OH)*D* or 1,25-(OH)*D3 were identical to those in Fig. 5 (data not shown). 0.02
DISCUSSION
2* E x N 5
' 0.00
:
The present report provides for the first time a route to chemically synthesized radiolabeled 25-OH-D2 and 1,25-(OH)zD2 of high specific activity. 25-OH-[26,27-‘HID3 can also be quickly and largely converted enzymatically to radiolabeled 1,25-(OH)2-[26,273H]D2 of high specific activity. These preparations now make possible detailed examination of vitamin D2 metabolism previously not possible. It provides material for studying receptor interaction with the hormonal form of vitamin Dz, and makes available radiolabeled vitamin D2 metabolites for standardizing assays of members of the vitamin Dz series. The advantages of the present preparation are that the compounds prepared are unambiguous because of the chemical route used; the tritium is introduced at the terminal reaction of the series, eliminating the need to work with high levels of radioactivity during synthesis; and the tritium is located on the 26 and 27 carbons with high specific activity permitting truly physiological experiments.
'2
0.01
; e
0
0.02 2, 0.01
'P * E a O‘O I
0.00 Elution
volume
(ml)
FIG. 4. HPLC of (A) authentic 1,25-(OH)2D2 and (24R)- I ,25-(OH)zD2; (B) in vitro produced 1,25-(OH)Z[26,27-‘H]Dz; and (C) in vitro produced (24R)-1,25(OH)2-[26,27-3H]D2. Chromatographic conditions were as described in the text. The solid lines represent optical density at 254 nm, while each bar represents radioactivity of an aliquot of each 0.8-ml fraction.
vitro or synthetically was further confirmed by testing its binding ability to chick intestinal receptor for 1,25~(0H)~D~ (3). As shown in Fig. 6, 1,25-(OH)2-[26,27-3H]D2 bound to chick intestinal receptor was .05D
,
0
I
1
I
I
I
I
1
1
1
1
I
1
1
I
0
2
4
6
8
IO
12
14
16
I8
20
22
24
26
TIME
(min)
FIG. 5. HPLC of the 24-epimers of 1,25-(OH)2-[26,27-3H]D2. The purified product was chromatographed on a 4.6 mm X 25 cm Zorbax-Sil column using a solvent of 7.5% 2-propanol in hexane.
102
SICINSKI ET AL.
It is curious that 24-epi-25-OH-D2 elutes before 25-OH-D2 on silica during HPLC, while 24-epi- 1,25-(OH)zDz elutes after 1,25-(OH)2D2 in the same system. Exactly why this occurs is not clear, but the l-hydroxyl function certainly has a dominating influence on silica acid chromatography of the vitamin D compounds. These results also show that chromatographic behavior of these compounds is not entirely predictable. REFERENCES unlabelled
1,25-(OH),
D, or 1,25-(OH),D,(pmol/tube)
FIG. 6. Displacement of 1,25-(OH)z-[26,27-3H]D2 from chick intestinal receptor protein for 1,25-(OH)*D3 by unlabeled 1,25-(OHhDz (0) or 1,25-(OH)zD3 (0). Each point represents the mean value (*SD) oftriplicate determinations.
The only disadvantage is that side chain cleaved metabolites (if they exist) cannot be detected using this label. The synthetic route used is straightforward and thus unremarkable (6,7). The compounds obtained are radiochemically pure and are certainly authentic, as revealed by competitive binding studies, cochromatography, and enzymatic conversion to the expected 1a-hydroxylated derivatives. Thus, these preparations should provide the basis for rapid advances in our understanding of vitamin D2 metabolism. The present preparation provided radiolabeled 24-epi-25-OH-D2 and 24-epi- 1,25(OH)*D2 of high specific activities. Although this may not appear to be of importance currently, we have learned that 24-epi-1,25(OH)2D2 has interesting biological activities; thus the 24-epi-l,25-(OH)2-[26,27-3H]Dz may provide material useful in understanding the basis of its peculiar biological activities.
1. DeLuca, H. F., Paaren, H. E., and Schnoes, H. K. (1979) in Topics in Current Chemistry, pp. l-65, Springer-Verlag, Berlin; (1980) Annu. Rev. Med. Chem. l&288-301. 2. DeLuca, H. F., and Schnoes, H. K. (1983) Annu. Rev. Biochem. 52,4 1 l-439. 3. Suda, T., DeLuca, H. F., Schnoes, H. K., and Blunt, J. W. (1969) Biochemistry 8, 35 15-3520. 4. Jones, G., Schnoes, H. K., and DeLuca, H. F. (1975) Biochemistry 14, 1250-1256. 5. Jones, G., Schnoes, H. K., Levan, L., and DeLuca, H. F. (1980) Arch. Biochem. Biophys. 202, 450-457. 6. Morzycki, J. W., Schnoes, H. K., and DeLuca, H. F. (1984) J. Org. Chem. 49,2148-2151. 7. Sicinski, R. R., Tanaka, Y., Schnoes, H. K., and DeLuca, H. F. (1985) Bioorg. Chem. 13, 158-169. 8. Napoli, J. L., Fivizzani, M. A., Hamstra, A. J., Schnoes, H. K., and DeLuca, H. F. (1979) Anal. Biochem. 96,481-488. 9. Napoli, J. L., Mellon, W. S., Fivizzani, M. A., Schnoes, H. K., and DeLuca, H. F. (1980) Biochemistry 19, 25 15-252 1. 10. Omdahl, J., Holick, M., Suda, T., Tanaka, Y., and DeLuca, H. F. (1971) Biochemistry 10, 2935-2940. 11. Tanaka, Y., and DeLuca, H. F. (198 1) Anal. Biothem. 110, 102-107. 12. Shepard, R. M., Hors& R. L., Hamstra, A. J., and DeLuca, H. F. (1979) Biochem. J. 182, 55-69. 13. Jones, G., Schnoes, H. K., and DeLuca, H. F. (1976) J. Biol. Chem. 251, 24-28. 14. Eisman, J. A., and DeLuca, H. F. (1977) Steroids 30,245-257.
161, 96-102
(1987)
Synthesis of 25-Hydroxy-[26,27-3H]vitamin D2, 1 ,25-Dihydroxy-[26,27-3H]vitamin D2 and Their (24R)-Epimers’ RAFAL R. SICINSKJ~YOKOTANAKA,~
MARYPHELPS,HEINRICHK.SCHNOES, AND HECTOR F. DELUCA
Department of Biochemistry, University of Wisconsin, College ofAgricultural 420 Henry Mall, Madison, Wisconsin 53706
and L&e Sciences,
Received August 11, 1986 Synthesis of a C-24-epimeric mixture of 25-hydroxy-[26,27-3H]vitamin D2 and a C-24-epimeric mixture of 1,25-dihydroxy-[26,27-‘HIvitamin D2 by the Grignard reaction of the corresponding 25-keto-27-nor-vitamin D2 and lo-acetoxy-25keto-27-nor-vitamin D3 with tritiated methyl magnesium bromide is described. Separation of epimers by high-performance liquid chromatography afforded pure radiolabeled vitamins of high specific activity (80 Ci/mmol). The identities and radiochemical purities of 25-hydroxy-[26,27-3H[vitamin D2 and 1,25-dihydroxy-[26,27-3H]vitamin D2 were established by cochromatography with synthetic 25hydroxyvitamin Dz or 1,25-dihydroxyvitamin D 2. Biological activity of 25-hydroxy[26,27-3H]vitamin D2 was demonstrated by its binding to the rat plasma binding protein for vitamin D compounds, and by its in vitro conversion to l,25-dihydroxy-[26,27-3H]vitamin D2 by kidney homogenate prepared from vitamin D-deficient chickens. The biological activity of 1,25-dihydroxy-[26,27-3H]vitamin D2 was demonstrated by its binding to the chick intestinal receptor for 1,25-dihydroxyvitamin D3. 0 1987 Academic FWSS. hc. KEY WORDS: vitamin D; calcium metabolism; intestine; bone; calcium regulating hormones; phosphorus metabolism.
Metabolism of vitamin D3 has received considerable attention during the past two decades. Substantial advances in our understanding of vitamin D3 metabolic pathways and its mode of action have been closely connected with the development of synthetic methods providing radiolabeled vitamin D3 derivatives ( 1,2). The availability of these radioactive compounds of still higher specific activity is currently of considerable interest because they serve as convenient tools for a
number of detailed biomedical experiments. During the extensive investigations on the vitamin D3 compounds, parallel work on vitamin D2 metabolism was restricted because of a lack of available labeled compounds in spite of the common clinical use of vitamin Dz . The data available so far indicate that the main metabolic activation of vitamins D2 and D3 in mammals and birds is identical (1,2). Although 25-hydroxyvitamin D2 (25 OH-D2)4 and 1,25-dihydroxyvitamin D2 ( 1,25-(OH)2D2) were isolated and chemically identified in 1969 and 1974, respectively (3,4), they have not been tested extensively and their biological significance remains unclear. Jones et al. (5) reported the isolation
’ This work was supported by Grant AM-32701 and Program Project Grant AM-14881 from the National Institutes of Health, and by the Harry Steenbock Research Fund of the Wisconsin Alumni Research Foundation. ’ Present address: Department of Chemistry, University of Warsaw, u 1, Pasteura 1,02-093 Warsaw, Poland. 3 Present address: VA Medical Center, 113 Holland Avenue, Albany, NY 12208. 0003-2697187
$3.00
Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.
4 Abbreviations used: 1,25-(OHhD3, 1,25-dihydroxyvitamin D3; 25-OH-Dr , 25-hydroxyvitamin D3 ; CT3MgBr, tritiated methyl magnesium bromide. 96
SYNTHESIS
OF
VITAMIN
of 24-OH-D2 and (24R),25-(OH)2D2, but no information on their activity is available so far. Random labeled vitamin [3H]D2 (8.6 mCi/mmol) was obtained by Suda et al, (3) by exposing vitamin D2 to tritiated acetic acid, whereas vitamin [3a-3H]D2 (1.2 Ci/ mmol) was synthesized by Jones et al. (4) using sodium borotritide reduction of 3/3acetoxyergosta-3,5,7,22-tetraene. The corresponding ~~-OH-[~CX-~H]D~ was generated in vivo in small amounts in rats by the administration of vitamin [3a-3H]D2. We, therefore, turned our attention to a chemical synthesis of radioactive 25-OH-D2 of high specific activity. The first unpublished synthesis of 25OH-D2 has been achieved at the Upjohn Co., but this method practically excluded the possibility of radiolabeling at the side chain because its construction was performed at a very early step in the process (1). A very recent synthesis of 25-OH-D2 and its 24epimer, which has been carried out in our laboratory by Morzycki et al. (6), provided a direct precursor for both unlabeled and labeled compounds. Similarly, a convenient synthesis of 1,25-(OH)2D2 has been devised that also provides a direct precursor (7). Using these approaches, we obtained (24S)and (24R)-25-OH vitamin [26,27-3H]D2 (compounds 2a and 3a, Fig. 1) and (24S)and (24R)- 1,25-(OH)~-[26,27-~H]D~ (compounds 2b and 3b) of high specific activity (80 Ci/mmol). MATERIALS
AND METHODS
Radioactive determinations were performed with a Model 3255 Packard Tri-Carb liquid scintillation counter fitted with an automatic external standardization system (Packard Instruments, Downers Grove, IL). Radioactive samples were counted in a 3a70B counting solution (Research Products Int. Corp., Elk Grove, IL). Ultraviolet spectra were recorded in ethanolic solutions with a Beckman DB-G recording spectrophotometer (Beckman Instruments, Lincolnwood, IL).
D2 COMPOUNDS
97
Vitamin D compounds. 25-OH-D3 was a gift from the Upjohn Company (Kalamazoo, MI), while 1,25-(OH)2D3 was a gift from the Hoffmann-LaRoche Company (Nutley, NJ). 25OH-D2 and its 24-epimer were synthesized using the method of Morzycki et al. (6) while 1,25-(OH)*D* and its epimer were synthesized using the method of Sicinski et al. (7). 25-OH-[26,27-3H]D3 and 1,25-(OH)2-[26,27-3H]D3 were synthesized using previously described methods (8,9). Animals. One-day-old white Leghorn chicks were purchased from Northern Hatcheries (Beaver Dam, WI) and fed a vitamin D-deficient diet (10) for 4 weeks for a source of renal 25-OH-D3- 1-hydroxylase and chick intestinal 1,25-(OHhD3 receptor. In vitro production of 1,25-(OH)2-[26,27‘HID* and 24-epi-I,25(OH)2-[26,27-‘HID*. Kidneys from vitamin D-deficient chickens were pooled and a 20% (w/v) homogenate was prepared in an ice-cold 15 mM Tris-acetate buffer (pH 7.4 at room temperature) containing 0.19 M sucrose and 1.9 mM magnesium acetate. The incubation was carried out in a 125-ml Erlenmeyer flask containing 1.2 g of kidney tissue, 0.19 M sucrose, 15 mM Tris-acetate (pH 7.4 at room temperature), 1.9 mM magnesium acetate, and 25 mM succinate in a 9-ml final volume as described by Tanaka and DeLuca (11). The reaction was initiated by addition of 120 PCi ( 1.5 nmol) of either 25-OH-[26,27-3H]D2 or 24-epi-25OH-[26,27-3H]D2 dissolved in 60 ~1 of 95% ethanol. The reaction mixtures were incubated at 37°C with shaking at 100 oscillations/min. The reaction was terminated by the addition of 18 ml of MeOH and 9 ml of dichloromethane. Extraction was carried out by the method described by Tanaka and DeLuca (11). The extract was prepurified on a small column of Sephadex LH-20 as described earlier ( 11). The purified extract from a Sephadex LH-20 column was injected into a Waters Associates Model ALC/GPC 204 HPLC system (Milford, MA) equipped with a DuPont Zorbax-Sil column (4.6 mm X 25 cm) (Wilmington, DE) developed with a solvent system of 8% 2-propanol in hexane at a
98
SICINSKI ET AL.
HO . . . . c la, R=ti lb, R=OH
FIG.
HO . . . . p
2a, 2b,
1. Synthesis of 25-OH-[26,27-3H]D2,
R=H R=OH
3a, R=H 3b,
1,25-(OH)2-[26,27-3H]Dz,
pressure of 800 psi. A 254~nm ultraviolet monitor was used to detect authentic compounds or compounds produced in vitro, while fractions of 0.8 ml each were collected, and an aliquot of each fraction was used for detection of tritium. Displacement of 2.5-OH-[26,27-3H]D2 for rat plasma binding protein for vitamin D compounds by 25-OH-D* or 25-OH-Dj. Diluted rat plasma ( 1:5000) was incubated with 25-OH-[26,27-3H]D2 and graded amounts of unlabeled 25OH-D* or 25-OH-D3 dissolved in 50 ~1 of 95% ethanol at 4’C for 16 h, as described by Shepard et al. (12). Duplicate determinations for each concentration of either compound were carried out. Displacement of 1,25-(0H)~-[26,27-~H]D~ from chick intestinal cytosol receptor protein for 1,25-(OH)*D3. Chick intestinal cytosol was incubated with 1,25-(0H)2-[26,273H]D2 and various concentrations of unlabeled 1,25-(OH)*DZ or 1,25-(OH)2D3 dissolved in 50 ~1 of 95% ethanol at 4°C for 16 h as described by Shepard et al. ( 12). Triplicate determinations for each concentration were carried out. Synthesis of 25-OH-[26,27-‘HID2 compounds (2a and 3a). 25-Keto-27-nor-vitamin Dz (la) (Fig. 1) was obtained by the method of Morzycki et al. (6) as a mixture of 24epimers. A solution of la (1 mg) in 2 ml of dry ether was added dropwise with a syringe through a septum into a solution of 0.5 mmol of tritiated methyl magnesium bromide5 (CT3MgBr) in 0.5 ml of dry ether. The 5 The methylation step was carried out in the laboratories of New England Nuclear (Boston, MA). Workup of the reaction and subsequent experiments were performed in our laboratory.
R=OH
and their C-24-epimers.
mixture was maintained at 0°C for the period of addition, then stirred at room temperature for 2 h. The solvent was evaporated under vacuum, and labile tritium was removed by three successive evaporations under vacuum of 2 ml portions of methylene chloride-methanol (1: 1). The mixture was then frozen in a liquid nitrogen bath. After addition of 1 ml of saturated aqueous ammonium chloride, the mixture was allowed to warm to room temperature. It was then dried and labile tritium was removed as before. The crude mixture was stored temporarily in 10 ml of ethanol. It was partitioned between 60 ml of ether (previously washed with water) and 3 X 20 ml of water. The aqueous layers were combined and reextracted with ether. An assay showed that 95% of the radioactivity was in the main extract. The crude reaction product was redissolved in 1 ml of hexane:methanol:chloroform (9: 1: 1) and applied to a 2X 30-cm Sephadex LH-20 column, Chromatography of the radioactive compounds was carried out with the same solvent mixture at a flow rate of 1 ml/min and 9-ml fractions were collected. The fractions in the 25-OH-D2 region (fractions 1 1- 18) were pooled, and the solvent was removed under vacuum. This material was further purified by high-performance liquid chromatography on a DuPont 830 LC (DuPont Instruments, Wilmington, DE) equipped with a DuPont Zorbax-Sil column (4.6 mm X 25 cm) using 1% 2-propanol in hexane as the solvent (flow rate 2 ml/min, 1100 psi); the eluted compounds were monitored at 254 nm. The peaks of radioactive 25-hydroxyvitamin D compounds overlapped; the (24S)-epimer was collected at
SYNTHESIS
OF VITAMIN
Dz COMPOUNDS
99
Both C-24-epimers of 25-OH-[26,27‘HID2 with specific activities of 80 Ci/mmol were obtained by the Grignard reaction of 25-keto-27-nor-vitamin D2 with CT3MgBr.6 Separation of the epimers was achieved by
HPLC. Radioactive purity and identification of 25-OH-[26,27-3H]D2 and its C-2bepimer were provided by cochromatography of the radiolabeled compound and synthetic isomers of unlabeled 25-OH-D2. One microcurie of 25-OH-[26,27-3H]D2 was mixed with 50 pg of the unlabeled (24R)-epimer of 25-OH-D2 and 25 pg of unlabeled 25-OH-D* to give a final specific activity of 0.04 wCi/bg. The mixture was subjected to an HPLC on a Zorbax-Sil semipreparative column. As shown in Fig. 2, a baseline separation of the C-24-epimers of 25-OH-D2 was achieved when fractions of each isomer were recycled three times by placing the HPLC system in a recycle mode. The recovered 25-OH-D2 had not changed in specific activity and no trace of radioactive (24R)-epimer was found. The biological identity of 25-OH-[26,273H]D2 was established by its competitive binding activity for rat plasma vitamin D transport protein. It has been shown that the binding abilities of 25-OH-D2 and 25-OH-D3 to the rat plasma protein are identical (Tanaka and DeLuca, unpublished results). As shown in Fig. 3, displacement curves were obtained by competition with increasing concentration with unlabeled 25-OH-D2 for binding sites previously saturated with a fixed amount of 25-OH-[26,27-3H]D2. Displacement curves obtained by competition of unlabeled 25-OH-D2 or 25-OH-D, for binding sites saturated with 25-OH-[26,273H]D3 were identical to those in Fig. 3 (data not shown). Evidence for biological activity of 25-OH[26,27-3H]D2 was further established by its conversion to 1,25-(OH)2-[26,27-3H]D2 in vitro by vitamin D-deficient chick kidney homogenate. It has been demonstrated by Jones et al. (13) that an in vitro chick kidney mitochondrial system converts 25-OH-D2 to 1,25-(OH)2D2. As shown in Fig. 4A, a baseline separation of chemically synthesized isomers of 1,25-(OH)2D2 was achieved by HPLC with a Zorbax-Sil column using a sol-
6 The introduction of label by the Grignard reaction produces an asymmetric center on C-25 with the label entirely on one carbon. Since the introduction is not
stereospecific, the label will be found on both C-26 and C-27 and is so designated.
68.5-71.5 ml and the (24R)-compound was collected at 73.5-79 ml. Both fractions (about 90% purity) were rechromatographed separately by HPLC in the same system yielding pure stereoisomers. Both compounds 2a and 3a exhibited the expected ultraviolet absorption (X,,, 264.5 nm) and identical constant specific activity of 80 Ci/ mmol. The identities and radiochemical purities of 25-OH-[26,27-3H]D2 and its 24epimer were also confirmed by HPLC cochromatography with synthetic 25-OH-D2 epimers. Synthesis of 1,25-(OH)2-[26,27-‘H]DJ compounds. The syntheses of 1,25-(OH)2[26,27-3H]D2 and its 24-epimer were carried out as described for 25-OH-[26,27-3H]D3 above except that la-hydroxy-25-keto-27nor-vitamin D2 (lb) as l-acetate (7) was used as the starting material. The radiolabeled crude 1,25-(OH)2-[26,27-3H]D2-epimeric mixture was chromatographed on a precalibrated 1 X 60-cm Sephadex LH-20 column packed and developed in CHC13:hexane (65:35). Fractions containing a mixture of the two isomers were collected and further purified by HPLC using a 4.6 mm X 25 cm Zorbax-Sil column (DuPont, Wilmington, DE), with 9% 2-propanol in hexane as the solvent system. The combined peaks were collected and chromatographed again on Zorbax-Sil with a solvent of 7% 2-propanol in hexane. This provided baseline separation of the epimers which were collected separately (2b, 3b). On rechromatography, each gave a single symmetrical peak (not shown). Both the 24-epi-1,25-(OH)2-[26,27-3H]D2 and 1,25-(OH)2-[26,27-3H]D2 had specific activities of 78 Ci/mmol. RESULTS
100
SICINSKI ET AL.
I.0 24-epi-25-OH-D2 f 0.6 -
E c N s
.
0.6
”
25-OH-D, --
I’
I
0.4.
a-
s 8 0.2 -
D
Jc
; 50
/ 100
Elution
150
volume
200
250
300
(ml)
FIG. 2. Chromatography of 25-OH-[26,27-3H]D2 with authentic C-2Cepimers of 25-OH-D2 on HPLC. Chromatography was performed on a Zorbax-Sil semipreparative column (6.2 mm X 25 cm) developed with 2% 2-propanol in hexane at a pressure of 1500 psi and a flow rate of 4 ml/min. The solid line represents optical density at 254 nm. The effluent was recycled three times. Fractions of 1.6 ml each were collected between 270 and 320 ml, and an aliquot of each fraction was counted as shown by the broken line (inset). The final elution volume of unlabeled 25-OH-D2 was 303 ml, while that of unlabeled (24S)-25-OH-D2 was 305 ml, due to an isotope effect. Fractions containing 25-OH-D* were combined to measure the vitamin D triene by means of a uv spcctrophotometer to calculate specific activity.
vent system of 8% 2-propanol in hexane. A compound biologically generated from 25-OH-[26,27-3H]D2 was detected by monitoring radioactivity and uv absorption, and was eluted at the position of authentic
unlobelled
25-OH-D,
or 25-OH-D,
(pmol/tube)
FIG. 3. Displacement of25-OH-[26,27-‘HID2 from rat plasma vitamin D binding protein by unlabeled 25-OHD2 (0) or 25-OH-D, (0). Each point represents the mean value of triplicate determinations.
1,25-(OH)2D2 as shown in Figure 4B. The area of the peak detected by uv absorption was measured and the amount of the compound produced in vitro was calcuiated from a standard curve. This curve was constructed by plotting the area under the uv absorption peak versus the amount of authentic 1,25-(OH)*D* injected into HPLC as described by Tanaka et al. (11). It was found that 41 &i/220 ng of 1,25-(OH)zD2 was produced in vitro, giving a specific activity of 80 Ci/mmol. Similarly, the (24Qepimer of 1,25-(OH)zDz was produced from 24-epi-25OH-[26,27-3H]D2 as shown in Fig. 4C. Its specific activity was found to be identical to that of its substrate or 48 &i/225 ng. Direct chemical syntheses of 1,25-(OH)*[26,27-3H]D2 and its 24-epimer were also accomplished (7). Figure 5 illustrates the final separation of the two synthetic epimers. The synthetic 1,25-(OH)2-[26,27-3H]D2 and the enzymatically produced 1,25-(OH)2-[26,27‘HID2 exactly corn&ate on HPLC (results not shown). The identity of 1,25-(OH)*D* produced in
SYNTHESIS
OF VITAMIN
101
Dz COMPOUNDS
equally displaced by various concentrations of unlabeled 1,25-(OH)2D2 or 1,25-(OH)zD3. Displacement curves for 1,25-(OHh-[26,273H]D3 from the receptor protein by 1,25-(OH)*D* or 1,25-(OH)*D3 were identical to those in Fig. 5 (data not shown). 0.02
DISCUSSION
2* E x N 5
' 0.00
:
The present report provides for the first time a route to chemically synthesized radiolabeled 25-OH-D2 and 1,25-(OH)zD2 of high specific activity. 25-OH-[26,27-‘HID3 can also be quickly and largely converted enzymatically to radiolabeled 1,25-(OH)2-[26,273H]D2 of high specific activity. These preparations now make possible detailed examination of vitamin D2 metabolism previously not possible. It provides material for studying receptor interaction with the hormonal form of vitamin Dz, and makes available radiolabeled vitamin D2 metabolites for standardizing assays of members of the vitamin Dz series. The advantages of the present preparation are that the compounds prepared are unambiguous because of the chemical route used; the tritium is introduced at the terminal reaction of the series, eliminating the need to work with high levels of radioactivity during synthesis; and the tritium is located on the 26 and 27 carbons with high specific activity permitting truly physiological experiments.
'2
0.01
; e
0
0.02 2, 0.01
'P * E a O‘O I
0.00 Elution
volume
(ml)
FIG. 4. HPLC of (A) authentic 1,25-(OH)2D2 and (24R)- I ,25-(OH)zD2; (B) in vitro produced 1,25-(OH)Z[26,27-‘H]Dz; and (C) in vitro produced (24R)-1,25(OH)2-[26,27-3H]D2. Chromatographic conditions were as described in the text. The solid lines represent optical density at 254 nm, while each bar represents radioactivity of an aliquot of each 0.8-ml fraction.
vitro or synthetically was further confirmed by testing its binding ability to chick intestinal receptor for 1,25~(0H)~D~ (3). As shown in Fig. 6, 1,25-(OH)2-[26,27-3H]D2 bound to chick intestinal receptor was .05D
,
0
I
1
I
I
I
I
1
1
1
1
I
1
1
I
0
2
4
6
8
IO
12
14
16
I8
20
22
24
26
TIME
(min)
FIG. 5. HPLC of the 24-epimers of 1,25-(OH)2-[26,27-3H]D2. The purified product was chromatographed on a 4.6 mm X 25 cm Zorbax-Sil column using a solvent of 7.5% 2-propanol in hexane.
102
SICINSKI ET AL.
It is curious that 24-epi-25-OH-D2 elutes before 25-OH-D2 on silica during HPLC, while 24-epi- 1,25-(OH)zDz elutes after 1,25-(OH)2D2 in the same system. Exactly why this occurs is not clear, but the l-hydroxyl function certainly has a dominating influence on silica acid chromatography of the vitamin D compounds. These results also show that chromatographic behavior of these compounds is not entirely predictable. REFERENCES unlabelled
1,25-(OH),
D, or 1,25-(OH),D,(pmol/tube)
FIG. 6. Displacement of 1,25-(OH)z-[26,27-3H]D2 from chick intestinal receptor protein for 1,25-(OH)*D3 by unlabeled 1,25-(OHhDz (0) or 1,25-(OH)zD3 (0). Each point represents the mean value (*SD) oftriplicate determinations.
The only disadvantage is that side chain cleaved metabolites (if they exist) cannot be detected using this label. The synthetic route used is straightforward and thus unremarkable (6,7). The compounds obtained are radiochemically pure and are certainly authentic, as revealed by competitive binding studies, cochromatography, and enzymatic conversion to the expected 1a-hydroxylated derivatives. Thus, these preparations should provide the basis for rapid advances in our understanding of vitamin D2 metabolism. The present preparation provided radiolabeled 24-epi-25-OH-D2 and 24-epi- 1,25(OH)*D2 of high specific activities. Although this may not appear to be of importance currently, we have learned that 24-epi-1,25(OH)2D2 has interesting biological activities; thus the 24-epi-l,25-(OH)2-[26,27-3H]Dz may provide material useful in understanding the basis of its peculiar biological activities.
1. DeLuca, H. F., Paaren, H. E., and Schnoes, H. K. (1979) in Topics in Current Chemistry, pp. l-65, Springer-Verlag, Berlin; (1980) Annu. Rev. Med. Chem. l&288-301. 2. DeLuca, H. F., and Schnoes, H. K. (1983) Annu. Rev. Biochem. 52,4 1 l-439. 3. Suda, T., DeLuca, H. F., Schnoes, H. K., and Blunt, J. W. (1969) Biochemistry 8, 35 15-3520. 4. Jones, G., Schnoes, H. K., and DeLuca, H. F. (1975) Biochemistry 14, 1250-1256. 5. Jones, G., Schnoes, H. K., Levan, L., and DeLuca, H. F. (1980) Arch. Biochem. Biophys. 202, 450-457. 6. Morzycki, J. W., Schnoes, H. K., and DeLuca, H. F. (1984) J. Org. Chem. 49,2148-2151. 7. Sicinski, R. R., Tanaka, Y., Schnoes, H. K., and DeLuca, H. F. (1985) Bioorg. Chem. 13, 158-169. 8. Napoli, J. L., Fivizzani, M. A., Hamstra, A. J., Schnoes, H. K., and DeLuca, H. F. (1979) Anal. Biochem. 96,481-488. 9. Napoli, J. L., Mellon, W. S., Fivizzani, M. A., Schnoes, H. K., and DeLuca, H. F. (1980) Biochemistry 19, 25 15-252 1. 10. Omdahl, J., Holick, M., Suda, T., Tanaka, Y., and DeLuca, H. F. (1971) Biochemistry 10, 2935-2940. 11. Tanaka, Y., and DeLuca, H. F. (198 1) Anal. Biothem. 110, 102-107. 12. Shepard, R. M., Hors& R. L., Hamstra, A. J., and DeLuca, H. F. (1979) Biochem. J. 182, 55-69. 13. Jones, G., Schnoes, H. K., and DeLuca, H. F. (1976) J. Biol. Chem. 251, 24-28. 14. Eisman, J. A., and DeLuca, H. F. (1977) Steroids 30,245-257.