Synthesis of novel carborane-hybrids based on a triazine scaffold for boron neutron capture therapy

Article abstract of DOI:10.1055/s-2004-822897

Cyanuric chloride has been used as a scaffold for the synthesis of potential agents for BNCT containing a carborane cluster, a sugar moiety as hydrophilic arm and an amino acid for the conjugation with bioactive molecules; this approach is, in principle, particularly suitable for a combinatorial approach.

Full text of DOI:10.1055/s-2004-822897

LETTER
1007
Synthesis of Novel Carborane-hybrids Based on a Triazine Scaffold for Boron
Neutron Capture Therapy
T
riazine Scaf
f
i
f
or
l
BNCTvia Ronchi,a Davide Prosperi,b Federica Compostella,c Luigi Panza*a
a
b
c
DISCAFF, Università del Piemonte Orientale, viale Ferrucci 33, 28100 Novara, Italy
Istituto di Scienze e Tecnologie Molecolari – CNR, via Golgi 19, 20133 Milano, Italy
Dipartimento dii Chimica, Biochimica e Biotecnologie per la Medicina, Università di Milano, via Saldini, 50, 20133 Milano, Italy
Fax +39(0321)375821; E-mail: panza@pharm.unipmn.it
Received 12 December 2003
to the different reactivity of the chlorine atoms, allows a
sequential introduction of various substituents, opening
the way to introduce diversity into the products. For this
reason, cyanuric chloride or triazine derivatives are often
used in combinatorial synthesis.8 It has to be noted that, to
the best of our knowledge, only one example of triazine
derived carboranes has appeared in the literature.9 The au-
thors showed that it is possible to directly join the carbo-
rane cage to a triazine core, but the other substituents
reported were only simple secondary amines. It is remark-
able that, in their reactions, the carborane cage seems to be
quite resistant to the presence of secondary amines, as
they did not observe the degradation of the closo form
with formation of anionic nido derivatives.
Abstract: Cyanuric chloride has been used as a scaffold for the syn-
thesis of potential agents for BNCT containing a carborane cluster,
a sugar moiety as hydrophilic arm and an amino acid for the con-
jugation with bioactive molecules; this approach is, in principle,
particularly suitable for a combinatorial approach.
Key words: carborane cluster, glycosides, amino acids, boron, me-
dicinal chemistry
Boron neutron capture therapy (BNCT) is a therapy for
the treatment of tumors based on the selective irradiation
of molecules containing 10B atoms with a thermal neu-
trons beam.1 This nuclide shows a large capture cross sec-
tion relative to the more abundant endogenous nuclei (1H,
12C, 16O, 31P, 14N).2 In order to be therapeutically useful,
an ideal boronated candidate should have the following
properties: high tumor targeting selectivity; low cytotox-
icity; high water solubility required for intra-arterious ad-
ministration of the BNCT agent; high uptake by cancer
cells.3
In order to verify the possibility of generating differently
functionalised triazines, we prepared 2-carboranyl ethyl-
amine as well as protected glucose and lactose aminopro-
pyl glycosides.
Among the different methods developed for the synthesis
of aminoalkyl carboranes,10 the preparation of the deriva-
tive 2 was performed according to the procedure suggest-
ed by Soloway and co-workers,11 with a modification of
the azide reduction step, carried out without acid added.
Tosylation of 3-butyn-1-ol, carborane formation by treat-
ment with decaborane-acetonitrile adduct and substitution
with azide anion gave 2-carboranyl azido ethane 1. The
azido function was converted into the corresponding ami-
no group with H2-Pd/C in dry THF. Excess of cyanuric
chloride and Hünigs’ base were directly added into the re-
action mixture containing the freshly prepared amine at
room temperature,12 affording smoothly the desired
monosubstituted triazine 3 in good yields (Scheme 1).
As boron moiety, our attention has been focused mainly
on carboranes. Such boron containing structures allow to
obtain chemical constructions that carry a large number of
boron atoms per molecule.4 On the other hand, carboranes
are highly hydrophobic compounds, thus requiring them
to be conjugated with hydrophilic counterparts in order to
have structures with adequate water solubility for their ad-
ministration. Among the different classes of molecules
able to give the desired solubility properties to the carbo-
rane containing derivatives, carbohydrates are particular-
ly interesting compounds5 as they could also target
specific receptors found on the surface of tumor cells, and
usually show low toxicities. Moreover, the introduction of
an amino acid unit could enable the incorporation of car-
boranes into peptides.6 To date, only a few examples of
carborane derivatives containing both a sugar and an ami-
no acid have appeared in the literature, and their syntheses
are usually quite laborious.7
The aminopropyl glycosides were obtained by conven-
tional Lewis acid catalysed glycosylation of commercial-
ly available peracetylated b-D-glucose (4a) and b-D-
lactose (4b), with 3-benzyloxycarbonylaminopropanol.13
Although the yields were not very high, each product was
obtained in a single step. Benzyloxycarbonyl protecting
group of compounds 5a and 5b was removed by catalytic
hydrogenolysis. The reaction mixture containing the free
amino derivative 6a (or 6b, respectively) was added drop-
wise to a solution of compound 3 and Hünigs’ base in
THF at 0 °C and the resulting mixture was allowed to
warm to room temperature. We were pleased to observe
the clean formation of the disubstituted triazines 7a and
We have chosen to exploit cyanuric chloride as a scaffold
for the easy introduction of a carborane, a sugar and an
amino acid onto the same molecule. Such a scaffold, due
SYNLETT 2004, No. 6, pp 1007–1010
0
6.
0
5.
2
0
0
4
Advanced online publication: 25.03.2004
DOI: 10.1055/s-2004-822897; Art ID: G35303ST
© Georg Thieme Verlag Stuttgart · New York

1008
S. Ronchi et al.
LETTER
7b both with glucose and lactose derivatives. Under these respectively in 70–75% yield. The newly introduced
conditions the carborane cage demonstrated complete carboxylic function would make this molecule ready for
stability in the presence of free amino groups.
conjugation with biopolymers.
The displacement of the third chlorine atom is usually Encouraged by these results, we decided to try to perform
more difficult: in our hand, any attempt to attach the pri- the third substitution reaction with cysteine. This amino
mary e-amino group of the side chain of the Z-a-N-L- acidic moiety, allowed us to generate a compound, which
lysine invariably failed, even after refluxing in acetonitrile could not only be easily bound to a protein, but also insert-
for two days.14 So we decided to exploit a better nucleo- ed into a synthetic peptide.
phile such as a thiolate anion for this reaction. In order to
verify the reaction conditions we initially employed the
thiolate of methyl thioglycolate, generated with sodium
hydride in dry acetonitrile. After addition of compound 7a
or 7b, the reaction mixture was refluxed overnight. Aque-
ous work-up and purification by flash chromatography af-
forded the desired trisubstituted compounds 8a and 8b,
Thus, the reaction was repeated essentially under the same
conditions employed for methyl thioglycolate, using N-
Boc-L-cysteine and 2 equivalents of sodium hydride.
After refluxing overnight in dry acetonitrile, a 5% HCl so-
lution was added dropwise until pH 3 was reached, then
the mixture was diluted with dichloromethane and ex-
tracted as previously. Again, we were pleased to observe
R1
O
OAc
OAc
O
O
O
•
iii
O
O
O
O
RO
AcO
RO
AcO
O
O
NHCbz
O
O
O
OAc
OAc
OAc
4a,b
1 R1 = N3
2 R1 = NH2
i
5a,b
Cl
N
Cl
iv
ii
N
N
OAc
O
RO
NH
O
NH2
AcO
O
O
O
O
•
OAc
6a,b
O
O
O
O
3
O
O
v
OAc
H
N
O
RO
N
Cl
O
AcO
OAc
N
N
NH
O
O
O
O
O
•
7a,b
O
O
O
O
O
OAc
vi
H
O
RO
N
N
S
COOCH3
O
AcO
OAc
N
N
NH
O
O
O
8a,b
O
= CH
= C
O = BH
O
•
O
O
O
O
O
•
OAc
NHBoc
COOH
H
N
a R = OAc
O
RO
N
S
O
AcO
vii
OAc
OAc
OAc
b R =
N
N
NH
O
AcO
OAc
O
O
O
9a,b
O
O
•
O
O
O
O
O
Scheme 1 (i) H2, cat. Pd/C, dry THF; (ii) DIPEA, 3 equiv cyanuric chloride, 85% for two steps; (iii) 3-benzyloxycarbonylaminopropanol,
BF3·Et2O, dry CH2Cl2, 47% for 5a, 41% for 5b; (iv) H2, cat. Pd/C, dry THF; (v) DIPEA, dry THF, 0 °C, then r.t., 88% for 7a, 60% (not opti-
mised) for 7b; (vi) NaH, methyl thioglycolate, dry MeCN, 80 °C, 75% for 8a, 71% for 8b; (vii) 2 equiv NaH, N-Boc-cysteine, dry MeCN, 80 °C,
72% for 9a, 70% for 9b.
Synlett 2004, No. 6, 1007–1010 © Thieme Stuttgart · New York

LETTER
Triazine Scaffold for BNCT
1009
After the addition, the ice bath was removed and the mixture was al-
lowed to warm to r.t. for 15 min. A solution of 7a,b (0.14 mmol) in
dry MeCN (3.5 mL) was added. The resulting mixture was refluxed
overnight under argon, then cooled to r.t. and concentrated to 1 mL
volume. To this mixture H2O (5 mL) was added followed by 5%
HCl until pH 3 was reached. The mixture was diluted with CH2Cl2
(15 mL) and extracted. The organic layer was washed with H2O,
then dried over Na2SO4, filtered and concentrated. The crude
product was purified flash chromatography (EtOAc–MeOH, 9:1),
furnishing 9a and 9b in 72% (91 mg) and 70% (124 mg) yields,
respectively.
the formation of the expected compounds 9a and 9b in
comparable yields with respect to the previous reaction. In
the 11B NMR spectra, compounds 8a,b and 9a,b showed
signals in the range d = –2 to –15 ppm,15 while no peaks
were observed in the range d = –30 to –40 ppm, diagnostic
for the nido-carborane.16 MALDI-TOF mass experiments
confirmed the presence of the intact carborane cage.
In conclusion, this paper describes a simple and effective
way to introduce a carborane, a sugar and a carboxylic
acid or an amino acid onto a triazine scaffold. Such com-
pounds open the possibility to generate easily a remark-
able diversity employing a combinatorial approach. It is in
fact possible to introduce various oligosaccharidic
structures as well as to use this class of derivatives for the
synthesis of biologically relevant peptides.
Acknowledgement
We thank MIUR (COFIN 2001), Università del Piemonte Orientale
and CNR for financial support.
Work is in progress to extend the scope of the procedure References
to more complex derivatives directly involved in tumor
cell surface recognition and endocytosis phenomena.
(1) Locher, G. L. Am. J. Roentgenol. 1936, 36, 1.
(2) For some general reviews see e.g.: (a) Bregadze, V. I.
Chem. Rev. 1992, 92, 209. (b) Valliant, J. F.; Guenther, K.
J.; King, A. S.; Morel, P.; Schaffer, P.; Sogbei, O. O.;
Stephenson, K. A. Coord. Chem. Rev. 2002, 232, 173.
(c) Yamamoto, Y. Pure Appl. Chem. 1991, 63, 423.
(d) Barth, R. F.; Soloway, A. H.; Fairchild, R. G. Cancer
Res. 1990, 50, 1061.
(3) (a) Soloway, A. H.; Tjarks, W.; Barnum, B. A.; Rong, F.-G.;
Barth, R. F.; Codogni, I. M.; Wilson, J. G. Chem. Rev. 1998,
98, 1515. (b) Hawthorne, M. F. Angew. Chem., Int. Ed. Engl.
1993, 32, 950.
(4) (a) See ref. 3a. (b) Morin, C. Tetrahedron 1994, 50, 12521.
(5) For general information on oligosaccharides synthesis and
carbohydrates biology see e.g.: (a) Seeberger, P. H.; Haase,
W.-C. Chem. Rev. 2000, 100, 4349. (b) Nishimura, S. Curr.
Opin. Chem. Biol. 2001, 5, 325. (c) Sears, P.; Wong, C.-H.
Science 2001, 291, 2344. (d) Dove, A. Nat. Biotechnol.
2001, 19, 913. (e) Barchi, J. J. Jr. Curr. Pharm. Des. 2000,
6, 485.
(6) Valliant, J. F.; Guenther, K. J.; King, A. S.; Morel, P.;
Schaffer, P.; Sogbei, O. O.; Stephenson, K. A. Coord. Chem.
Rev. 2002, 232, 173.
(7) (a) Thimon, C.; Panza, L.; Morin, C. Synlett 2003, 1399.
(b) Das, B. C.; Das, S.; Li, G.; Bao, W.; Kabalka, G. W.
Synlett 2001, 1419.
(8) (a) Gustafson, G. R.; Baldino, C. M.; O’Donnell, M.-M.;
Sheldon, A.; Tarsa, R. J.; Verni, C. J.; Coffen, D. L.
Tetrahedron 1998, 54, 4051. (b) Xia, Y.; Mirzai, B.;
Chackalamannil, S.; Czarniecki, M.; Wang, S.; Clemmons,
A.; Ahn, H.; Boykow, G. C. Bioorg. Med. Chem. Lett. 1996,
6, 919. (c) Ichihara, K.; Naruta, Y. Chem. Lett. 1995, 631.
(9) Lee, C.-H.; Lim, H.-G.; Lee, J.-D.; Lee, Y.-J.; Ko, J.;
Nakamura, H.; Kang, S. O. Appl. Organometal. Chem. 2003,
17, 539.
(10) (a) Maderna, A.; Herzog, A.; Knobler, C. B.; Hawtorne, M.
F. J. Am. Chem. Soc. 2001, 123, 10423. (b) Lee, J.-D.; Lee,
Y.-J.; Jeong, H.-J.; Lee, J. S.; Lee, C.-H.; Ko, J.; Kang, S. O.
Organometallics 2003, 22, 445. (c) Malmquist, J.; Sjöberg,
S. Inorg. Chem. 1992, 31, 2534.
Preparation of Compound 3
Compound 1 (516 mg, 2.42 mmol) was dissolved in 35 mL of dry
THF. Catalytic hydrogenation (Pd/C, 20 mg) furnished compound
2 after 2.5 h. Disappearance of 1 was monitored by TLC (petroleum
ether/EtOAc 8:2). Hünig’s base (422 mL, 2.42 mmol) and cyanuric
chloride (1339 mg, 7.26 mmol) were added directly to the reaction
mixture at r.t. under nitrogen. After 14 h, the mixture was diluted
with THF, filtered over celite and the solvent evaporated. The resi-
due was purified by flash chromatography (petroleum ether–
EtOAc, 85:15) giving 690 mg of 3 as a white solid (85% yield).
General Procedure for the Preparation of Compounds 7a and
7b
(Z)-aminopropyl glycoside 5a or 5b (0.90 mmol) was dissolved in
dry THF (25 mL) A catalytic amount of Pd on activated charcoal
was added and the amino group was deprotected under hydrogen at-
mosphere (3 h). The mixture containing the free amine 6a (or 6b re-
spectively) was transferred into a solution of 3 (300 mg, 0.90 mmol)
and Hünig’s base (156 mL, 0.90 mmol) in THF (15 mL) at 0 °C. The
mixture was allowed to warm to r.t. and stirred overnight, then di-
luted with THF and filtered over celite. The solvent was evaporated
and the brown solid purified by flash chromatography (petroleum
ether–EtOAc, 1:1 for 7a and 3:7 for 7b). Compound 7a and 7b were
obtained as white solids in 88% (554 mg) and 60% (533 mg, not op-
timised) yields, respectively.
General Procedure for the Preparation of Compounds 8a and
8b
95% NaH (7 mg, 0.28 mmol) was added to a solution of methyl
thioglycolate (27 mL, 0.28 mmol) in 2 mL of dry MeCN in an ice
bath. After the addition, the ice bath was removed and the mixture
was allowed to warm to r.t. for 15 min. A solution of 7a,b (0.14
mmol) in dry MeCN (3 mL) was added. The resulting mixture was
refluxed overnight under argon, then diluted with CH2Cl2 and H2O
and extracted. The organic layer was dried over Na2SO4, filtered
and concentrated. The crude product was purified flash chromatog-
raphy (petroleum ether–EtOAc, 1:1 for 8a and 4:7 for 8b), giving
8a and 8b in 75% (82 mg) and 71% (76 mg) yields, respectively.
(11) Wilson, J. G.; Anisuzzaman, A. K. M.; Alam, F.; Soloway,
A. H. Inorg. Chem. 1992, 31, 1955.
(12) It is well known that the presence of the carboranyl moiety
has a strong electron withdrawing effect, which induces a
deactivation of the amine nucleofilicity. See e.g.: Oliva, J.
M.; Viñas, C. J. Mol. Struct. 2000, 556, 33; and references
cited therein.
General Procedure for the Preparation of Compounds 9a and
9b
95% NaH (15 mg, 0.57 mmol) was added to a solution of N-Boc-
cysteine (63 mg, 0.28 mmol) in 2 mL of dry MeCN in an ice bath.
Synlett 2004, No. 6, 1007–1010 © Thieme Stuttgart · New York

1010
S. Ronchi et al.
LETTER
(13) Mereyala, H. B.; Gurrala, S. R. Carbohydr. Res. 1998, 307,
351.
C32H56B10N6O14S: 889.0; found: 928.44 [M + K]+; 12b calcd
for C44H72B10N6O22S: 1177.2; found: 1216.3 [M + K]+. 11
B
(14) Reaction conditions: Z-a-N-L-lysine, 2 equiv DIPEA,
MeCN, 15 min, r.t.; then, 7a or 7b, 48 h, 80 °C.
(15) Selected data for final compounds. MALDI-MS: 8a calcd
for C27H47B10N5O12S: 773.9; found: 796.0 [M + Na]+, 812.8
[M + K]+; 8b calcd for C39H63B10N5O20S: 1062.1; found:
1085.2 [M + Na]+, 1101.2 [M + K]+; 9a calcd for
NMR (CD3OD): 8a –3.62 (2 B), –10.54 (4 B), –12.36 (4 B);
8b –3.59 (2 B), –10.47 (4 B), –12.22 (4 B); 9a –3.55 (2 B),
–10.50 (4 B), –12.25 (4 B); 9b –3.64 (2 B), –10.45 (4 B),
–12.18 (4 B).
(16) Yoo, J.; Hwang, J.-W.; Do, Y. Inorg. Chem. 2001, 40, 568.
Synlett 2004, No. 6, 1007–1010 © Thieme Stuttgart · New York