through the THP group provides a traceless linker for the
alcohols 8a-af and the final compounds 2a-aj. Preparation
of resin 5 was achieved by combining a 5-fold excess of
8-oxo-geraniaol 4 with support 3 in the presence of 0.2 equiv
of PPTS at 60 °C. Under these conditions, excess alcohol 4
can be recovered for reuse. The optimized reaction conditions
gave a resin loading of 72% for the solid support-linked R,â-
unsaturated aldehyde 5. Aldehyde loading was determined
by cleavage and recovery of alcohol 4 by treatment of resin
5 with PPTS/MeOH/DCE at 60 °C. Attempts to generate
resin-bound aldehyde 4 connected via a silyl linker to the
solid support were unsuccessful.
ing chlorides followed by diphosphorylation to provide
diphosphates.13,24,25 Utilizing this approach reduces some of
the advantages of SPOS, as it requires two solution-phase
manipulations after release of the anilinogeraniols 8a-af
from the resin to produce the desired diphosphates. In
previous solution-phase work, we employed an excess of
Ph3PCl2 to convert alcohol 8a to its corresponding chloride,
which was immediately transformed to the diphosphate 2a.13
Direct cleavage of the resin-linked THP ethers to the
corresponding allylic chlorides or bromides would reduce
the number of solution steps to one, diphosphorylation. We
observed that addition of Ph3PCl2 to the resin-bound ethers
and subsequent diphosphorylation resulted in poor yield of
diphosphates. Alternatively, stirring the THP resin 7a-aj
with 3 equiv of Ph3PBr2 in CH2Cl2 for 4 h followed by
addition of 10 equiv of tris(tetra-n-butylammonium)hydro-
gendiphosphate in CH3CN gave the desired diphosphates
2a-aj in moderate yield. Employing Ph3PBr2 not only
reduced the reaction time but also increased the yield of the
desired diphosphates. We observed that the yield of the
diphosphates falls precipitously if the bromination is allowed
to continue for longer times. We attribute these observations
to the formation of undesired side products due to the
enhanced reactivity of the Ph3PBr2. The release of bromides
9a-aj from the THP resin by Ph3PBr2 provides a traceless
linker path to the FPP analogues 2a-aj. In practice, bromides
9a-aj were not isolated but were converted directly to the
anilinogeranyl diphosphates 2a-aj by addition of tris(tetra-
n-butylammonium) hydrogen diphosphate25 in dry CH3CN
in the same reaction vessel.
With resin 5 in hand, we employed a solid-phase parallel
synthetic strategy to prepare a directed library of the
anilinogeranyl farnesol analogues 8a-af. Diversity was
introduced into the library by selecting the commercially
available substituted anilines 6a-aj listed in Table 1.
Optimum yields of the desired resin-bound anilinogeraniols
7a-aj were obtained by performing the reductive amination
with 10 equiv of the anilines and 12 equiv of acetic acid in
1:1 THF/DCE and subsequent reduction with 10 equiv of
NaBH(OAc)3. Increasing the number of equivalents of acetic
acid or substituting PPTS or other Lewis acids both reduced
the yield of the desired anilinogeraniols and led to additional
uncharacterized byproducts. Moreover, reducing the number
of equivalents of anilines for imine formation or decreasing
the fold excess of NaBH(OAc)3 employed for the subsequent
reduction resulted in longer reaction times that gave reduced
yields. Cleavage of the anilinogeraniols 8a-af from the
resin-bound amines 7a-af was achieved by treatment of the
solid support with DCE/MeOH/PPTS at 60 °C. Methanol,
ethanol, and n-butanol were all found to be equally effective
in promoting the cleavage reaction. The final yields of the
anilinogeraniols 8a-af after silica gel column chromatog-
raphy are reported in Table 1. Anilinogeraniol 8a and
p-nitroanilinogeraniol 8s have previously been prepared by
solution methods in 85 and 64% yield, respectively. This is
in contrast to results from the SPOS where the presence of
an electron-withdrawing group in the para position leads to
higher yield from the reductive amination (Table 1, entries
7, 18 and 19). The optimum yield of anilinogeraniols
prepared by SPOS reductive amination requires both elevated
temperatures and increased reaction times relative to solution
methods.
The resulting diphosphates 2a-aj were first converted to
+
the NH4 form by ion exchange chromatography and then
purified by reverse-phase HPLC. Complete removal of the
tetra-n-butylammonium counterions from the sample prior
to reverse-phase HPLC was required for effective purifica-
+
tion. However, efficient recovery of the NH4 form of the
diphosphates 2a-aj from the ion exchange chromatography
was highly dependent on the buffer conditions employed.
This observation is unsurprising, as the various diphosphates
2a-aj are expected to have a wide range of solubilites in
aqueous buffer related to the structure of the parent anilines
6a-aj. Optimization of the ion exchange conditions by
altering the concentration of NH4HCO3 and including
compound-dependent proportions of MeCN cosolvent had
a profound influence on the yield of individual diphosphates
2a-aj recovered. However, the yields of diphosphates
2a-aj reported in Table 1 are for a single, uniform, ion
exchange condition utilized to maximize throughput. The
mixed SPOS-solution-phase route outlined here provides
sufficient material for biochemical screening and is shorter
and more convenient than what we have previously reported.
1
Integration of the H NMR spectra of amines 8a-af
indicated that a mixture of cis/trans isomers about the 6,7
double bond were formed in an approximately 1:9 ratio. This
result was expected since similar cis/trans mixtures were
observed for unsubstituted and para-substituted anilinogera-
nyl acetates prepared by solution methods. Interestingly,
bulky ortho substituents on anilines 6h, 6j, 6p, 6t, and 6ab
did not lead to any alteration of the cis/trans ratio for the
corresponding anilinogeraniols 8h, 8j, 8p, 8t, and 8ab. As
is typical of the previously prepared anilinogeraniols, we
were unsuccessful in separating the cis/trans isomers of
8a-af by HPLC methods.
(21) Lepre, C. A.; Peng, J.; Fejzo, J.; Abdul-Manan, N.; Pocas, J.; Jacobs,
M.; Xie, X.; Moore, J. M. Comb. Chem. High Throughput Screening 2002,
5, 583-590.
(22) Thompson, L. A.; Ellman, J. A. Chem. ReV. 1996, 96, 555-600.
(23) Ma, S.; Duan, D.; Wang, Y. J. Comb. Chem. 2002, 4, 239-247.
(24) Mu, Y.; Gibbs, R. A.; Eubanks, L. M.; Poulter, C. D. J. Org. Chem.
1996, 61, 8010-8015.
In solution, FPP analogues are typically prepared by
sequential conversion of the allylic alcohols into correspond-
(25) Davisson, V. J.; Woodside, A. B.; Neal, T. R.; Stremler, K. E.;
Muehlbacher, M.; Poulter, C. D. J. Org. Chem. 1986, 51, 4768-4779.
Org. Lett., Vol. 7, No. 11, 2005
2111
Subramanian, Thangaiah
