and purity. In light of these limitations, we surmised that if
one could increase the electrophilicity of the activated
carboxylic acid, one could subsequently increase the ef-
ficiency of these problematic Im-NH2/BocPyOH couplings.
In 2002, Jung et al. reported an Fmoc-based synthesis of
cyclic peptides containing sterically hindered secondary
amines on a solid support in which BTC [bis(trichlorom-
ethyl)carbonate or triphosgene] was used as the coupling
agent.25,26 The BTC reagent putatively forms acid chlorides
in situ in high yield and has shown application as a highly
efficient coupling agent for the Fmoc-mediated synthesis of
aromatic oligoamides using automated peptide synthesizers.27
It has also been shown that BTC performs better than TFFH28
and exclusively our BTC protocol, we were pleased to prepare
polyamide 1 in 33% yield after CBz deprotection of the γ-turn
motif; i.e. this is a 330-fold increase in isolated yield for 1 using
the BTC coupling protocol.
Encouraged by this result, we then investigated the
preparation of polyamides using solid supports that do not
install an A•T/T•A encoding ꢀ-Ala tail on the C-termini of
polyamides. This is an important requisite for biological
applications as the presence of ꢀ-Ala tails is known to
correlate with generally poor cellular uptake as well as
enforcing an A•T/T•A encoding end sequence.4,10,17 The
Dervan group recognized this limitation and developed a
polyamide synthetic protocol which utilized the Kaiser resin
to delete the ꢀ-Ala tail.32 As a consequence of their increased
stability in strongly acidic and basic conditions coupled with
a mild oxidative resin release protocol that enables the
installation of various tail functionalities, aryl hydrazide
resins offer potential advantages over Kaiser oxime resins
currently used for truncated polyamide synthesis.31 We
evaluated the compatibility of the BTC coupling strategy with
aryl hydrazide resins via the preparation of polyamide 2.
Polyamide 2 was chosen as it comprises a diverse range of
couplings not encountered in the synthesis of 1 such as an
Im-Im coupling, in addition to the known challenging Resin-
ImNH2-BocPyOH coupling late in the synthesis. The solid
phase synthesis of polyamide 2 proceeded smoothly using the
BTC protocol. Polyamide release from the hydrazine resin was
then achieved by oxidative activation of the resin in the presence
of 2.0 equiv of NBS in pyridine for 10 min at room temperature.
Cleavage of the polyamide from the activated diazene solid
support with dimethylamino propylamine (Dp) at 40 °C for 5 h
afforded the crude polyamide product 2 in 73% purity. After
semipreparative HPLC purification, the isolated yield of 2 was
29% yield and with a purity of 98% (Figure 2). The versatility
29
and POCl3 in the solid phase synthesis of difficult se-
quences.
The in situ formation of the putative acid chloride also enables
the utilization of acid-labile protecting groups such as tert-butyl
esters, which are reported to have a limited shelf life in the
presence of acid chlorides.30 Although there have been no
reports of its use in Boc-based solid phase synthesis, we
reasoned that the increased electrophilicity of the acid chloride
generated in situ by the reaction of BTC with the appropriate
aromatic carboxylic acid would considerably enhance coupling
efficiencies of polyamides and reduce reaction times. Indeed
the BTC method was found to be far superior to current
benzotriazole-based protocols in all couplings tested (Table 1).
Activation times of both the BocPyOH and BocImOH only
required 1 min when 0.33 equiv of BTC was used, compared
to 2 h for DCC/HOAt-mediated activations.20 Coupling times
of 20 min were typically required for quantitative conversion
to coupled products using BTC according to HPLC analysis,
which enabled each deprotection-coupling-wash cycle to be
effected well within 1 h. To the best of our knowledge, this is
the first demonstration of a BTC-mediated solid phase synthesis
protocol using Boc chemistry.
Encouraged by our model studies, we then investigated
whether a BTC-based coupling methodology could be applied
to the preparation of challenging hairpin polyamide sequences.
We chose to investigate the synthesis of polyamide 1 which
targets the DNA sequence 5′-WWGWGCW-3′ (where W is
either A or T) with nanomolar affinity.31 Compound 1 was
prepared in only 0.1% reported yield using the ꢀ-Ala PAM resin
via a traditional Boc-chemistry/benzotriazole-based ester pro-
tocol.30 The low yield is most likely attributed to a challenging
Resin-ImNH2/BocPyOH coupling late in the synthesis sequence,
which was confirmed in our laboratory using the conventional
coupling methodology. This polyamide sequence also comprises
other typical coupling sequences such as Py-Py, Py-Im, and
Im-aliphatic couplings which enable us to ascertain the scope
of the BTC coupling methodology. Using Boc-based chemistry
Figure 2. HPL chromatogram of (A) crude and (B) purified
polyamide 2. The retention time of 2 is 13.02 min.33
(25) Thern, B.; Rudolph, J.; Jung, G. Angew. Chem., Int. Ed. 2002, 41,
of the aryl hydrazine method was also demonstrated by the
NBS-mediated oxidative release of polyamides outfitted with
amino- (3) and alkyne-modified (4) tail structures in 22% and
12% yields, respectively.
According to the high purities of crude 3 and 4 by HPLC
analysis, the lower isolated yield of these polyamides
compared to polyamide 1 most likely correlates to their
2307
.
(26) Thern, B.; Rudolph, J.; Jung, G. Tetrahedron Lett. 2002, 43, 5013
.
(27) Konig, H. M.; Gorelik, T.; Kolb, U.; Kilbringer, A. F. M. J. Am.
Chem. Soc. 2007, 129, 704.
(28) Carpino, L. A.; Elfaham, A. J. Am. Chem. Soc. 1995, 117, 5401.
(29) Rijkers, D. T. S.; Adams, H. P. H. M.; Hemker, H. C.; Tesser,
G. I. Tetrahedron 1995, 51, 11235.
(30) Falb, E.; Yechezkel, T.; Salitra, Y.; Gilon, C. J. Pept. Res. 1999,
53, 507.
3912
Org. Lett., Vol. 11, No. 17, 2009