Simple, high-yield synthesis of polyhedral carborane amino acids

Full text of DOI:10.1021/ja9534260

J. Am. Chem. Soc. 1996, 118, 1223-1224
1223
Scheme 1a
Simple, High-Yield Synthesis of Polyhedral
Carborane Amino Acids
Stephen B. Kahl* and Ramesh A. Kasar
Department of Pharmaceutical Chemistry
a Reagents: (a) (i) n-BuLi; (ii) CH3Li, CH3ONH2‚HCl; (iii) H2O;
(b) benzophenone/PTSA; (c) (i) n-BuLi; (ii) ClCOOCH3; (d) (i) aqueous
KOH; (ii) Pd-C/H2.
UniVersity of California, San Francisco
513 Parnassus AVenue
San Francisco, California 94143-0446
Scheme 2a
ReceiVed October 11, 1995
Boron neutron capture therapy (BNCT) is a form of binary
cancer therapy that offers the potential of delivering spatially
selective, high linear energy transfer radiation to the target cells
while sparing surrounding normal tissue.1 BNCT is based on
the nuclear reaction that occurs when the stable isotope 10B
absorbs a thermal neutron to yield an R particle and a recoil
7Li nucleus. These highly energetic particles have mean free
paths of 9 and 5 µm, respectively, resulting in confinement of
their kinetic energies to approximately one cell diameter. The
neutrons used for activation of this process are of subionizing
energy (<0.025 eV), and radiation damage to cells that do not
contain 10B is minimal. It has been estimated that a gross 10B
content of approximately 35 µg/g of cancerous tissue is sufficient
to provide a therapeutic gain of about 4.2
The most formidable obstacle preventing the widespread use
of BNCT has been the development of new boron-containing
compounds that are truly selective for tumor cells. Despite the
recent apparently successful treatments of several patients in
the United States with glioblastoma multiforme and malignant
melanoma, this goal remains essentially unattained as the
boronated amino acid used in these early clinical trials,
p-boronophenylalanine (BPA), is not highly tumor-selective.
Typical tumor-to-blood and tumor-to-normal brain ratios in
animals and humans are about 3:1.3 Moreover, the single boron
atom present in BPA produces a low weight percentage of boron
in the molecule (∼5%) and very large doses must be admin-
istered in order to obtain therapeutically useful tumor boron
concentrations. This is a particular disadvantage for BPA since
it is poorly soluble at physiologic pH and is being administered
as the fructose complex, further decreasing the weight percent-
age of boron. The ideal boron compound for BNCT would be
water-soluble, nontoxic, and highly tumor-selective and would
carry multiple boron atoms on each molecule. Hawthorne has
recently published an excellent review on the role of boron
chemistry in the development of agents for BNCT.4
a Reagents: (a) (i) n-BuLi; (ii) CO2; (iii) HCl; (b) (PhO)2PON3/TEA/
t-BuOH/reflux.
diverse class of biologically specific agents with numerous
potential therapeutic applications, we decided to focus our
synthetic work on carborane amino acids.6
Recent reports of the synthesis of rod-like molecules contain-
ing two to five p-carborane cages7,8 led us to consider first the
preparation of 1-amino-12-carboxy-p-carborane (3). The syn-
thetic strategy is shown in Scheme 1. Initially our approach
was to synthesize 1-amino-p-carborane (2) from the correspond-
ing mono acid under Curtius conditions. The monoacid has
reportedly been prepared by stepwise lithiation and carboxyla-
tion of p-carborane (1),9 but in our hands this method consis-
tently produced mixtures of the mono- and diacids. Application
of the method of Michl et al. (lithiation under high-dilution
conditions followed by treatment with methyl chloroformate and
base cleavage of the ester)8 gave the desired monoacid in 83%
yield, but we were unable to convert this acid into the amine 2
using Curtius conditions. Direct amination of organolithium
compounds has been reported using methoxyamine,10 and this
method was successfully applied to the synthesis of 1-amino-
p-carborane 2, albeit in low yield and accompanied by signifi-
cant amounts of the diamine. Protection of the amine group as
the diphenylimine followed by lithiation, treatment with methyl
chloroformate, and stepwise deprotection gave the desired
compound 3. However, the overall yield of amino acid 3 was
unacceptably low, and we turned our attention to other potential
routes.
A far simpler route, as shown in Scheme 2, successfully
avoided the complications imposed by attempts to monolithiate
the carborane. p-Carborane (1) was dilithiated with n-butyl-
lithium (2.4 equiv) and converted to 1,12-bis(hydroxycarbonyl)-
p-carborane (4) in almost quantitative yield by reaction with
carbon dioxide followed by acidification. Shiori and co-workers
have reported a simple, one-step conversion of carboxylic acids
to urethanes using 1 equiv each of the acid, diphenylphosphoryl
azide (DPPA), and triethylamine in the presence of an alcoholic
solvent.11 When this modified Curtius reaction is carried out
Heterobifunctional polyhedral carboranes appeared to us to
be logical synthetic targets in our continuing investigations of
compounds for BNCT. Carboranes bearing dissimilar polar
functional groups on the two carbons of the icosahedral cage
might be water-soluble and possible tumor localizers in them-
selves, but could be even more important synthons for attach-
ment to known tumor seekers such as porphyrins.5 Unfortu-
nately, such compounds are exceedingly rare in the chemical
literature and general methods for their preparation are nonexist-
ent. Since peptides have long been recognized as an amazingly
(6) Varadarajan, A.; Hawthorne, M. F. Bioconjugate Chem. 1991, 2,
242-253.
(1) Barth, R. F.; Soloway, A. H.; Fairchild, R. G. Cancer Res. 1990, 50,
1061-1070.
(7) Yang, X.; Jiang, W.; Knobler, C. B.; Hawthorne, M. F. J. Am. Chem.
Soc. 1992, 114, 9719-9721.
(2) Fairchild, R. G.; Bond, V. P. Int. J. Radiat. Oncol. Biol. Phys. 1985,
11, 831-840.
(8) Mu¨ller, J.; Ba¨se, K.; Magnera, T. F.; Michl, J. J. Am. Chem. Soc.
1992, 114, 9721-9722.
(3) (a) Mallesch, J. L.; Moore, D. E.; Allen, B. J.; McCarthy, W. H.;
Jones, R.; Stenning, W. A. Int. J. Radiat. Oncol. Biol. Phys. 1994, 28, 1183-
1188. (b) Fukuda, H.; Hiratsuka, J.; Honda, C. Radiat. Res. 1994, 138,
435-442.
(9) Zakharkin, L. I.; Kalinin, V. N.; Podvisotskaya, L. S. IzV. Akad. Nauk.
SSSR, Ser. Khim. 1968, 2661.
(10) (a) Beak, P.; Basa, A.; Kokko, B.; Loo, D. J. Am. Chem. Soc. 1986,
108, 6016-6023. (b) Beak, P.; Kokko, B. J. Org. Chem. 1982, 47, 2822-
2823.
(4) Hawthorne, M. F. Angew. Chem., Int. Ed. Engl. 1993, 32, 950-984.
(5) Hill, J. S.; Kahl, S. B.; Kaye, A. H.; Stylli, S. S.; Koo, M.-S.;
Gonzales, M. B.; Vardaxis, N. J.; Johnson, C. I. Proc. Natl. Acad. Sci.
U.S.A. 1992, 89, 1785-1789.
(11) (a) Ninomiya, K.; Shiori, T.; Yamada, S. Tetrahedron 1974, 30,
2151-2157. (b) Shiori, T.; Ninomiya, K.; Yamada, S. J. Am. Chem. Soc.
1972, 94, 6203-6205.
0002-7863/96/1518-1223$12.00/0 © 1996 American Chemical Society

1224 J. Am. Chem. Soc., Vol. 118, No. 5, 1996
Communications to the Editor
Scheme 3a
Scheme 5a
a Reagents: (a) (i) n-BuLi; (ii) CO2; (b) (PhO)2PON3/TEA/t-BuOH/
PTC.
a Reagents: (a) Cs2CO3/PhCH2Br/CH3CN/reflux; (b) (i) TFA/
CH2Cl2; (ii) TEA/CH2Cl2.
Scheme 6a
Scheme 4a
a Reagents: (a) (i) (CH3)3SiN3; (ii) t-BuOH; (b) (i) n-BuLi; (ii) CO2;
(iii) HCl.
a Reagents: (a) (i) n-BuLi; (ii) CO2; (iii) HCl; (b) (PhO)2PON3/DIEA/
t-BuOH/reflux.
Treatment of o-carborane (11) with excess n-butyllithium gave
the dilithio salt 12 which, instead of being converted into the
diacid as with m-carborane, was treated with DPPA and
diisopropylethylamine in the presence of a phase transfer catalyst
in refluxing tert-butyl alcohol to afford the desired Boc-protected
amino acid 13 in 54% yield.
However, the yields in this reaction varied widely, and
another, more reproducible approach was sought. The reaction
sequence shown in Scheme 6 proved useful. o-Carboranecar-
bonyl chloride 14, easily prepared from the monoacid, was
refluxed with trimethylsilyl azide followed by subsequent
addition of tert-butyl alcohol to give quantitatively the Boc-
protected amine 15. Treatment of 15 with n-butyllithium
followed by carbon dioxide reproducibly gave the desired amino
acid 13 in 70-80% yield.
In summary, we have demonstrated a versatile, general
method for the conversion of o-, m-, and p-carborane to their
corresponding Boc-protected amino acids. Heterobifunctional
polyhedral carboranes are exceedingly rare in the literature, and
the amino acids prepared by this general method may prove to
be valuable synthons for use in the synthesis of tumor-seeking
compounds for BNCT or PDT. Moreover, these conforma-
tionally constrained amino acids should be particularly interest-
ing for use in peptide synthesis. The dihedral angle between
the carbon atoms of these polyhedra increases in the order 60°
(ortho), 110° (meta), and 180° (para), allowing the peptide
chemist to select a desired conformation. Efforts are currently
underway in our laboratory to demonstrate both uses and will
be reported at a later date.
in tert-butyl alcohol, the intermediate isocyanate collapses to a
Boc-protected amine group, giving a protected amino acid
ideally suited for peptide synthesis. Thus the diacid 4 was
reacted with DPPA/TEA and a catalytic amount of 4-(dimethyl-
amino)pyridine (DMAP) in refluxing tert-butyl alcohol under
high-dilution conditions to afford the N-protected amino acid
5 in an overall yield of 87% from p-carborane. If desired, the
free amino acid 3 can be obtained easily by treatment of this
compound with acid.
It is also possible to invert the functional group protection to
expose the amine group as shown in Scheme 3. Treatment of
N-protected amino acid 5 with benzyl bromide and cesium
carbonate in refluxing acetonitrile protects the carboxyl group
as its benzyl ester 6. Acid-induced cleavage of the Boc
carbamate followed by neutralization provides the C-protected
amino acid 7. In this manner, both the N- and C-protected
amino acids become readily available for linkage to tumor-
seeking functions or for peptide synthesis.
In order to test the generality of this method, it was applied
to m- and o-carborane, 8 and 11, respectively. The diacids of
these compounds are readily prepared by treatment with excess
butyllithium followed either by carbon dioxide or by methyl
chloroformate and base. Yields for these reactions are generally
in excess of 95%. Reaction of m-carborane dicarboxylic acid
(9) with DPPA and diisopropylethylamine (DIEA) in the
presence of a catalytic amount of DMAP in tert-butyl alcohol
gave the desired amino acid as a mixture of the tert-butyl and
phenyl carbamates. These products can be separated by silica
gel chromatography, but pure Boc-protected amino acid 10 can
be obtained by further modification of the reaction sequence,
as shown in Scheme 4. The diacid 9 was reacted with 1 equiv
of ethyl chloroformate in the presence of diisopropylethylamine
in acetone at high dilution at room temperature. Further
treatment in situ with aqueous sodium azide afforded the mono
acyl azide. After removal of the acetone in Vacuo at room
temperature followed by acidification, the product was extracted
with ethyl acetate-toluene (30:70). The extraction solution was
refluxed in the presence of tert-butyl alcohol overnight to give
the Boc-protected amino acid 10 in more than 80% yield from
m-carborane (8).
Acknowledgment. This research was supported by the National
Institutes of Health (CA-37961) and the Department of Energy Office
of Energy Research (DE-FG03ER60873-91). Mass spectra were
provided by the UCSF Mass Spectrometry Facility (A. L. Burlingame,
Director) supported by the Biomedical Research Technology Program
of the National Center for Research Resources, NIH NCRR BRTP
RR01614.
Supporting Information Available: Experimental details for the
synthesis and characterization of 2, 4, 5, 7, 9, 10, 12, 13, and 15 (6
pages). This material is contained in many libraries on microfiche,
immediately follows this article in the microfilm version of the journal,
can be ordered from the ACS, and can be downloaded from the Internet;
see any current masthead page for ordering information and Internet
access instructions.
Application of the DPPA/TEA method to o-carboranedicar-
boxylic acid produced only the phenyl carbamate in quantitative
yield. However, modification of the reaction sequence as shown
in Scheme 5 gave the desired Boc-protected amino acid 13.
JA9534260