An expeditious route to 1α,25-dihydroxyvitamin D3 and its analogues by an aqueous tandem palladium-catalyzed a-ring closure and suzuki coupling to the C/D unit

Article abstract of DOI:10.1002/chem.200902972

Chemical equation presented Daily vitamins: A mild, general, and highly stereoselective Pd⁰-catalyzed cascade to the triene system of the hoi mone 1α,25-dihydroxyvitamin D₃ and six representative analogues is reported. The intramolecular cyclization of an enol-triflate (lower fragment) followed in situ by Suzuki Miyaura coupling with an alkenyl boronic ester (upper fragment, also efficiently prepared by Pd⁰-catalyzed coupling) in equimolar amounts under protic conditions is ideal for the preparation of small amounts of new vitamin D analogues for biological testing (see scheme).

Full text of DOI:10.1002/chem.200902972

DOI: 10.1002/chem.200902972
An Expeditious Route to 1a,25-Dihydroxyvitamin D3 and Its Analogues by
an Aqueous Tandem Palladium-Catalyzed A-Ring Closure and Suzuki
Coupling to the C/D Unit
Pranjal Gogoi, Rita Sigꢀeiro, Silvina Eduardo, and Antonio MouriÇo*[a]
Dedicated to Professor Dieter Seebach
1a,25-Dihydroxyvitamin D3 (1), the hormonally active
metabolite of the seco-steroid vitamin D3, interacts with the
vitamin D nuclear receptor (VDR)[1] to initiate a cascade of
events that ultimately controls mineral homeostasis and a
multitude of cellular processes
palladium-catalyzed route introduced by Trost and co-work-
ers (Scheme 1, route B).[6] These methods have practical
drawbacks in that they either require an excess of the lower
(A ring) fragment (for small-scale work) or elevated temper-
including differentiation, anti-
proliferation, growth, apoptosis,
angiogenesis, and immunomo-
dulation.[2] Unfortunately, the
therapeutic applications of 1 in
pharmacological doses to cor-
rect dysfunction of one or more
of these processes are severely
limited by its potent calcemic
effects.[3] Efforts to develop an-
alogues with selectively reduced
calcemic effects for treatment
of, for example, cancer and skin
diseases or with selective activi-
ty on bone formation have led
Scheme 1. Synthetic routes to vitamin D compounds. M=metal, SC=side chain, Si=protecting group.
to more than 3000 synthetic analogues being tested, al-
though only a few have reached the pharmaceutical market
or advanced clinical trials.[4]
The most useful convergent methods to synthesize the
triene moiety in vitamin D analogues include the Wittig–
Horner approach devised by Lythgoe and developed by the
Hoffmann La Roche group (Scheme 1, route A)[5] and the
atures that equilibrate vitamin D with its previtamin D
form. We have recently also developed a palladium-cata-
lyzed process for the construction of the vitamin D triene
that couples enol–triflates with alkenyl zinc intermediates
(Scheme 1, route C), but this method was still limited by
problems with reproducibility on the small-scale and re-
quires two equivalents of the upper (C/D) fragment even to
afford moderate yields.[7,8]
Prompted by these considerations and the continuing
need for simple, small-scale access to an array of test com-
pounds for rapid screening to generate clinical candidates,
we envisaged the possibility of employing alkenyl-boronic
esters (2, M=B(OR)2; Scheme 1), instead of the corre-
[a] Dr. P. Gogoi, R. Sigꢀeiro, S. Eduardo, Prof. Dr. A. MouriÇo
Departamento de Quꢁmica Orgꢂnica
Universidad de Santiago de Compostela
y Unidad Asociada al CSIC
Avda de las Ciencias s/n
15782 Santiago de Compostela (Spain)
Fax : (+34)981-595012
sponding alkenyl zinc intermediates of our earlier work.[9]
A
Suzuki coupling with the palladium intermediate resulting
from the initial cyclization of enol–triflate 3a[10] as a precur-
sor of the A-ring fragment would construct the triene unit
E-mail: antonio.mourino@usc.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902972.
1432
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 1432 – 1435

COMMUNICATION
Table 1. Pd0-catalyzed preparation of alkenyl boronates 2 from alkenyl
bromides 5.[a]
stereoselectively in one pot. We can now report that this
strategy does indeed circumvent the problems associated
with the previous synthetic approaches and provides a gen-
eral method for the small-scale preparation of a wide variety
of vitamin D analogues in a practical, economical, and re-
producible fashion.
To validate the approach, we were first concerned with
the synthesis of the natural hormone 1. Our synthesis com-
mences with the known alkenyl bromide 5a that is derived
from ketone 4a in ꢀ55% yield by using Trost conditions
(Scheme 2).[6] The moderate yield in the Wittig reaction
Entry Substrate Catalyst
Ligand Solvent
T
t
Yield[b]
[8C] [h] [%]
1
5a
5a
5a
5a
5a
5a
5a
5a
5b
5c
G
E
toluene 50
6
6
0[c]
35
2[d]
3
N
G
–
DMSO 80
dioxane 80
U
PCy3
12 23
4[d]
5
DMSO RT 12 28
–
–
dioxane 80
DMSO 80
dioxane 80
DMSO 80
DMSO 80
DMSO 80
12 28
6
6
6
3
3
3
68
60
91
92
80
7[d]
8[d]
9[d]
10[d]
[a] dba=trans,trans-dibenzylideneacetone. dppf=1,1’-bis(diphenylphos-
phino) ferrocene. SK-CC02A=2-(dimethylaminomethyl)ferrocen-1-ylpal-
ladium(II) chloride dinorbornylphosphine complex. PCy3 =tricyclohexyl-
phosphine. DMSO: dimethylsulfoxide. [b] Isolated yields. [c] KOPh
(1.5 equiv). [d] Catalyst complexed to CH2Cl2.
worked equally well with alkenyl bromides 5b,c to give bor-
onate esters 2b,c (Table 1, entries 9 and 10).
With 2a and 3a in hand, we proceeded to the critical Pd-
catalyzed tandem cyclization–Suzuki coupling process to the
triene system of the natural hormone 1. Treatment of a mix-
ture of 2a (0.23 mmol) and 3a (0.27 mmol) in aqueous
Scheme 2. Synthesis of alkenyl bromides 5a–c. TES=SiEt3. a) Trost con-
ditions: Ph3P+CH2BrBrꢁ, NaHMDS (HMDS=hexamethyldisilazane),
THF; b) new conditions: Ph3P+CH2BrBrꢁ washing with toluene/CH2Cl2
under ultrasonication, then KOtBu, ꢁ15!08C.
K3PO4 (2m)/THF with a catalytic amount of [PdCl2ACHTUNGTRENNUNG(Ph3P)2]
(5 mol%) at RT for 1 h delivered, after standard desilyla-
prompted us to scrutinize this important step, and eventually
we found that purification of the phosphonium salt by wash-
ing with toluene/CH2Cl2 under sonication, followed by gen-
eration of the ylide with KOtBu in toluene (instead of the
previously used NaHMDS in THF) and subsequent reaction
with ketones 4a–c, which incorporate diverse functionally at
the side chains, provided the corresponding alkenyl bro-
mides 5a–c in good yields.
tion, the natural hormone 1 in 81% yield (Scheme 3). Any
The alkenyl bromide 5a could be converted to boronate
2a by lithiation–transmetallation according to a procedure
reported by Sato and co-workers,[11] but the fact that this re-
action sequence is not compatible with functionalities such
as the hydroxy and ester groups present in alkenyl bromides
5b and 5c led us to explore better procedures for this trans-
formation. We first studied the conditions developed by
Miyaura and co-workers that were previously used to pre-
pare arylboronic esters by a Pd0-catalyzed cross-coupling re-
action between bis(pinacolato) diboron and haloarenes.[12]
However, these conditions afforded the desired boronate 2a
in low yields (Table 1, entries 1–3). After several attempts
using different catalysts and ligands, we finally found that
the use of tricyclohexylphosphine (PCy3) as the ligand to-
gether with the Miyaura catalyst provided boronate ester 2a
in excellent yield (Table 1, entry 8), and furthermore
Scheme 3. Synthesis of the natural hormone 1: a) [PdCl
(5 mol%), 2m K3PO4 (aq)/THF, RT, 1 h; b) nBu4NF, THF, RT, 24 h.
TIPS=Si(iPr)3.
2ACHTUNGTRENNUNG(Ph3P)2]
AHCTUNGTRENNUNG
problems associated with excess of one or the other synthet-
ic building block, exclusion of moisture, and/or elevated
temperature are dismissed in this remarkable reaction, and
the way is opened up for making analogues with modified
structures under these standard conditions.[13]
The 6-methyl-derivative 6a, which could not be prepared
by routes A[14] or B,[7] was selected as a test for our method-
ology (Scheme 4). Gratifyingly, enyne 3b[7] reacted with 2b
under the standard conditions to give the target compound
6a in 73% yield after desilylation. The mildness of the
Chem. Eur. J. 2010, 16, 1432 – 1435
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1433

A. MouriÇo et al.
was chosen as the test example of a vitamin D analogue
modified at the C ring, since the original synthesis of ana-
logues with a b-substituent at C-11 in Vandewalleꢄs labora-
tory provided consistently low yields (20–40%) by the
Wittig–Horner approach (Scheme 1, route A).[17]
In summary, a concise, general, and stereoselective entry
to the vitamin D triene system of the natural hormone 1 and
six representative analogues has been achieved by an effi-
cient strategy featuring a highly stereoselective intramolecu-
lar cyclization of an enol triflate (A ring or lower fragment)
followed in situ by a Suzuki–Miyaura coupling of the result-
ing palladium intermediate with an alkenyl boronic ester
(CD side-chain upper fragment). The method employs equi-
molar quantities of both fragments under protic conditions
and can be used for the preparation of small amounts of
new vitamin D analogues for biological testing. Further syn-
thetic studies pertaining to even more challenging vitamin D
analogues are underway in our laboratory.
Scheme 4. Synthesis of 1a-OH-6-methyl-vitamin D3: a) standard condi-
tions; b) nBu4NF, THF, RT, 2 h (73% yield for the two steps); c) equili-
bration time in CD3OD at RT and ratio vitamin D/previtamin D deter-
1
mined by H NMR spectroscopy: (3 d, 6a/6b=1/1; 15 d, 6a/6b=1/60).
Acknowledgements
method is strikingly demonstrated by the fact that analogue
6a equilibrates on standing in CD3OD with its previtamin D
form 6b.
We thank the Spanish Ministry of Education and Science (grant no.
SAF2007-67205) and Xunta de Galicia (projects INCITE08PXIB-
209130PR and ACEUIC-2006/XA050) for financial support and Dishman
Netherlands for the gift of vitamin D2. P.G. thanks Xunta de Galicia for a
fellowship. R.S. and S.E. thank the Spanish MEC for predoctoral fellow-
ships.
The versatility and flexibility of the new route is illustrat-
ed by the efficient construction of the triene moieties of the
variety of vitamin D analogues shown in Scheme 5. As rep-
resentative examples of analogues modified at the A ring,
we prepared 7[15] (78% yield) (a member of the superagonist
set of analogues) and 8[16] (66%) (which has a congested A-
ring fragment). For examples with other functionalities at
the side chain we chose compounds 9 and 10. Compound 11
Keywords: homogeneous catalysis
· natural products ·
palladium · synthesis design · vitamins
[1] a) N. Rochel, J. M. Wurtz, A. Mitschler, B. Klaholz, D. Moras, Mol.
Cell 2000, 5, 173–179; b) S. Christakos, P. Dhawan, B. Benn, A.
Porta, M. Hediger, G. T. Oh, E. B. Jeung, Y. Zhong, D. Ajibade, K.
Dhawan, S. Joshi, Ann. N. Y. Acad. Sci. 2007, 1116, 340–348.
[2] a) Vitamin D (Eds.: D. Feldman, F. H. Glorieux, J. W. Pike). Elsevi-
er, New York, 2005; b) “Proceedings of the 12th and 13th Workshop
on Vitamin D” (Eds.: R. Bouillon, A. W. Norman, J. R. Pasqualini):
J. Steroid Biochem. Mol. Biol. 2007, 103, 201–822; 2004, 89–90, 1–
633; and the previous 11 volumes in these series.
[3] a) M. S. Stein, J. D. Wark, Expert Opin. Invest. Drugs 2003, 12, 825–
840; b) K. K. Deeb, D. L. Trump, C. S. Johnson, Nat. Rev. Cancer
2007, 7, 684–700.
[4] a) M. M. Kabat, R. Radinov, Curr. Opin. Drug Discov. Devel. 2001,
4, 808–833; b) A. J. Brown, Am. J. Kidney Dis. 2001, 38, S3–S19;
c) C. Carlberg, A. MouriÇo, Expert Opin. Ther. Pat. 2003, 13, 761–
772; d) R. Bouillon, W. H. Okamura, A. W. Norman, Endocr. Rev.
1995, 16, 200–257.
[5] E. G. Baggiolini, J. A. Iacobelli, B. M. Hennessy, A. D. Batcho, J. F.
Sereno, M. R. Uskokovic, J. Org. Chem. 1986, 51, 3098–3108.
[6] B. M. Trost, J. Dumas, M. Villa, J. Am. Chem. Soc. 1992, 114, 9836–
9845.
[7] C. Gꢅmez-Reino, C. Vitale, M. Maestro, A. MouriÇo, Org. Lett.
2005, 7, 5885–5887.
[8] For selected reviews on the synthesis of vitamin D compounds, see:
a) G.-D. Zhu, W. H. Okamura, Chem. Rev. 1995, 95, 1877–1952;
b) S. Krause, H.-G. Schmalz, Organic Synthesis Highlights (Ed.: H.-
G. Schmalz), Wiley-VCH, Weinheim, 2000, pp. 212–217; c) G. H.
Posner, M. Kahraman, Eur. J. Org. Chem. 2003, 3889–3895.
[9] a) A. M. Garcꢁa, J. L. MascareÇas, L. Castedo, A. MouriÇo, J. Org.
Chem. 1997, 62, 6353–6358; b) A. M. Garcꢁa, J. L. MascareÇas, L.
Scheme 5. Illustrative examples of vitamin D analogues modified at the
A ring, C ring, and side chain. Yields for the two steps, Pd-catalyzed cyc-
lization-coupling and desilylation, are shown in parentheses. No desilyla-
tion was carried out for examples 9, 10, and 11. TBS=SiMe2tBu.
1434
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 1432 – 1435

1a,25-Dihydroxyvitamin D3 and Its Analogues
COMMUNICATION
Castedo, A. MouriÇo, Tetrahedron Lett. 1995, 36, 5413–5416; c) J. L.
MascareÇas, A. M. Garcꢁa, L. Castedo, A. MouriÇo, Tetrahedron
Lett. 1992, 33, 7589–7592.
[15] Compound 7 was synthesized in 67% yield by the Trost method: K.
Konno, S. Maki, T. Fujishima, Z.-P. Liu, D. Miura, M. Chokki, H.
Takayama, Bioorg. Med. Chem. Lett. 1998, 8, 151–156.
[10] A. MouriÇo, M. Torneiro, C. Vitale, S. Fernꢂndez, J. Pꢆrez-Sestelo,
S. Anne, C. Gregorio, Tetrahedron Lett. 1997, 38, 4713–4716.
[11] T. Hanazawa, A. Koyama, K. Nakata, S. Okamoto, F. Sato, J. Org.
Chem. 2003, 68, 9767–9772.
[16] Compound 8 was synthesized by the Trost method in 63% yield as a
mixture with its 3-epimer: T. Fujishima, A. Kittaka, K. Yamaoka, K.
Takeyama, S. Kato, H. Takayama, Org. Biomol. Chem. 2003, 1,
1863–1869.
[12] T. Ishiyama, M. Murata, N. Miyaura, J. Org. Chem. 1995, 60, 7508–
7510.
[13] For reviews on Suzuki–Miyaura coupling reactions, see: a) N.
Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483; b) N.
Miyaura, J. Organomet. Chem. 2002, 653, 54–57.
[17] R. Bouillon, K. Allewaert, J. P. T. M. van Leeuwen, B.-K. Tan, D. Z.
Xiang, P. De Clercq, M. Vandewalle, H. A. P. Pols, M. P. Bos, H. van
Baelen, J. C. Birkenhꢇger, J. Biol. Chem. 1992, 267, 3044–3051.
Received: September 28, 2009
[14] Unpublished results from our laboratory.
Published online: December 28, 2009
Chem. Eur. J. 2010, 16, 1432 – 1435
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1435