Chemical Synthesis of Side-Chain-Hydroxylated Vitamin D3 Derivatives and Their Metabolism by CYP27B1

Article abstract of DOI:10.1002/cbic.202100250

1α,25-Dihydroxyvitamin D₃ (abbreviated here as 1,25D₃) is a hormonally active form of vitamin D₃ (D₃), and is produced from D₃ by CYP27 A1-mediated hydroxylation at C25, followed by CYP27B1-mediated hydroxylation at C1. Further hydroxylation of 25D₃ and 1,25D₃ occurs at C23, C24 and C26 to generate corresponding metabolites, except for 1,25R,26D₃. Since the capability of CYP27B1 to hydroxylate C1 of side-chain-hydroxylated metabolites other than 23S,25D₃ and 24R,25D₃ has not been examined, we have here explored the role of CYP27B1 in the C1 hydroxylation of a series of side-chain-hydroxylated D₃ derivatives. We found that CYP27B1 hydroxylates the R diastereomers of 24,25D₃ and 25,26D₃ more effectively than the S diastereomers, but shows almost no activity towards either diastereomer of 23,25D₃. This is the first report to show that CYP27B1 metabolizes 25,26D₃ to the corresponding 1α-hydroxylated derivative, 1,25,26D₃. It will be interesting to examine the physiological relevance of this finding.

Full text of DOI:10.1002/cbic.202100250

Full Papers
doi.org/10.1002/cbic.202100250
ChemBioChem
Chemical Synthesis of Side-Chain-Hydroxylated Vitamin D3
Derivatives and Their Metabolism by CYP27B1
Ryota Sakamoto,[a] Akiko Nagata,[a] Haruki Ohshita,[b] Yuka Mizumoto,[a] Miho Iwaki,[a]
Kaori Yasuda,*[b] Toshiyuki Sakaki,*[b] and Kazuo Nagasawa*[a]
1α,25-Dihydroxyvitamin D3 (abbreviated here as 1,25D3) is a
hormonally active form of vitamin D3 (D3), and is produced from
D3 by CYP27 A1-mediated hydroxylation at C25, followed by
CYP27B1-mediated hydroxylation at C1. Further hydroxylation
of 25D3 and 1,25D3 occurs at C23, C24 and C26 to generate
corresponding metabolites, except for 1,25R,26D3. Since the
capability of CYP27B1 to hydroxylate C1 of side-chain-hydroxy-
lated metabolites other than 23S,25D3 and 24R,25D3 has not
been examined, we have here explored the role of CYP27B1 in
the C1 hydroxylation of a series of side-chain-hydroxylated D3
derivatives. We found that CYP27B1 hydroxylates the
R
diastereomers of 24,25D3 and 25,26D3 more effectively than the
S diastereomers, but shows almost no activity towards either
diastereomer of 23,25D3. This is the first report to show that
CYP27B1 metabolizes 25,26D3 to the corresponding 1α-hy-
droxylated derivative, 1,25,26D3. It will be interesting to
examine the physiological relevance of this finding.
Introduction
Vitamin D3 (VD3) is a steroidal hormone originally discovered as
an anti-rickets agent in 1928.[1] VD3 is biosynthetically generated
from 7-dehydrocholesterol in the skin upon exposure to
sunlight, which causes photolytic cleavage of the B ring
followed by thermal isomerization (Scheme 1). Then, VD3 is
metabolized to 1α,25-dihydroxyvitamin D3 [(1b), abbreviated
here as 1,25D3; other hydroxylated derivatives are designated
similarly], an active form of VD3 with roles in the control of
calcium homeostasis and bone formation through binding to its
receptor, VDR.[2] Hydroxylation of VD3 at C25 is mediated by
CYP2R1 or CYP27 A1, followed by hydroxylation at C1 mediated
by CYP27B1 (Scheme 1).[3]
Scheme 1. Metabolic conversion of vitamin D3 (VD3) into 25D3 (1a) and
1,25D3 (1b) by CYPs.
The side chain of 1,25D3 (1b) is further metabolized by
CYP24 A1 and CYP3 A4 to generate C23-, C24- or C26-hydroxy-
lated derivatives (Figure 1 and path A in Scheme 2). That is,
1,25D3 (1b) is metabolized by CYP24 A1 to produce 1,23S,25D3
(2b) and 1,24R,25D3 (5b), which are converted into the final
metabolites, 1,25D3-26,23 lactone (8) and calcitroic acid (9),
[a] R. Sakamoto, A. Nagata, Y. Mizumoto, M. Iwaki, Dr. K. Nagasawa
Department of Biotechnology and Life Science
Graduate School of Technology,
Tokyo University of Agriculture and Technology
2-24-16, Naka-cho, Koganei,
184-8588, Tokyo (Japan)
E-mail: knaga@cc.tuat.ac.jp
[b] H. Ohshita, Dr. K. Yasuda, Prof. T. Sakaki
Department of Pharmaceutical Engineering, Faculty of Engineering
Toyama Prefectural University, 5180 Kurokawa, Imizu,
Toyama 939-0398 (Japan)
E-mail: kyasuda@pu-toyama.ac.jp
Scheme 2. Metabolic pathways of vitamin D3 (VD3) and its derivatives 2–7
with CYPs.
tsakaki@pu-toyama.ac.jp
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cbic.202100250
This article is part of a Special Collection in collaboration with the Asian
Chemical Biology Initiative (ACBI). Please see our homepage for more articles
in the collection.
ChemBioChem 2021, 22, 1–6
1
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cbic.202100250
ChemBioChem
25R,26D3 (7a), would be converted into the corresponding 1α-
hydroxylated metabolites (Scheme 2, this work).
Here, we explored the metabolism of 23R,25D3 (3a),
24S,25D3 (4a), 25S,26D3 (6a), and 25R,26D3 (7a) to the
corresponding 1α-hydroxylated metabolites by CYP27B1. The
synthetic routes to 25D3 as well as 1,25D3 derivatives 2a–7a
and 2b–7b were also re-examined.
Results and discussion
Two major strategies have been reported for a synthesis of VD3
and its derivatives, i.e., the classical biosynthetic-type strategy
from 7-dehydrocholesterols via ring-opening cleavage of the B-
ring, and a convergent strategy involving coupling reactions
between the A–ring or its precursors and the corresponding
CD–ring structures.[12,13,14] Both strategies have been applied for
the synthesis of 23-, 24-, and 26-hydroxylated 25D3 derivatives
(Scheme 3).
In our synthesis of 2–7, we adopted the convergent
strategy, employing Wittig-type coupling between A–ring
phosphonates 13[15] and CD–ring ketones 10–12, and we also
explored concise synthetic approaches for the CD–ring struc-
tures.
The CD–ring structures 10 and 11 for 23,25D3 (2 and 3) and
24,25D3 (4 and 5) were synthesized by slightly modifying
previously reported methods (Scheme 4 and Scheme 5).[12b,13]
Installation of the hydroxyl group at C23 was carried out by
reaction of the aldehyde 14 and ethyl bromoacetate in the
Figure 1. Structures of 23-, 24-, and 26-hydroxylated D3 derivatives 2–7, and
1,25(OH)2D3-23,26 lactone (8) and calcitroic acid (9).
respectively (Figure 1).[4] On the other hand, CYP3 A4 is reported
to convert 1,25D3 (1b) into 1,23R,25D3 (3b) and 1,24S,25D3
(4b).[5] A 1α-hydroxylated metabolite, 1,25,26D3 (6b), was
identified in VDR null mice, and its stereochemistry was
assigned as 25S, although the relevant CYP species was not
identified.[6] The 25R metabolite of 1,25R,26D3 (7b) has not
been discovered in vivo so far.
presence of indium as a Lewis acid under ultrasound
irradiation.[16] Under these conditions, β-hydroxy esters 15a and
15b with 23S and 23R configurations, respectively, were
obtained in 57% and 43% yields as mixtures, which were
separable on a silica gel column.[12b] Then, 15a and 15b were
converted into the ketones 10a and 10b, respectively, via the
following 4 steps: (i) installation of the methyl group at C25
with methyl magnesium Grignard reagent, (ii) acetylation of the
hydroxyl group at C23 with acetic anhydride and DMAP, (iii)
deprotection of the TBS ether with PTS, and (iv) oxidation of the
Metabolism in the side chain of 25D3 (1a) has also been
reported; CYP24 A1 generates 23S- and 24R-hydroxylated
metabolites 2a and 5a, and CYP3 A4 generates 23R- and 24S-
hydroxylated metabolites 3a and 4a, analogously to the case of
the 1α-hydroxylated metabolites 2b-5b (Scheme 2, path B).[7]
The C26-hydroxylated metabolites 6 and 7 have not been well
studied compared to the C23- and C24-hydroxylated metabo-
lites 2–5, although 25,26D3 (6a or 7a) was identified in the early
1970s.[8] In 1981, Baggiolini and co-workers determined the
stereochemistry of 25,26D3 as 25S based on a synthetic
approach.[9] Recently, Sakaki and co-workers have found that
26-hydroxylated 25S,26D3 (6a) and 25R,26D3 (7a) are generated
by CYP27 A1 from 25D3 (1a).[10] Thummel reported that these
metabolites 6a and 7a were also produced from 1a by
CYP3 A4.[7] Among these metabolites, further metabolism of 2a
and 5a to the corresponding 1α-hydroxylated metabolites by
CYP27B1 was studied. In the case of 24R,25D3 (5a), conversion
to 5b occurred efficiently, but CYP27B1 did not catalyze the
hydroxylation of 23S,25D3 (2a).[11] These preliminary observa-
tions prompted us to explore whether other hydroxylated
analogs, 23R,25D3 (3a), 24S,25D3 (4a), 25S,26D3 (6a), and
Scheme 3. Convergent strategy for the synthesis of 23-, 24-, and 26-
hydroxylated 25D3 2–7.
ChemBioChem 2021, 22, 1–6
www.chembiochem.org
2
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cbic.202100250
ChemBioChem
single diastereomers, respectively. The selectivity was confirmed
by HPLC. The resulting diols 18 were further converted into
ketones 11a and 11b by deprotection of the benzoate ester
with DIBALÀ H, protection of the diol as an acetonide, and
oxidation of the secondary alcohol with TPAP. Then, the
coupling reaction of these CD–rings 11 with A–ring 13a
afforded the corresponding 24S,25D3 (4a) and 24R,25D3 (5a) in
69% and 68% yield, respectively, after deprotection of
acetonide and silyl ether under acidic conditions. The reference
compounds 1,24S,25D3 (4b) and 1,24R,25D3 (5b) were similarly
synthesized from A–ring 13b and CD–rings 11.
Scheme 4. Synthesis of 23-hydroxylated VD3 2 and 3. Reagents and
The 26-hydroxylated CD–ring compounds 12a and 12b
were synthesized from the triols (23R,25S)-19a and (23S,25R)-
19b, as reported by our group[19] (Scheme 6). The 1,2-diol
moiety of the triols 19 was protected with acetone in the
presence of PTS, followed by reduction of the secondary alcohol
at C23 using the Barton method via xanthate to give 20a and
20b in 50% and 48% yield through two steps, respectively.
Then, deprotection of the TBS ether in 20 was carried out by HF
treatment. This also removes the acetonide, so C25-26 were re-
protected with acetonide to give the secondary alcohol, which
was oxidized with TPAP to afford ketone 12a and 12b in 84%
and 83% yield from 20. These CD–ring compounds 12a and
12b were coupled with the A–ring 13a to give 25S,26D3 (6a)
and 25R,26D3 (7a) in 63% and 57% yield, respectively, after
deprotection of the acetonide group and silyl ether under acidic
conditions. 1,25S,26D3 (6b) and 1,25R,26D3 (7b) were similarly
synthesized using A–ring compound 13b.
°
conditions: (a) MeMgBr (5.8 equiv.), THF, 0 C; (b) Ac2O (15 equiv.), Et3N
(15 equiv.), DMAP (0.5 equiv.), CH2Cl2, reflux; (c) HF aq., CH2Cl2/CH3CN=9:1,
rt; (d) TPAP (0.1 equiv.), NMO (4.0 equiv.), MS4 A, CH2Cl2, rt; (e) 13a or 13b
°
(2.5 equiv.), n-BuLi (2.4 equiv.), THF, À 78 C; (f) p-TsOH·H2O (3.0 equiv.),
MeOH, rt; (g) K2CO3 (3.0 equiv.), MeOH, rt. THF=tetrahydrofuran,
Ac2O=acetic anhydride, Et3N=triethylamine, DMAP=4-dimeth-
ylaminopyridine, TPAP=tetrapropylammonium perruthenate, NMO=4-
methylmorpholine N-oxide.
Next, the hydroxylation of 23,25D3 (2a and 3a), 24,25D3 (4a
and 5a), and 25,26D3 (6a and 7a) at C1 with CYP27B1 was
investigated, using a reconstituted system of CYP27B1, adreno-
doxin reductase, and adrenodoxin (see Supporting Informa-
tion).
As shown in Supporting Figure S1, 24S,25D3 (4a), 24R,25D3
(5a), 25S,26D3 (6a) and 25R,26D3 (7a) were each obtained as a
single product whose retention time coincided with that of the
1α-hydroxylated product. In contrast, no metabolite was formed
from 23S,25D3 (2a) or 23R,25D3 (3a). The 1α-hydroxylase
activities of CYP27B1 towards the side-chain-hydroxylated
Scheme 5. Synthesis of 24-hydroxylated VD3 4 and 5. Reagents and
°
conditions: (a) conc. H2SO4 (22 equiv.), 1,4-dioxane, 0 C; (b) AD-mix-α or AD-
°
mix-β, MeSO2NH2 (1.0 equiv.), t-BuOH/H2O=1:1, 0 C; (c) DIBALÀ H
°
(4.0 equiv.), toluene, 0 C; (d) p-TsOH·H2O (0.5 equiv.), acetone; (e) TPAP
(0.1 equiv.), NMO (4.0 equiv.), MS4 A, CH2Cl2, rt; (f) 13a or 13b (2.5 equiv.), n-
BuLi (2.4 equiv.), THF, À 78 C; (g) p-TsOH·H2O (3.5 equiv.), MeOH, rt. DIBAL-
°
H=diisobutylaluminum hydride.
resulting secondary alcohol with TPAP. These CD–ring synthons
were reacted with the A–ring moiety 13a to give the
corresponding coupling products, which were further subjected
to deprotection of the silyl ether and acetyl group with PTS,
followed by treatment with potassium carbonate to give 2a
and 3a in 92% and 84% yield, respectively, in three steps. The
1α-hydroxylated analogs 2b and 3b, which were used as
reference compounds, were synthesized by reacting A–ring
13b and CD–rings 10a and 10b in 53% and 66% yield,
respectively.
For the synthesis of 24-hydroxylated CD–rings 11a and
11b, the asymmetric dihydroxylation developed by Sharpless[17]
was applied to the tri-substituted olefin 17, which was
synthesized by dehydration of tertiary alcohol 16.[18] Upon
treatment of 16 with concentrated sulfonic acid in 1,4-dioxane
Scheme 6. Synthesis of 26-hydroxylated VD3 6 and 7; Reagents and
conditions: (a) p-TsOH·H2O (0.05 equiv.), acetone, rt; (b) NaH (10 equiv.), CS2
°
(14 equiv.), MeI (10 equiv.), THF, 50 C; (c) AIBN (0.5 equiv.), Bu3SnH
°
at 0 C, the dehydrated product 17 was obtained selectively in
°
(5.0 equiv.), toluene, 110 C; (d) HF aq., CH2Cl2/CH3CN=9:1, rt; (e) p-TsOH·H2O
65% yield. Then, asymmetric dihydroxylation was carried out by
utilizing AD-mix-α or AD-mix-β. Under these conditions, 24S-
18a and 24R-18b were obtained in 79% and 95% yield as
(0.05 equiv.), acetone, rt; (f) TPAP (0.1 equiv.), NMO (4.0 equiv.), MS4 A,
°
CH2Cl2, rt; (g) 13a or 13b (2.5 equiv.), n-BuLi (2.4 equiv.), THF, À 78 C; (h) p-
TsOH·H2O (4.0 equiv.), MeOH, rt. AIBN=2,2’-azobis(isobutyronitrile).
ChemBioChem 2021, 22, 1–6
www.chembiochem.org
3
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cbic.202100250
ChemBioChem
metabolites 6b and 7b. Further study will be needed to
Table 1. CYP27B1-dependent 1α-hydroxylation activity towards 25D3 and
its 23-, 24-, or 26-hydroxylated derivatives.
establish the physiological relevance of this finding.
Entry
1α-Hydroxylation activity
(nmol/min/nmol-P450)[a]
Acknowledgements
1
2
3
4
5
6
7
25D3 (1a)
0.96�0.28
<0.05
<0.05
0.44�0.07
1.50�0.16
0.43�0.02
0.81�0.04
23S,25D3 (2a)
23R,25D3 (3a)
24S,25D3 (4a)
24R,25D3 (5a)
25S,26D3 (6a)
25R,26D3 (7a)
This work was partially supported by JST-OPERA, and was inspired
by the international and interdisciplinary environments of JSPS
Asian CORE Program of ACBI (Asian Chemical Biology Initiative).
This work was also supported by JSPS KAKENHI (19H02889, and
19 K05829).
[a] 1α-Hydroxylation activity was measured at the substrate concentration
of 5 μM. Data are the mean�SD from three separate experiments.
Conflict of Interest
substrates 2a-7a were calculated from the peak areas in the
HPLC chromatograms (Supporting Figure S1). In this assay,
metabolism of 25D3 (1a) to 1,25D3 (1b) was also tested. The
results are summarized in Table 1.
The authors declare no conflict of interest.
Keywords: CYP27B1
dihydroxyvitamin D3
dihydroxyvitamin D3
·
·
vitamin D3 metabolites
24,25-dihydroxyvitamin D3
·
23,25-
In the case of C23-hydroxylated 23,25D3 (2a and 3a), the
1α-hydroxylation activity of CYP27B1 towards 23S,25D3 (2a)
was quite low, and almost no hydroxylated 2b was obtained, in
accordance with the report by Sakaki.[10] Interestingly, the 23R-
isomer of 3a did not serve as a substrate for CYP27B1. Thus,
CYP27B1 is inactive towards 23-hydroxylated D3 regardless of
the stereochemistry. The efficacy of hydroxylation at C1 by
CYP27B1 was different depending on the stereochemistry at
C24 in the case of 24,25D3 (4a and 5a). In the case of 24R,25D3
(5a), as reported by Sakaki,[10] high 1α-hydroxylation activity
was observed with a turnover number of 1.50 minÀ 1, whereas a
lower turnover number of 0.44 minÀ 1 was obtained for 24S,25D3
(4a) as a substrate (Table 1). The 26-hydroxylated D3 derivatives
(6a and 7a) were both 1α-hydroxylated to afford 6b and 7b by
CYP27B1. Interestingly, the efficiency of metabolism of 25S-6a
and 25R-7a was different, and the R-isomer of 7a was ca. two-
fold more effectively metabolized compared to 6a. The order of
activity of CYP27B1 towards side-chain-hydroxylated VD3 me-
tabolites was 24R,25D3 (5a)>25D3 (1a)>25R,26D3 (7a)>
24S,25D3 (4a)>25S,26D3 (6a)@23S,25D3 (2a), 23R,25D3 (3a).
These results suggest that the 1,25R,26D3 (7b) might exist in
human serum. In addition, 1α-hydroxylated metabolites of
24,25D3 (4a and 5a) and 25,26D3 (6a and 7a) might be
produced in vivo, raising the interesting possibility that some of
them might be physiologically relevant.
· 25,26-
[1] A. Windaus, O. Linsert, Ann. Chem. 1928, 465, 148.
[2] a) R. Bouillon, W. H. Okamura, A. W. Norman, Endocr. Rev. 1995, 95,
1877; b) S. Kim, M. Yamazaki, L. A. Zella, N. K. Shevde, J. W. Pike, Mol.
Cell. Biol. 2006, 26, 6469.
[3] T. Sakaki, N. Kagawa, K. Yamamoto, K. Inouye, Front. Biosci. 2005, 10,
119–134.
[4] a) T. Sakaki, N. Sawada, Y. Nonaka, Y. Ohyama, K. Inouye, Eur. J. Biol.
Chem. 1999, 262, 43–48; b) T. Sakaki, N. Sawada, K. Komai, S. Shiozawa,
S. Yamada, K. Yamamoto, Y. Ohyama, K. Inouye, Eur. J. Biol. Chem. 2000,
267, 6158–6165; c) S. Yamada, K. Nakayama, H. Takayama, T. Shinki, Y.
Takasaki, T. Suda, J. Biol. Chem. 1984, 259, 884–889; d) S. Ishizuka, A. W.
Norman, J. Biol. Chem. 1987, 262, 7165–7170; e) O. B. Yu, L. A. Arnold,
ACS Chem. Biol. 2016, 11, 2665–2672; f) M. Kaufmann, C. Martineau, A.
Arabian, M. Traynor, R. St-Arnaud, G. Jones, J. Steroid Biochem. Mol. Biol.
2019, 188, 23.
[5] Y. Xu, T. Hashizume, M. C. Shuhart, C. L. Davis, W. L. Nelson, T. Sakaki,
T. F. Kalhorn, P. B. Watkins, E. G. Schuetz, K. E. Thummel, Mol. Pharmacol.
2006, 69, 56–65.
[6] B. Endres, S. Kato, H. F. DeLuca, Biochemistry 2000, 39, 2123–2129.
[7] Z. Wang, Y. S. Lin, X. E. Zheng, T. Senn, T. Hashizume, M. Scian, L. J.
Dickmann, S. D. Nelson, T. A. Baillie, M. F. Hebert, D. Blough, C. L. Davis,
K. E. Thummel, Mol. Pharmacol. 2012, 81, 498–509.
[8] T. Suda, H. F. DeLuca, H. K. Schnoes, Y. Tanaka, M. F. Holick, Biochemistry
1970, 9, 4776–4780.
[9] J. J. Partridge, Shian-Jan Shiuey, N. K. Chadha, E. G. Baggiolini, J. F.
Blount, M. R. Uskokovic, J. Am. Chem. Soc. 1981, 103, 1253.
[10] a) Y.-D. Guo, S. Strugnell, D. W. Back, G. Jones, Proc. Natl. Acad. Sci. USA
1993, 90, 8668–8672; b) N. Sawada, T. Sakaki, M. Ohta, K. Inouye,
Biochem. Biophys. Res. Commun 2000, 273, 977–984.
[11] a) N. Sawada, T. Sakaki, S. Kitanaka, K. Takeyama, S. Kato, K. Inouye, Eur.
J. Biochem. 1999, 265, 950–956; b) T. Sakaki, N. Sawada, K. Takeyama, S.
Kato, K. Inouye, Eur. J. Biochem. 2001, 259, 731–738.
Conclusion
[12] Synthesis of 23-hydroxylated VD3 (2, 3) with classical strategy: a) N.
Ikekawa, T. Eguchi, Y. Hirano, J. Chem. Soc. Chem. Commun. 1981, 22,
1157–1159; Synthesis of 23-hydroxylated VD3 (2, 3) using the
convergent strategy: b) F. Kawagoe, T. Sugiyama, K. Yasuda, M. Uesugi,
T. Sakaki, A. Kittaka, J. Steroid Biochem. Mol. Biol. 2019, 186, 161–168.
[13] Synthesis of 24-hydroxylated VD3 (4, 5) using the classical strategy:
a) H. Y. Lam, H. K. Schnoes, H. F. DeLuca, T. C. Chen, Biochemistry 1973,
12, 4851–4855; b) J. Redel, P. Bell, F. Delbarre, E. Kodicek, C. R. Seances
Acad. Sci. 1974, 279, 529–531; c) M. Seki, N. Koizumi, M. Morisaki, N.
Ikekawa, Tetrahedron Lett. 1975, 16, 15–18; d) S. C. Eyley, D. H. Williams,
J. Chem. Soc. Perkin Trans. 1976, I, 731–735; e) J. Redel, N. Bazely, F.
Delbarre, Y. Calando, C. R. Seances Acad. Sci. 1976, 283, 857–860; f) G. M.
Segal, I. V. Torgov, Bioorganicheskaya Khimiya 1979, 5, 1668–1676; g) H.
Takayama, M. Ohmori, S. Yamada, Facile, Tetrahedron Lett. 1980, 21,
We synthesized 23-, 24- and 26-hydroxylated VD3 derivatives 2–
7, and examined their susceptibility to C-1 hydroxylation by
CYP27B1. The 24R and 25R metabolites 5a and 7a were more
effectively hydroxylated at C1 by CYP27B1 compared to their S
diastereomers 4a and 6a, respectively. On the other hand,
CYP27B1 showed almost no activity towards either of the
diastereomers of the 23-hydroxylated derivative. This is the first
observation that CYP27B1 metabolizes 26-hydroxylated D3
derivatives 6a and 7a to the corresponding 1α-hydroxylated
ChemBioChem 2021, 22, 1–6
www.chembiochem.org
4
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cbic.202100250
ChemBioChem
5027–5028; h) M. Ishiguro, N. Koizumi, M. Yasuda, N. Ikekawa, J. Chem.
[15] a) H. T. Toh, W. H. Okamura, J. Org. Chem. 1983, 48, 1414–1417; b) S.
Hatakeyama, H. Irie, T. Shintani, Y. Noguchi, H. Yamada, M. Nishizawa,
Tetrahedron 1994, 50, 13369–13376.
[16] P. H. Lee, K. Bang, K. Lee, S. Sung, S. Chang, Synth. Commun. 2001, 31,
3781–3789.
[17] a) Y. Ogina, H. Chen, E. Manoury, T. Shibata, M. Beller, D. Lübben, K. B.
Sharpless, Tetrahedron Lett. 1991, 32, 5761–5764; b) K. B. Sharpless, W.
Amberg, Y. L. Bennani, G. A. Crispino, J. Hartung, K. S. Jeong, H. L.
Kwong, K. Morikawa, Z. M. Wang, J. Org. Chem. 1992, 57, 2768–2771.
[18] N. Saito, T. Matsunaga, T. Fujishima, M. Anzai, H. Saito, K. Takenouchi, D.
Miura, S. Ishizuka, H. Takayama, A. Kittaka, Org. Biomol. Chem. 2003, 1,
4396–4402.
Soc. Chem. Commun. 1981, 115–117; i) N. Koizumi, M. Ishiguro, M.
Yasuda, N. Ikekawa, J. Chem. Soc. Perkin Trans. 1983, I, 1401–1410; j) M.
Seki, J. Rubio-Lightbourn, M. Morisaki, N. Ikekawa, Chem. Pharm. Bull.
1973, 21, 2783–2785; Synthesis of 24-hydroxylated VD3 (4, 5) with
convergent strategy: k) W. Stepanenko, J. Wicha, Tetrahedron Lett. 1998,
39, 885–888; l) M. Odrzywolska, M. Chodynski, J. Zorgdrager, J.-P.
Van de Velde, A. Kütner, Chirality 1999, 11, 701–706; m) J. Pérez-Sestelo,
I. Cornella, O. de Uña, A. Mouriño, L. A. Sarandeses, Chem. Eur. J. 2002,
8, 2747–2752; n) I. Cornella, R. M. Suárez, A. Mouriño, J. Pérez-Sestelo,
L. A. Sarandeses, J. Steroid Biochem. Mol. Biol. 2004, 89–90, 19–23; o) C.
Fernández, Z. Gándara, G. Gómez, B. Covelo, Y. Fall, Tetrahedron Lett.
2007, 48, 2939–2942; p) Y. Ono, H. Watanabea, I. Tairab, K. Takahashi, J.
Ishihara, S. Hatakeyama, N. Kubodera, Steroids 2006, 71, 529–540.
[14] Synthesis of 26-hydroxylated VD3 (6, 7) using the classical strategy: a) N.
Koizumi, M. Morosaki, N. Ikekawa, Tetrahedron Lett. 1978, 32, 2899–
2902; b) R. Barner, J. Hübscher, J. J. Daly, P. Schönholzer, Helv. Chim.
Acta 1981, 64, 915–938; and ref. [8]; Synthesis of 26-hydroxylated VD3
(6, 7) using the convergent strategy: c) R. Sigüeiro, P. González-
Berdullas, J. Loureiro, A. Mouriño, M. A. Maestro, J. Steroid Biochem. Mol.
Biol. 2018, 177, 247–249.
[19] A. Nagata, Y. Akagi, S. S. Masoud, M. Yamanaka, A. Kittaka, M. Uesugi, M.
Odagi, K. Nagasawa, J. Org. Chem. 2019, 84, 7630–7641.
Manuscript received: May 25, 2021
Revised manuscript received: July 7, 2021
Accepted manuscript online: July 11, 2021
Version of record online: ■■■, ■■■■
ChemBioChem 2021, 22, 1–6
www.chembiochem.org
5
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPERS
R. Sakamoto, A. Nagata, H. Ohshita,
Y. Mizumoto, M. Iwaki, Dr. K.
Yasuda*, Prof. T. Sakaki*, Dr. K.
Nagasawa*
1 – 6
Chemical Synthesis of Side-Chain-
Hydroxylated Vitamin D3 Deriva-
tives and Their Metabolism by
CYP27B1
The role of CYP27B1 in C1 hydroxyla-
tion of a series of side-chain-hydroxy-
lated D3 derivatives was explored. We
found that the 24R and 26R metabo-
lites were more effectively hydroxy-
lated at C1 by CYP27B1 compared to
the corresponding S diastereomers.
However, CYP27B1 showed almost no
activity towards either of the diaster-
eomers of the 23-hydroxylated deriva-
tive. This is the first report to show
that CYP27B1 metabolizes 26-hy-
droxylated D3, converting 25,26D3 to
the corresponding 1α-hydroxylated
compound, 1,25,26D3.