An improved aldehyde linker for the solid phase synthesis of hindered amides

Article abstract of DOI:10.1021/jo051742j

A novel aldehyde dual-linker system has been developed for the solid phase synthesis of sterically hindered amides. The linker [5-(4-formyl-3- hydroxyphenoxy)pentanoic acid] exploits an intramolecular oxygen-nitrogen acyl transfer mechanism to prepare com

Full text of DOI:10.1021/jo051742j

An Improved Aldehyde Linker for the Solid Phase Synthesis of
Hindered Amides
Mark J. Liley,*,† Tony Johnson,† and Susan E. Gibson§
MediVir UK Ltd., Chesterford Research Park, Little Chesterford, Essex C10 1XL, United Kingdom,
and Department of Chemistry, Imperial College London, South Kensington Campus,
London SW7 2AY, United Kingdom
mark.liley@mediVir.com
ReceiVed August 18, 2005
A novel aldehyde dual-linker system has been developed for the solid phase synthesis of sterically hindered
amides. The linker [5-(4-formyl-3-hydroxyphenoxy)pentanoic acid] exploits an intramolecular oxygen-
nitrogen acyl transfer mechanism to prepare compounds that are unattainable with current commercially
available linkers. A dual linker system, exploiting the hyper-acid labile Sieber amide linker as part of the
construct, enabled the initial reductive alkylation reactions of hindered amines and their subsequent
acylation with a range of carboxylic acids with varying stereoelectronic properties to be monitored. Simple
acylation conditions (HBTU/HOBt/NMM) sufficed to provide near quantitative reaction of test acids
with support-bound hindered amines, reaction conditions which failed when commercial linkers were
used.
Introduction
In our current research, amino acids are used as starting points
for the design and synthesis of potential enzyme inhibitors. We
required a simple solid-phase method that would enable an
amino acid to be functionalized at both its N- and C-termini
with minimal epimerization. We considered the most facile way
to achieve this was through monovalent attachment of the
conserved nitrogen to a linker, thus allowing simple synthetic
manipulation of both termini prior to resin cleavage. Using this
approach, libraries of compounds could be prepared that would
not contain an unwanted functional group derived from the
amino acid attachment point to the linker.
Such linkers have been developed to enable the solid phase
synthesis of C-terminally modified peptides3 and although the
polymer support may vary, there are essentially only four types
of this linker commercially available. (Figure 1).
These have been utilized in the literature to facilitate the syn-
thesis of a variety of different compounds4 with some specific
examples including benzimidazoles5 and alkoxyketones.6
Although all of these linkers have been shown to be effective,
they all have limited use for the formation of amide bonds. This
In recent years there has been an explosion of interest in the
research and development of novel linkers for polymer supported
synthesis1 and several reviews have been published covering
this area.2 Originally, linkers were designed almost exclusively
to aid peptide and nucleotide synthesis. However, the advent
of combinatorial chemistry techniques has led to a demand for
a wider range of linkers for solid-phase synthesis. This is
because the original linkers have one major drawback: most
leave a functional group attached to the cleaved molecule that
was the attachment point for the linker (e.g., a carboxylic acid,
an amide, or an alcohol). This is not always desirable for
combinatorial library synthesis.
* Address correspondence to this author.
† Medivir UK Ltd..
§ Imperial College London.
(1) (a) Yan, L. Z.; Mayer, J. P. J. Org. Chem. 2003, 68, 1161. (b)
Golisdlae, A.; Herforth, C.; Quirijen. L.; Maes, L.; Link, A. Bioorg. Med.
Chem. 2002, 10, 159. (c) Chhabra, S. R.; Parekh, H.; Khan, A. N.; Bycroft,
B. W.; Kellam, B. Tetrahedron Lett. 2001, 42, 2189. (d) Andrews, S. P.;
Ladlow, M. J. Org. Chem. 2003, 68, 5525. (e) Portal, C.; Launay, D.;
Merritt, A. J. Comb. Chem. 2005, 7, 554.
(2) (a) Blaney, P.; Grigg, R.; Sridharan, V. Chem. ReV. 2002, 102, 2607.
(b) Guillier, F.; Orain, D.; Bradley, M. Chem. ReV. 2000, 100, 2091.
(3) Jenson, K. J.; Alsina, J.; Songster, M. F.; Vagner, J.; Albericio, F.;
Barany, G. J. Am. Chem. Soc. 1998, 120, 5441.
10.1021/jo051742j CCC: $33.50 © 2006 American Chemical Society
Published on Web 01/27/2006
1322
J. Org. Chem. 2006, 71, 1322-1329

Solid Phase Synthesis of Hindered Amides
FIGURE 2. 4-(4-Formyl-3-methoxyphenoxy)butanoic acid (MeOBAL)
(1).
FIGURE 1. Commercially available backbone amide linkers (BAL).
SCHEME 1
FIGURE 3. Coupling of MeOBAL to Sieber amide resin.
was then cleaved from the hyper acid labile linker with a low
concentration of acid. Under these conditions (typically 1%
TFA, 5% TES in DCM) the aldehyde linker is completely stable
(data not published).
SCHEME 2. “Double Linker” Strategy
The construct of choice to define the limitations of the
commercially available linker is shown below (Figure 4). Valine
was chosen as a typical sterically hindered amine as subsequent
acylation could be problematic with an electron deficient or
structurally hindered acid. A number of polymer supported
reductive alkylation protocols have been described in the
literature.7-9 Our method of choice utilized sodium cyanoboro-
hydride in 1% AcOH/DMF for 18 h.
To ensure that quantitative reductive alkylation had occurred,
a small portion of resin was cleaved and the resulting filtrate
analyzed by HPLC and mass spectrometry techniques. Only one
major peak was seen by HPLC, and electrospray mass spec-
trometry (ESI) confirmed this to be the predicted product (5).
The HPLC chromatogram (Figure 5) showed no evidence of
the starting aldehyde linker.
Following successful reductive alkylation, acylation was
attempted with use of typical peptide synthesis techniques. To
test the scope of this resin-linker construct, several acids were
chosen for coupling with varying steric and electronic properties
(Figure 6).
Each acid was mixed with 4 for 64 h, using HBTU as an in
situ coupling agent in the presence of NMM and HOBt in DMF.
The degree of acylation was determined by analysis of the
cleaved products, using analytical HPLC/ESI.
problem is illustrated in Scheme 1: reductive alkylation
proceeds smoothly (e.g., with phenylalanine) but subsequent
acylation is achieved only when using highly activated coupling
reagents.3 Even so, acylation (e.g., with the relatively unhindered
leucine) is still not quantitative.3
Results and Discussion
Our research initially focused on defining the limitations of
one of the commercially available aldehyde linkers with respect
to the formation of amide bonds. To aid this investigation, a
solid phase “double linker” strategy was developed to allow
the rapid analysis of solid supported postreaction products. The
principle of this technique is shown in (Scheme 2).
This strategy enables the progress of reactions (e.g., reductive
alkylation and acylation) to be monitored via simple cleavage
and HPLC analysis of the resulting product(s). Since the
aldehyde linker (1) is UV active, both starting material and
products can be readily detected. The commercially available
aldehyde linker chosen for the initial studies was 4-(4-formyl-
3-methoxyphenoxy)butanoic acid (MeOBAL) (1) (Figure 2).
The aldehyde handle was first attached to polystyrene resin
functionalized with the hyper-acid labile Sieber amide linker
(Figure 3). Reductive alkylation and acylation were then
performed as described below. The newly synthesized molecule
Evaluation of the data showed that for all the acids chosen
(with the exception of acetic acid) <5% acylation had occurred
(Table 1, entries 1-8). The major product in each case (>95%)
was the linker-Val-OBut starting material (Val-OMe also gave
<5% acylation, Table 1, entries 9 and 10). This provided
conclusive evidence that acylation of sterically hindered amines
bound to MeOBAL proceeded at very low rates.
More forcing reaction conditions were employed in an attempt
to improve acylation onto the hindered valine. The symmetrical
anhydride of Z-leucine was prepared and compared to the
coupling of acetic anhydride. As expected the use of acetic
(4) (a) del Fresno, M.; Alsina, J.; Royo, M.; Barany, G.; Albericio, F.
Tetrahedron Lett. 1998, 38, 2639. (b) Farrant, E.; Rahman, S. S. Tetrahedron
Lett. 2000, 41, 5383. (c) Tolborg, J. F.; Jenson, K. J. Chem. Commun. 2000,
147. (d) Giovannoni. J.; Subra, G.; Amblard, M.; Martinez, J. Tetrahedron
Lett. 2001, 42, 5389. (e) Herpin, T. F.; Van Kirk. K. G.; Salvino, M. J.;
Yu, S. T.; Labaudiniere, R. F. J. Comb. Chem. 2000, 2, 513. (f) Estep, K.
G.; Neipp, C. E.; Stephens Stramiello, L. M.; Adam, M. D.; Allen, M. P.;
Robinson, S.; Roskamp, E. J. J. Org. Chem. 1998, 63, 5300.
(5) (a) Smith, J. M.; Krchnak, V. Tetrahedron Lett. 1999, 40, 7633. (b)
Muzurov, A. Bioorg. Med. Chem. Lett. 2001, 10, 67.
(6) Fenwick, A. E.; Garnier, B.; Gribble, A. D.; Ife, R. J.; Rawlings, A.
D.; Witherington, J. Bioorg. Med. Chem. Lett. 2001, 11, 195.
(7) Ho, P. T.; Chang, D.; Zhong, J. W. X.; Musso, G. F. Peptide Res.
1993, 6, 10.
(8) Coy, D. H.; Hocart, S. J.; Sasaki, Y. Tetrahedron 1988, 44, 835.
(9) (a) Gordon, D. W.; Steele, J. Biorg. Med. Chem. Lett. 1995, 5, 47.
(b) Szardenings, A. K.; Burkoth, S.; Look, G. C.; Campbell, A. J. Org.
Chem. 1996, 61, 6720. (c) Abdel-Magid, A. F.; Carson, K. G.; Harris, B.
D.; Maryanoff, C. A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849.
J. Org. Chem, Vol. 71, No. 4, 2006 1323

Liley and Johnson
FIGURE 4. Valine-OBut reductively alkylated onto “double linker” to give construct (4).
TABLE 1. Acylation Reactions with MeOBAL (reaction time 64 h)
amine reductively
alkylated
coupling
reagent
coupling
additive
coupling
base
coupling
solvent
acid
coupled
% starting amine
remaining
entry
1
2
3
4
5
6
7
8
9
valine-OBut
valine-OBut
valine-OBut
valine-OBut
valine-OBut
valine-OBut
valine-OBut
valine-OBut
valine-OMe
valine-OMe
valine-OBut
valine-OBut
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
NMM
NMM
NMM
NMM
NMM
NMM
NMM
NMM
NMM
NMM
DIPEA
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DCM
DCM
Z-Val-OH
Z-Leu-OH
o-anisic
>95
>95
>95
>95
>95
>95
<25
>95
>95
>95
<5
p-anisic
o-nitrobenzoic
p-nitrobenzoic
acetic
diphenylacetic
Z-Val-OH
Z-Leu-OH
acetic anhydride
Z-Leu-OH
10
11
12
acetic anhydride
symmetrical anhydride of Z-leucine
>85
SCHEME 3. Oxygen-Nitrogen Acyl Transfer
aldehyde linker,10 4-(4-formyl-3-hydroxyphenoxy)pentanoic acid
(OHBAL, Hydroxy, Backbone Amide Linker) (6) that would
potentially overcome these problems (Scheme 4). We have
described the synthesis of 6 in a patent,10 but no experimental
details of its use were included. Limited application of a similar
linker has been reported.11 In this work, simple unhindered
amines were attached via reductive alkylation and coupled with
simple amino acids. No demanding coupling reactions with
respect to steric or electronic considerations were reported.
It was envisaged that linker 6 could be simply attached to
any free amine resin via typical SPPS conditions (HBTU, HOBt,
NMM, DMF, e.g., Figure 3).
FIGURE 5. HPLC chromatogram of sample 5 from resin cleavage.
When OHBAL is used, acylation with both electron poor and
sterically hindered acids should be possible via initial acylation
of the 3-hydroxy function followed by O to N acyl transfer of
the incoming acid.12 This concept is illustrated in Scheme 3.
The synthetic route to OHBAL itself is outlined in Scheme
4. Selective alklyaltion was achieved with conditions described
by Mendelson et al.13 Methyl 5-bromopentanoate, spray dried
potassium fluoride, and 2,4-dihydroxybenzaldehyde were heated
under reflux in acetonitrile. The crude methyl ester (7) was then
purified by flash chromatography to give a white solid in good
FIGURE 6. Selected acids for acylation onto H-Val-(OBut)-resin (4).
anhydride led to almost quantitative acylation (Table 1, entry
11). However, the use of the highly electrophilic symmetrical
anhydride of the more hindered Z-leucine yielded <15% of the
desired acylated product by HPLC area (Table 1, entry 12).
HATU activation was also attempted as reported in the
literature3 but this gave little improvement over HBTU activation
(results not shown). Almost identical results were observed when
using BAL and Di-MeOBAL (Figure 1). Indole BAL (Figure
1) proved worse still, with only partial reductive alkylation being
noted even after resubmission of the resin to the reductive
conditions.
(10) Johnson, T.; Quibell, M. W. Patent Appl. WO 98/17628 (File date:
1997).
(11) Okayama, T.; Burritt, A.; Hruby, V. J. Org. Lett. 2000, 2, 1787.
(12) Johnson, T.; Quibell, M.; Owen, D.; Sheppard, R. C. J. Chem. Soc.,
Chem. Commun. 1993, 369.
(13) Mendelson, W. L.; Holmes, M.; Dougherty, J. Synth. Commun. 1996,
26, 593.
Having established the limitations of the commercially avail-
able linkers, work commenced on the investigation of a new
1324 J. Org. Chem., Vol. 71, No. 4, 2006

Solid Phase Synthesis of Hindered Amides
SCHEME 4. Synthesis of 5-(4-Formyl-3-hydroxyphenoxy)-
pentanoic Acid (OHBAL) (6) and Subsequent Attachment to
the Poylmer Support (8)
FIGURE 7. Two potential products derived from reductive alkylation
with OHBAL.
With use of the same eight carboxylic acid probes as
previously (Figure 6) and utilizing identical standard coupling
chemistry, the synthetic utility of the resin was tested. The
percentage acylation was determined by analysis of the cleaved
products by HPLC/ESI.
The data clearly showed >80% acylation had occurred for
all eight carboxylic acid probes (Table 2, entries 1-8) compared
to <5% for the commercial linkers under identical conditions
(Table 1). This provided conclusive evidence that the De-
ALOBAL had overcome the synthetic limitations that were seen
when using the commercial linkers for amide bond formation.
An example of one of the HPLC chromatograms is shown in
Figure 8.
yield. Hydrolysis of the ester via treatment with lithium
hydroxide in THF gave the desired acid (6) in good yield. Once
this had been isolated, the construct 8 was prepared on the solid
phase in a manner analogous to that used earlier to assemble
construct 4 from MeOBAL (1) (Figure 4).
Following acidic cleavage of a small portion of resin, HPLC
analysis revealed two peaks at 3.0 and 3.2 min in a relative
ratio of 3:2. Electrospray MS confirmed the major component
to be that of the desired product (9) and the minor product to
be the imine (10) (Figure 7). This phenomenon was not noted
when using the commercial linker and interestingly, Okayama
et al.11 did not report problems during reductive alkylation even
though an identical 2-hydroxyl functionality was present on their
linker. A variety of reported reductive alkylation conditions were
employed, including those used by Okayama et al.,11 and in
each case incomplete alkylation was seen with the hindered
amine. Even resubmission of the resin to the reductive conditions
failed to isolate the desired amine quantitatively.
This discovery led to a modification to the design and
synthesis of the new linker. It was proposed that temporary,
orthogonal protection of the hydroxyl function would facilitate
quantitative reductive alkylation. The revised structure and
synthetic route are outlined in Scheme 5. Methyl 5-(4-formyl-
3-hydroxyphenoxy)pentanoate (7), allyl bromide, and cesium
carbonate were stirred at room temperature in dry acetonitrile
for 3 h. The resulting crude product was purified by flash
chromatography to yield a white solid ester (11) in high yield.
Hydrolysis of the ester via treatment with lithium hydroxide in
THF gave the desired acid (12) in excellent yield (ALOBAL,
AllylOxy Backbone Amide Linker).
Pleasingly, reductive alkylation was found to occur quanti-
tatively. A small portion of resin was cleaved and the filtrate
analyzed by HPLC/ESI. One major peak at 3.1 min was
identified as the desired product (14) (Supporting Information)
and there was no evidence of the starting aldehyde linker.
The allyl protecting group was subsequently removed with a
mixture of tetrakistriphenylphosphine palladium(0) and phe-
nylsilane in DCM.14 To ensure quantitative deprotection had
occurred, a small portion of resin was cleaved and analyzed by
HPLC/ESI. One major peak at 2.6 min was identified as the
desired product (15) (Supporting Information) and there was
no evidence of the starting material.
The results clearly show the linker to be an extremely useful
tool for the formation of amide bonds. Both electron poor and
electron rich acids were coupled quantitatively to the sterically
hindered amines. Of particular note is the successful acylation
of a â-branched amino acid (Z-Val-OH) to the already sterically
hindered Val-OBut and Val-OMe (Table 2, entries 1 and 10).
One minor limitation of the linker was seen when coupling an
ortho-substituted benzoic acid. Both o-anisic acid and 2-ni-
trobenzoic acid (Table 2, entries 3 and 5) showed slightly
reduced acylation. A comparison of the degree of acylation with
ortho/para anisic acids and ortho/para nitro benzoic acids
(electron donating verses electron withdrawing) showed prefer-
1
ence for the latter. Full H and 13C NMR characterization of
one of the compounds prepared (16) (Table 2, entry 7) is detailed
in the Experimental Section.
Having used the dual linker system to demonstrate the
enhanced reactivity of the linker, the synthetic utility of this
approach was investigated by using ALOBAL directly attached
to the polymer (17) (Figure 9).
By using the same carboxylic acid probes as before (Figure
6) and utilizing identical standard coupling chemistry, seven
amides were prepared. Each was isolated by treatment of the
resin with 95% TFA, 5% TES. After workup the crude samples
were lyophilized and accurate weights determined. The yields
of the compounds (Table 3) were calculated by weight based
on the initial loading of the commercially available resin. Purity
was determined by HPLC peak area.
The purities of the amides were consistent with the data
observed with the dual linker system, but the overall yields were
a little lower than expected.
To test further the synthetic utility of the new linker, O to N
acyl transfer was attempted with 4-methoxy aniline since as a
class, anilines are poor nucleophiles compared to alkylamines.
The reductive alkylation chemistry described earlier for valine
worked quantitatively with the ALOBAL and MeOBAL and
worked well when using the Di-OMeBAL but did gave rise to
minor impurities. Acylation with two acids (Z-valine-OH and
p-nitrobenzoic acid), using HBTU chemistry, for both 24 and
(14) Martinez, P. G.; Dessolin, G. F.; Albericio, F. J. Chem. Soc., Perkin
Trans. 1 1999, 2871.
J. Org. Chem, Vol. 71, No. 4, 2006 1325

Liley and Johnson
SCHEME 5. Synthesis of 5-(3-Allyloxy-4-formylphenoxy)pentanoic Acid (ALOBAL) and Subsequent Attachment to the Solid
Support (13)
TABLE 2. Acylation Reactions with De-ALOBAL (reaction time 64 h)
amine reductively
alkylated
coupling
reagent
coupling
additive
coupling
base
coupling
solvent
% starting amine
remaining
entry
acid coupled
1
2
3
4
5
6
7
8
9
valine-OBut
valine-OBut
valine-OMe
valine-OMe
valine-OMe
valine-OMe
valine-OBut
valine-OMe
valine-OMe
valine-OMe
valine-OMe
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
NMM
NMM
NMM
NMM
NMM
NMM
NMM
NMM
NMM
NMM
NMM
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
Z-Val-OH
Z-Leu-OH
o-anisic
10
7
17
2
10
1
2
1
1
9
5
p-anisic
o-nitrobenzoic
p-nitrobenzoic
p-nitrobenzoic
acetic
diphenylacetic
Z-Val-OH
Z-Leu-OH
10
11
FIGURE 9. Valine-OMe reductively alkylated onto ALOBAL to give
construct 17.
The reductive alkylation chemistry described earlier for valine
was found to be inappropriate for this electron poor aniline.
Therefore, an alternative method was used involving dibutyltin
dichloride as a catalyst and phenysilane as reductant.15 Reductive
alkylation was quantitative with all three linkers. Acylation was
carried out for 64 h with 4-nitrobenzoic acid. The results are
given in Table 5.
The data showed no acylated product when using the
MeOBAL or Di-MeOBAL but quantitative acylation was seen
with De-ALOBAL.
FIGURE 8. HPLC chromatogram of sample from resin cleavage (16)
(entry 6, Table 2).
64 h then followed. As before, reductive alkylation with Indole
BAL was not quantitative and so acylation experiments were
not attempted. Following cleavage of the resins, the filtrates
were analyzed by HPLC and electrospray MS. The results are
tabulated below (Table 4).
The tabulated data clearly show a difference in performance
of the three linkers. Interestingly, the rate of acylation with the
Di-MeOBAL was superior to that of the MeOBAL linker (Table
4, entry 5 vs 3 and entry 6 vs 4), although neither commercial
linker was as effective as ALOBAL.
As before, the chemistry was then repeated with the ALOBAL
directly attached to the polymer (Figure 10). Acylation was
carried out for 64 h.
Following solid-phase synthesis, the amides were isolated by
treatment of the resin with 95% TFA, 5% TES. After workup
One final test for the ALOBAL involved replacing 4-meth-
oxyaniline (25) with the less nucleophilic 4-chloroaniline (26).
(15) Apodaca, R.; Xiao, W. Org. Lett. 2001, 3, 1745.
1326 J. Org. Chem., Vol. 71, No. 4, 2006

Solid Phase Synthesis of Hindered Amides
TABLE 3. Yield and Purity of Amides Prepared with ALOBAL
amine reductively
alkylated
coupling
reagent
coupling
additive
coupling
base
coupling
solvent
compd
acid coupled
% yield
% purity
18
19
20
21
22
23
24
valine-OMe
valine-OMe
valine-OMe
valine-OMe
valine-OMe
valine-OMe
valine-OMe
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
NMM
NMM
NMM
NMM
NMM
NMM
NMM
DMF
DMF
DMF
DMF
DMF
DMF
DMF
Z-Val-OH
Z-Leu-OH
o-anisic
66
87
68
47
51
73
72
84
95
93
78
84
91
89
p-anisic
o-nitrobenzoic
p-nitrobenzoic
diphenylacetic
TABLE 4. Acylation Reactions with Various Aldehyde Linkers (reaction time 24 & 64 h)
amine reductively
alkylated
coupling
reagent
coupling
additive
coupling
base
coupling
solvent
% amine converted
(24 and 64 h)
entry
linker
acid coupled
1
2
3
4
5
6
De-ALOBAL
De-ALOBAL
2-MeOBAL
2-MeOBAL
Di-MeOBAL
Di-MeOBAL
p-anisidine
p-anisidine
p-anisidine
p-anisidine
p-anisidine
p-anisidine
HBTU
HBTU
HBTU
HBTU
HBTU
HBTU
HOBt
HOBt
HOBt
HOBt
HOBt
HOBt
NMM
NMM
NMM
NMM
NMM
NMM
DMF
DMF
DMF
DMF
DMF
DMF
Z-Val-OH
p-nitrobenzoic
Z-Val-OH
p-nitrobenzoic
Z-Val-OH
p-nitrobenzoic
100 and 100
100 and 100
11 and 28
18 and 33
16 and 53
21 and 73
TABLE 5. Acylation Reactions (64 h) with Various Aldehyde Linkers
amine reductively
alkylated
coupling
reagent
coupling
additive
coupling
base
coupling
solvent
% amine converted
(64 h)
entry
linker
acid coupled
1
2
3
De-ALOBAL
2-MeOBAL
Di-MeOBAL
p-chloroaniline
p-chloroaniline
p-chloroaniline
HBTU
HBTU
HBTU
HOBt
HOBt
HOBt
NMM
NMM
NMM
DMF
DMF
DMF
p-nitrobenzoic
p-nitrobenzoic
p-nitrobenzoic
98
2
5
TABLE 6. Yield and Purity of Amides Prepared with ALOBAL
amine reductively
alkylated
coupling
reagent
coupling
additive
coupling
base
coupling
solvent
compd
acid coupled
% yield
% purity
27
28
29
p-methoxyaniline
p-methoxyaniline
p-chloroaniline
HBTU
HBTU
HBTU
HOBt
HOBt
HOBt
NMM
NMM
NMM
DMF
DMF
DMF
p-nitrobenzoic
Z-Val-OH
p-nitrobenzoic
85
76
76
94
87
91
methods typically used in peptide chemistry for amide bond
formation. Future work will be directed toward expanding the
range of structural motifs that can be prepared by utilizing the
unique O to N acyl transfer mechanism of De-ALOBAL.
Experimental Section
Acylation Experiments. All amide bonds were formed with the
standard acylation protocol: HBTU (5 equiv), HOBt (5 equiv),
and test acid (5 equiv) were pre-dissolved in a minimum volume
of DMF and added to DMF swollen resin (1 equiv). N-Methyl-
morpholine (NMM) (10 equiv) was then added and the resin gently
agitated for 64 h. When an alternative activating agent or base was
used, exactly the same protocol was followed. Following acylation,
resins were washed (20 mL/mmol resin) with standard washing
protocol A: DMF (× 10), 5% hydrazine/DMF (10 min), DMF (×
10), MeOH, DCM (10 × each solvent) and dried for 2 h in vacuo.
Sieber Linker Cleavage. Cleavage was achieved with standard
cleavage protocol A, using 10 mg of resin with gentle agitation:
1% TFA, 5% triethysilane (TES), and 94% DCM (total volume 2
mL) for 5 min. The filtrate was treated with an equal volume of
brine and after separation, the organics were sparged with nitrogen
to obtain a sample that was immediately analyzed by RP-HPLC
and ES-MS.
FIGURE 10. Anilines reductively alkylated onto ALOBAL to give
resins (25 and 26).
the crude samples were lyophilized and accurate weights
determined. The yield and purity (HPLC) of the compounds
are shown in Table 6.
Summary
A novel linker (ALOBAL) for the solid phase synthesis of
moieties linked through an amide bond has been prepared. This
linker overcomes the limitations of similar commercially
available handles in facilitating the quantitative acylation of both
hindered (e.g., valine) and electron poor (e.g., anilines) second-
ary amines via O-N acyl transfer. The use of a double linker
strategy enabled individual steps in the overall method to be
monitored and optimized readily. Initial problems with the
reductive alkylation step have been solved and the acylation
reaction reduced to a routine coupling reaction with reagents/
Strong Acid Cleavage. Cleavage was achieved with standard
cleavage protocol B, using 1.0 g of resin with gentle agitaion: 95%
TFA, 5% TES (total volume 5 mL) for 90 min. The filtrate was
concentrated in vacuo and taken up in acetonitrile/water (1:1) before
lyophilization.
Symmetrical Anhydride Formation. Fmoc-Leu-OH (0.1 mmol,
35.3 mg) was dissolved in 2 mL of DCM. 1,3-Diisopropylcarbo-
J. Org. Chem, Vol. 71, No. 4, 2006 1327

Liley and Johnson
diimide (0.05 mmol, 7.8 µL) was added and the solution was cooled
to 0 °C for 45 min. The urea precipitate was filtered off and the
filtrate added immediately to the resin.
Preparation of Methyl 5-(3-Allyloxy-4-formylphenoxy)pen-
tanoate (11). Compound 7 (0.5 g, 1.98 mmol) was dissolved in
dry acetonitrile (10 mL). To this was added allyl bromide (1.1 equiv,
2.18 mmol, d ) 1.4, 0.188 mL) and cesium carbonate (2.0 equiv,
3.96 mmol, 1.29 g). The reaction was stirred at room temperature
for 2 h. TLC: 20% EtOAc/heptane 1 spot. Rf 0.19 (desired product).
HPLC peak 1 Rt 4.13 min (97%). The reaction mixture was filtered
through Celite, concentrated in vacuo, and then transferred to a
separating funnel. EtOAc (50 mL) was added and washed succes-
sively with 1 M KHSO4 (3 × 20 mL) and brine (1 × 30 mL). The
organic layer was dried over MgSO4 and the solvent removed in
vacuo to yield 0.527 g of a cream colored solid (91%), mp 39-40
°C. Found: C, 65.54; H, 6.85. Calcd for C16H20O5: C, 65.74; H,
6.90. νmax(film)/cm-1 2949, 1735, 1677, 1601, 1260; m/z (ESMS)
291.23 (M - H 100%), 251.20 (M - CH2-CHdCH2), 177.15 [M
- (CH2)4-CO2H), 10]. m/z (EI) found 315.1208. C16H20O5Na
requires 315.1202. δH (400 MHz; CDCl3) 10.30 (1H, s) 7.88 (1H,
d, J ) 8.5 Hz), 6.54-6.50 (1H, m), 6.45 (1H, d, J ) 2.3 Hz),
6.13-6.01 (1H, m), 5.45 (1H, dq, J ) 17.0 Hz, J ) 1.50 Hz), 5.33
(1H, dq, J ) 10.5 Hz, J ) 1.5 Hz), 4.62 (2H, dt, J ) 5.00 Hz, J
) 1.5 Hz), 4.02 (2H, t, J ) 5.85 Hz), 3.70 (3H, s), 2.42-2.39 (2H,
m), 1.85-1.71 (4H, m). δC (400 MHz; CDCl3) 188.5, 173.9, 165.6,
162.9, 132.5, 130.7, 119.4, 118.3, 106.7, 99.6, 69.4, 68.0, 51.8,
33.8, 28.7, 21.7.
Preparation of 5-(3-Allyloxy-4-formylphenoxy)pentanoic Acid
(12). The crude solid (11) (0.527 g, 2.08 mmol) was dissolved in
dioxan (10 mL) and to this was added 1 M lithium hydroxide (5
mL). The reaction was stirred at room temperature for 2 h. After
cooling, the reaction mixture was concentrated in vacuo and then
transferred to a separating funnel. EtOAc (50 mL) was added and
washed successively with H2O (3 × 25 mL) and 1 M KOH (3 ×
25 mL). The organic layer was dried over Na2SO4 and the solvent
removed in vacuo to yield a white solid (0.466 g, 93%), mp 110-
111 °C. HPLC: Peak 1 Rt 2.93 min (96%). Found: C, 65.63; H,
6.45. Calcd for C15H18O5: C, 65.74; H, 6.52. νmax(film)/cm-1 2952,
1706, 1676, 1604, 1266. m/z (ESMS) 277.25 (M - H 100%),
177.15 [M - (CH2)4-CO2H), 98.6]. m/z (EI) found 301.1052.
C15H18O5Na requires 301.1047. δH (400 MHz; CDCl3) 10.30 (1H,
s), 7.73 (1H, d, J ) 9.3 Hz), 6.62-6.58 (1H, m), 6.52-6.48 (1H,
m), 6.04-6.16 (1H, m), 5.45 (1H, dq, J ) 17.00 Hz, J ) 1.5 Hz),
5.30 (1H, dq, J ) 10.54 Hz, J ) 1.50 Hz), 4.55 (2H, dt, J ) 5.0
Hz, J ) 1.50 Hz), 4.02 (2H, t, J ) 5.85 Hz), 2.41-2.38 (2H, m),
1.88-1.76 (4H, m). δC (400 MHz; CDCl3) 188.7, 176.1, 166.4,
163.4, 132.9, 130.1, 118.8, 117.0, 107.2, 99.2, 69.2, 68.1, 33.4,
28.4, 21.5.
Synthesis of Compound 15. Resin 13 (1.0 equiv) was treated
with 0.5 equiv of (PPh3)4Pd(0) and 25 equiv of phenylsilane in
DCM for 5 min. The resin was thoroughly washed following
washing protocol B: DMF (10 × 5 mL), 0.5% DIPEA, 0.5%
sodium diethyldithiocarbamate in DMF (15 min agitation), DMF
(10 × 5 mL), DCM (10 × 5 mL). The procedure was then repeated
and finally the resin dried for 2 h in vacuo. The resin was cleaved
by using the standard cleavage protocol A to generate compound
15. HPLC: Peak 1 Rt 2.58 min (91%). m/z (ESMS) 395.22 (MH+,
100%).
Synthesis of Compound 16. Allyl protection was removed from
resin 13 as described above. The resin was then reacted with
4-nitrobenzoic acid, using the standard acylation protocol. The resin
was washed using standard washing protocol A and then cleaved
using standard cleavage protocol A. The crude material was
lyophilised from acetonitrile/water (1:1 v/v) to yield a yellow solid
(Figure 8). HPLC: Peak 1 Rt 4.60 min (96%). m/z (ESMS) 544.37
(M + H 44.17%), 488.30 [M - But), 100%]. δH (400 MHz; CDCl3)
8.29 (1H, d, J ) 8.6 Hz), 7.67 (1H, d, J ) 8.6 Hz), 7.15 (1H, d,
J ) 8.6 Hz), 6.46 (1H, s), 6.37 (1H, d, J ) 8.6 Hz), 6.04 (1H, s),
5.64 (1H, s), 4.65 (1H, d, J ) 14.8 Hz), 4.54 (1H, d, J ) 14.8 Hz),
3.90-4.01 (2H, m), 3.71 (1H, d, J ) 13.3 Hz), 2.35-2.45 (1H,
m), 2.25-2.35 (2H, m), 1.82-1.88 (2H, m), 1.49 (9H, s), 0.79
(1H, d, J ) 6.2 Hz), 0.65 (1H, d, J ) 6.2 Hz). δC (400 MHz;
Synthesis of Resin 3. 9-Fmoc-amino-xanthen-3-yloxy-Merrifield
resin (Sieber, Novabiochem) was suspended in DMF (5 min) and
then treated with a minimum volume of 20% piperidine in DMF
(1 × 2 min, 1 × 5 min). After thorough washing (DMF × 10)
linker (1) was coupled using the standard acylation protocol
described above. The resin was then washed following standard
washing protocol A and cleaved with standard cleavage protocol
A.
Synthesis of Resin 4. Synthesis of resin 4 was carried out with
standard reductiVe alkylation protocol A: The amine (either
hydrochloride salt or free base) (5 equiv) was predissolved in the
minimum volume of 1% acetic acid in DMF and added to resin 3.
Sodium cyanoborohydride (5 equiv) was then added and the resin
was agitated for 18 h. After this time the resin was washed (20
mL/mmol resin) with standard washing protocol A and dried for 2
h in vacuo.
Synthesis of Compound 5. Resin 4 was cleaved with standard
cleavage protocol A to generate compound 5 (Figure 5). HPLC:
Peak 1 Rt 2.38 min (92%). m/z (ESMS) 395.12 (MH+, 100%).
Preparation of 5-(4-Formyl-3-hydroxyphenoxy)pentanoic Acid
(6). Compound 7 (1.0 g, 3.95 mmol) was dissolved in dioxan (10
mL) and to this was added 1 M lithium hydroxide (10 mL). The
reaction was stirred at room temperature for 2 h. After cooling,
the reaction mixture was concentrated in vacuo and then transferred
to a separating funnel. EtOAc (50 mL) was added and the mixture
was washed successively with H2O (3 × 25 mL) and 1 M KOH (3
× 25 mL). The organic layer was dried over Na2SO4 and the solvent
removed in vacuo to yield a white solid (0.837 g, 89%), mp 88-
89 °C. HPLC: Peak 1 Rt 2.82 min (97%). Found: C, 60.53; H,
5.86. Calcd for C12H14O5: C, 60.50; H, 5.92. νmax(film)/cm-1 2948
(sat. CH), 1691 (CO2H), 1631 (CHO). m/z (ESMS) 237.20 (M -
H 56%), 137.09 [M - (CH2)4-CO2H), 100]. m/z (EI) found
261.0739. C12H14O5Na requires 261.0732. δH (400 MHz; CDCl3)
9.78 (1H, s), 9.70 (1H, s), 7.51 (1H, d, J ) 8.5 Hz), 6.53 (1H, dd,
J ) 8.5, 2.3 Hz), 6.39 (1H, d, J ) 2.3 Hz), 4.00 (2H, t, J ) 5.8
Hz), 2.40-2.37 (2H, m), 1.85-1.71 (4H, m). δC (400 MHz; CDCl3)
194.5, 176.2, 166.6, 164.2, 135.1, 115.6, 108.2, 100.9, 68.1, 33.3,
28.3, 21.5.
Preparation of Methyl 5-(4-Formyl-3-hydroxyphenoxy)pen-
tanoate (7). 2,4-Dihydroxybenzaldehyde (10 g, 7.5 mmol), methyl
5-bromopentanoate (12.9 mL, 1.2 equiv, 9 mmol), and spray dried
potassium fluoride (8.85 g, 1.0 equiv, 7.5 mmol) were dissolved
in dry acetonitrile (100 mL) and heated under reflux for 18 h
(moisture excluded with calcium chloride drying tube). After
cooling, the reaction mixture was concentrated in vacuo and then
transferred to a separating funnel. EtOAc (150 mL) was added and
washed successively with H2O (3 × 75 mL) and 1 M KOH (3 ×
75 mL). The organic layer was dried over Na2SO4 and the solvent
removed in vacuo to yield a brown solid (16.3 g, 75%). TLC:
30%EtOAc/70%heptane: 3 spots, minor 1 Rf 0.32 (starting mate-
rial), minor 2 Rf 0.1 (acid, hydrolyzed from methyl ester), major Rf
0.36 desired product. HPLC: Peak 1 Rt 2.81 min (17%), peak 2 Rt
2.81 min (30%), peak 3 Rt 3.78 min (51%). Purification by flash
chromatography. The product was dry loaded and eluted with 20%
EtOAc/heptane. Appropriate fractions were combined and concen-
trated in vacuo to afford the three compounds stated above, all as
white/cream solids. Desired product (7.6 g, 42%), mp 65-66 °C.
Found: C, 61.90; H, 6.37. Calcd for C13H16O5: C, 61.90; H, 6.39.
ν
max(film)/cm-1 2949, 1734, 1630. m/z (ESMS) 251.23 (M - H
100%), 137.10 [M - (CH2)4-CO2H), 11.83]. m/z (EI) found
275.0895. C13H16O5Na requires 275.0890. δH (400 MHz; CDCl3)
11.45 (1H, s), 9.70 (1H, s), 7.40 (1H, d, J ) 8.50 Hz), 6.50 (1H,
dd, J ) 8.5 Hz, J ) 2.3 Hz), 6.40 (1H, d, J ) 2.3 Hz), 4.00 (2H,
t, J ) 5.8 Hz), 3.70 (3H, s), 2.42-2.38 (2H, m), 1.88-1.76 (4H,
m). δC (400 MHz; CDCl3) 188.5, 173.9, 165.6, 162.9, 130.7, 119.4,
106.7, 99.6, 68.0, 51.8, 33.8, 28.9, 21.7.
1328 J. Org. Chem., Vol. 71, No. 4, 2006

Solid Phase Synthesis of Hindered Amides
CDCl3) 176.5, 173.5, 168.5, 161.1, 157.9, 148.9, 141.4, 133.6,
128.8, 124.1, 113.9, 106.9, 103.1, 83.6, 69.8, 67.6, 43.9, 35.6, 28.71,
28.4, 28.2, 22.4, 20.0, 18.9.
The aniline (5 equiv) and dibutyltin dichloride (0.1 equiv) were
dissolved in the minimum volume of dry THF and added to the
resin. After gentle agitation (5 min) phenylsilane (5 equiv) was
added and the resin agitated for 18 h. After this time the resin was
washed (20 mL/mmol resin) using standard washing protocol A.
Synthesis of Compound 29.22 Resin 26 was treated with 0.5
equiv of (PPh3)4Pd(0) and 25 equiv of phenylsilane in DCM for 5
min. The resin was thoroughly washed using standard washing
protocol B and the procedure repeated. 4-Nitrobenzioc acid was
coupled using the standard acylation protocol. The resin was once
again washed using standard washing protocol A and after drying
in vacuo was subjected to the standard cleavage protocol B. The
crude material was lyophilized from acetonitrile/water to yield a
yellow solid (0.042 g, 76%). HPLC: Peak 1 Rt 5.22 min (99%).
m/z (ESMS) 277.45 (M + H 43.37%). m/z (EI) found 299.6593.
C19H28N2O5Na requires 299.6652. δH (400 MHz; DMSO) 8.35 (2H,
d, J ) 8.78 Hz), 8.16 (2H, d, J ) 9.28 Hz), 7.80 (2H, d, J ) 8.78
Hz), 4.42 (2H, d, J ) 9.28 Hz). δC (400 MHz; CDCl3) 164.6, 149.9,
141.0, 138.4, 129.9, 129.32, 128.5, 124.2, 122.7.
Synthesis of Resin 17. NovaSyn (Novabiochem) TGR (1.0 g,
0.2 mmol) was suspended in DMF (5 min) and then treated with a
minimum volume of 20% piperidine in DMF (1 × 2 min, 1 × 5
min). After thorough washing (DMF ×10) linker 12 was coupled
using the standard acylation protocol. The resin was then washed
using standard washing protocol A. After pre-swelling in DMF,
HCl.Val-OMe was reacted using standard reductive alkylation
protocol A and then washed using standard washing protocol A.
Synthesis of Compound 18.16 Resin 17 was treated with 0.5
equiv of (PPh3)4Pd(0) and 25 equiv of phenylsilane in DCM for 5
min. The resin was thoroughly washed using standard washing
protocol B and the procedure repeated. Z-Valine was coupled using
the standard acylation protocol. The resin was once again washed
using standard washing protocol A and then subjected to the
standard cleavage protocol B. The crude material was lyophilised
from acetonitrile/water to yield a clear oil (0.048 g, 66%). HPLC:
Peak 1 Rt 4.10 min (99%). m/z (ESMS) 365.03 (M + H 37.27%).
m/z (EI) found 387.1904. C19H28N2O5Na requires 387.1896. δH (400
MHz; CDCl3) 7.38-7.22 (5H, m), 5.32 (1H, s) 5.16 (2H, m), 4.56
(1H, m), 4.31 (1H, m), 3.73 (3H, s), 2.08 (2H, m), 1.00-0.86 (12H,
m). δC (400 MHz; CDCl3) 173.0, 172.6, 156.8, 129.8-129.2, 67.4,
59.1, 57.8, 52.6, 30.8, 19.7-18.2.
Supporting Information Available: Full characterization of all
numbered products along with detailed procedures for the additional
reactions.17-21 This material is available free of charge via the
Internet at http://pubs.acs.org.
JO051742J
Synthesis of Resin 26. NovaSyn (Novabiochem) TGR (1.0 g,
0.2 mmol) was suspended in DMF (5 min) and then treated with a
minimum volume of 20% piperidine in DMF (1 × 2 min, 1 × 5
min). After thorough washing (DMF ×10) linker 12 was coupled
using the standard acylation protocol. The resin was then washed
using standard washing protocol A. After preswelling in DMF,
4-chloroaniline was reacted using reductive alkylation protocol B:
(17) Govindachari, T. R.; et al. Tetrahedron 1967, 23, 4811-4815.
(18) Suga, H.; Shi, X.; Ibata, T.; Kakehi, A. Heterocycles 2001, 9, 1711-
1726.
(19) Fache, F.; Valot, F.; Milenkovic, A.; Lemaire, M. Tetrahedron 1996,
52, 9777-9784.
(20) Mitas, P.; Sedlak, M.; Kavalek, J. J. Chem. Soc., Commun. 1998,
63, 85-93.
(21) Mathis, C. A.; Wang, Y.; Holt, D. P.; Huang, G.-F.; Debnath, M.
L.; Klunk, W. E. J. Med. Chem. 2003, 46, 2740-2754.
(22) Walczyna, R.; Sokolowski, J. Pol. J. Chem. 1984, 58, 791-804
(16) Breit, B.; Laungani, A. C. Tetrahedron: Asymmetry 2003, 14, 3823-
3826.
J. Org. Chem, Vol. 71, No. 4, 2006 1329