The preparation of a new "safety catch" ester linker for solid-phase synthesis

Article abstract of DOI:10.1021/jo0012086

A new "safety catch" linker for esters has been synthesized on polystyrene resin. This 2-tert-butoxyphenol resin 10 may be acylated to give a relatively stable ester that will allow nucleophilic chemistry without reaction at the linking ester group. Removal of the tert-butyl group with acid unmasks a highly reactive 2-hydroxyphenyl ester that reacts readily with nucleophiles to cause release of the product from the resin. This sequence has been exemplified by acylating the resin with various bromo acids, carrying out nucleophilic displacements with thiols, phenols, or amines, activating the ester with trifluoroacetic acid and cleaving from the resin with amines to give the (nucleophile) substituted carboxamides in high yield and purity. Kinetic studies with a model ester revealed half-lives for reaction with morpholine of 119 h for the tert-butoxyphenyl ester and 1 min for the corresponding phenol.

Full text of DOI:10.1021/jo0012086

2240
J . Org. Chem. 2001, 66, 2240-2245
Th e P r ep a r a tion of a New “Sa fety Ca tch ” Ester Lin k er for
Solid -P h a se Syn th esis
Claire L. Beech, J ohn F. Coope, Gary Fairley, Philip S. Gilbert, Brian G. Main,* and
Karen Ple´†
AstraZeneca Pharmaceuticals Ltd., Alderley Park, Macclesfield, Cheshire SK10 4TG, United Kingdom
brian.main@astrazeneca.com
Received August 7, 2000
A new “safety catch” linker for esters has been synthesized on polystyrene resin. This 2-tert-
butoxyphenol resin 10 may be acylated to give a relatively stable ester that will allow nucleophilic
chemistry without reaction at the linking ester group. Removal of the tert-butyl group with acid
unmasks a highly reactive 2-hydroxyphenyl ester that reacts readily with nucleophiles to cause
release of the product from the resin. This sequence has been exemplified by acylating the resin
with various bromo acids, carrying out nucleophilic displacements with thiols, phenols, or amines,
activating the ester with trifluoroacetic acid and cleaving from the resin with amines to give the
(nucleophile) substituted carboxamides in high yield and purity. Kinetic studies with a model ester
revealed half-lives for reaction with morpholine of 119 h for the tert-butoxyphenyl ester and 1 min
for the corresponding phenol.
In tr od u ction
imposing steric hindrance on the ester group. Thus, ortho
benzyloxy8 and phenacyloxy9 groups reduce reactivity
markedly. Conversely, the introduction of o-hydroxy
groups into aryl esters increases the reactivity by some
degree due to anchimeric assistance of the hydroxy group
during reaction with nucleophiles, possibly by stabiliza-
tion of the tetrahedral intermediate. Cowell and J ones10
and Merrifield et al.11 have both, therefore, explored
protected o-hydroxyphenyl esters as safety catch linkers.
The first group used the benzyloxyphenyl ester and
released the free phenol by hydrogenolysis, while Mer-
rifield used the tert-butoxyphenyl ester and activated
proteolytically with trifluoroacetic acid (TFA). He also
made mention of application of these latter esters in cyclic
peptide synthesis on solid phase, but with no great
success and with no details given of the type of linker
employed.11
We wish to report our sucessful synthesis of a resin-
linker-tert-butoxyphenol system that is capable of being
esterified, is moderately stable to nucleophiles, and that
may be activated with TFA to give an extremely reactive
o-hydroxyphenyl ester which is cleaved rapidly by nu-
cleophiles. For convenience, we have called this “BuPhe”
resin. The use of this resin to make simple chemical
libraries will also be described.
The ester group is one of the most common types of
linker used in solid-phase synthesis, linking a substrate
to a polymeric support.1 The types of ester used vary in
their reactivity toward nucleophiles from the extremely
stable Merrifield2 and Wang3 esters to Kaiser’s oximino
esters used to promote peptide aminolysis.4 Others, such
as polystyrylmethylthiophenoxy esters,5 are known, but
their degree of reactivity has not been quantified beyond
the anecdotal. We were interested in designing an ester
that would be stable to nucleophiles during the course
of a synthesis, but would then be capable of activation
to a highly reactive state at the end of a synthesis when
release from the resin was required, that is a so-called
“safety catch” linker.
The concept of safety catch linkers has been explored
widely; for example, Kenner6 developed a peptide syn-
thesis by coupling amino acids to a sulfonamide resin as
a stable acyl sulfonamide that deprotonates in basic
conditions and is then stable toward nucleophilic attack.
Subsequent N-alkylation of the sulfonamide using dia-
zomethane produces a readily cleavable acyl derivative.
This concept was extended later by Ellman’s group, who
used iodomethane or iodoacetonitrile to activate the
sulfonamide linker.7 The whole concept of safety catch
linkers is discussed in J ames’ review.1
Aryl esters are reactive toward nucleophiles, but
several groups have sought to reduce this reactivity by
Resu lts a n d Discu ssion
The 2-tert-butoxyphenol 3 may be prepared in large
quantities from catechol by the route shown in Scheme
1, this being a modification of Merrifield’s method.11
Reaction of 3 (Scheme 2) with diazotized 4-amino benzene
sulfonic acid gave a water-soluble azo derivative that, in
* To whom correspondence should be addressed. Phone: +44 1625
512330. Fax: +44 1625 586707.
†CNRS-UPRESA 6013, CPCBAI, Moulin de la Housse, Baˆt 18, BP
1039, 51097 Reims Cedex 2, France.
(1) J ames, I. W. Tetrahedron 1999, 55, 4855-4946.
(2) Merrifield, R. B. J . Am. Chem. Soc. 1963, 85, 2149.
(3) Wang, S.-W. J . Am. Chem. So.c 1973, 95, 1328.
(4) DeGrado, W. F.; Kaiser, E. T. J . Org. Chem. 1980, 45, 1295.
(5) Marshall, D. L.; Liener, I. E. J . Org. Chem. 1970, 35, 867.
(6) Kenner, G. W.; McDermott, J . R.; Sheppard, R. C. J . Chem. Soc.,
Chem. Commun. 1971, 636-7.
(8) J ones, J . H.; Young, G. T. J . Chem. Soc. C 1968, 436-41.
(9) Trudelle, Y. J . Chem. Soc., Chem. Commun. 1971, 639-41.
(10) Cowell, R. D.; J ones, J . H. J . Chem Soc. C 1971, 1082-1090.
(11) Mitchell, A. R.; Altaf-Ur Rahman; Merrifield, R. B. Peptides
1990, 39-40; Giralt, E., Andreu, D., Eds.; Escom Science Publishers
BV: Leiden, The Netherlands, 1991.
(7) Backes, B. J .; Ellman, J . A. J . Am. Chem. Soc. 1996, 118, 3055-
6.
10.1021/jo0012086 CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/14/2001

“Safety Catch” Ester Linker for Solid-Phase Synthesis
J . Org. Chem., Vol. 66, No. 7, 2001 2241
Sch em e 1
Sch em e 3
Sch em e 2
Ta ble 1. Bu ild in g Block s Used
situ, was reduced with sodium dithionite to give the
amine 4. This could be extracted from the reaction
mixture directly, but it proved advantageous to include
benzaldehyde in order to form the Schiff base 5, in situ,
and provide better extraction of the imine, which was also
more stable than the aminophenol. Benzoylation, fol-
lowed by hydrolysis of the imine, gave very pure protected
aniline 7, directly suitable for coupling to the resin bound
acid 8.
This acid 8 was easily prepared from chlorometh-
ylpolystyrene and ethyl 4-hydroxyphenyl propionate,
followed by hydrolysis with sodium hydroxide. After the
aniline 7 was coupled to this resin, the benzoate 9 was
converted to free phenol 10 (“BuPhe” resin) using sodium
methoxide in methanol/THF. When chloromethylpoly-
styrene with a loading of 3.6 mmol/g was used, the
resulting resin contained approximately 1.8 mmol phenol/
g.
Esters of the BuPhe resin 12 could be prepared
(Scheme 3) using either acid chlorides or carboxylic acids
coupled with 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyl-
uronium hexafluorophosphate (HBTU). Various bro-
moesters of the resin were prepared, and these were then
allowed to react with thiols, phenols, or amines to give
the corresponding thioethers, ethers, or amines 13 (X )
S, O, N), without cleavage of the ester bond. Activation
with TFA to give the phenol 14 was followed by nucleo-
philic cleavage from the resin using amines, the amide
products 15 being very pure, typically better than 90%,
without any need for purification.
Examples of the building blocks used are shown in
Table 1, and as shown in Table 2, the synthesis works
with R- and ω-haloesters, along with esters containing a
benzyl halide, although reaction of alkanethiols with the
unreactive ω-haloesters causes an unacceptably high
degree of ester cleavage. Both primary and secondary
amines may be used to cleave the activated esters, but
anilines do not work under the conditions studied. In
many cases, the aminolysis could be performed using a
sub-equivalent amount of amine (relative to the loading
of ester on the resin), leading to complete loss of amine
from the reaction solution; alternatively, a slight excess
could be used, followed by treatment of unreacted amine

2242 J . Org. Chem., Vol. 66, No. 7, 2001
Beech et al.
Ta ble 3. Kin etic Stu d ies: Ra te of Rea ction of
Ta ble 2. Sa m p le Libr a r y
Resin -Bou n d Ester s w ith Mor p h olin e
most reactive studied being the strongly electron-
withdrawing sulfone 16e, which is 300 times more
reactive than the ether 16c, with the thioether 16d
having intermediate reactivity. The sulfone shows similar
reactivity to the ester of Kaiser’s oxime 16f, and both of
these are very synthetically useful solid-phase active
ester systems, although it would not be possible to do
any nucleophilic chemistry on their derivatives without
causing cleavage from the resin.
The BuPhe resin derivative 16g is relatively resistant
to attack by piperidine, while the corresponding phenol
16h is more reactive by many orders of magnitude.
Useful chemistry can thus be performed without ester
cleavage, followed by nucleophilic release from the resin
after activation by TFA. We have illustrated this prin-
ciple above by creating several simple libraries from
bromoesters, performing the nucleophilic displacement
of the bromine without loss from the resin and then
removing the tert-butyl group and cleaving the products
as amides in very high yield and purity. The residual
resin is a catechol and, therefore, cannot be realkylated
and reused.
The balance of reactivity of the BuPhe resin esters in
the activated vs nonactivated states is governed critically
by the electronic nature of the linking group. Thus, the
nonactivated ester, with an electron neutral (σ 0) acy-
lamino group in the para position to the phenol group, is
700 times less reactive than the sulfone resin 16e, with
the electron-withdrawing (σ 0.72) group in the same
position. It would be possible to make the nonactivated
ester more resistant to nucleophilic attack by adding
either another ortho substituent, or another electron-
a
Not purified.
with a scavenger resin12 such as isocyanatomethyl poly-
styrene.
Kin etic Stu d ies. To assess more accurately the rela-
tive reactivities of our esters, we decided to quantify the
reactivity of a series of known esters. To this end, a series
of resin bound 4-methoxyphenylacetic esters 16 was
prepared (Table 3) and the rate of aminolysis was
measured using a 50-fold excess of piperidine. The
formation of 4-methoxyphenylacetyl piperidide was quan-
tified using HPLC, using standard solutions for calibra-
tion, to determine reaction rate and half-life. Because of
the large excess of nucleophile the reactions were as-
sumed to have pseudo-first-order kinetics and were
analyzed using the Enzfitter program developed for
enzyme kinetics.13 This allowed calculation of a half-life.
The reactivity of resin-bound esters toward nucleophilic
cleavage varies widely, as shown in Table 3. Merrifield
esters 16a are completely inert to the reaction conditions
used, and Wang esters 16b are little different. Various
aryl esters show greater reactivity, the degree depending
on the electronic character of the resin-phenol link, the
(12) Kaldor, S. W.; Siegel, M. G. Curr. Opin. Chem. Biol. 1997, 1,
101.
(13) Leatherbarrow, R. J 1987 Enzfitter, Elsevier Biosoft. For a
review, see: Leatherbarrow, R. J . Trends in Biochemical Science 1990,
455-8.

“Safety Catch” Ester Linker for Solid-Phase Synthesis
J . Org. Chem., Vol. 66, No. 7, 2001 2243
rt. At this stage, if desired, the reaction could be purged with
argon and kept overnight before proceeding. Aqueous NaOH
solution (40%, 50 mL) was now added, the reaction heated to
90 °C, and Na2S2O4 (126.3 g, 0.726 mol) added portionwise at
a rate to control both exotherm and effervescence until the
red color was discharged (sometimes slightly more or less than
the above amount). After cooling, the pH was adjusted to 7
with a little HCl, benzaldehyde (30.5 mL, 0.30 mol) was added,
and the mixture stirred vigorously for 30 min. The resulting
imine was extracted into CH2Cl2 (1 × 400 mL, 2 × 200 mL),
and the combined extracts were dried (MgSO4) and evaporated
to give 5 as a crude brown solid, wt 57.14 g, 71% yield. The
product proved too unstable to purify but MS showed a peak
at 270 (M + H)+, and 1H NMR (CDCl3) showed the expected
peaks at δ 1.46 (s, 9H), 5.73 (s, 1H), 6.97 (m, 3H), 7.45 (m,
3H), 7.90 (m, 2H), 8.44 (s, 1H) ppm.
donating group in the ring, but in both cases this would
also reduce the reactivity of the ester after activation.
We are currently looking at further uses for esters of
BuPhe resin.
Exp er im en ta l Section
1H NMR spectra were determined at 300 MHz, and 13C
NMR spectra of resins were determined at 400 MHz as gels
using a magic angle spinning probe (Bruker). Mass spectra
were determined using positive ion electrospray, and HRMS
was also performed using electrospray ionization (Quattro),
calibrated with PEG 200/400/600, and looking at peaks within
10 ppm. Evaporations were performed on a rotary evaporator.
For resin products combustion analysis proved satisfactory and
reproducible for N, S, and halogens, but C and H were too
variable to be useful.
2-Hyd r oxyp h en yl Ben zoa te (1). A mixture of catechol (22
g, 0.20 mol) and benzoic acid (48 g, 0.40 mol) was mixed with
polyphosphoric acid (500 g), heated 30 min on a steam bath,
and then poured into 1 L of H2O. The resulting solid was
collected, slurried with saturated NaHCO3 (200 mL) twice, and
washed with H2O (2 × 200 mL). After drying, the product was
recrystallized from toluene to give 19.6 g (36%) of 1 as a white
solid. Mp: 130 °C. Anal. Calcd for C13H10O3: C, 72.89; H, 4.71.
Found: C, 72.70; H, 4.81. MS: 215 (M + H)+. 1H NMR
(DMSO): δ 6.8 (m, 1H), 7.0 (m, 1H), 7.15 (m, 2H), 7.6 (m, 2H),
7.7 (m, 1H), 8.15 (m, 2H), 9.7 (s, 1H) ppm.
4-Am in o-2-ter t-bu toxyp h en yl Ben zoa te, Ben za ld eh yd e
Im in e (6). Phenol 5 (57.0 g, 0.21 mol) was dissolved in CH2-
Cl2 (550 mL) and cooled to 0 °C. Et3N (87.6 mL, 0.63 mol) was
added, followed by dropwise addition of benzoyl chloride (26.4
mL, 0.21 mol) in CH2Cl2 (100 mL) over 1 h. The reaction was
then allowed to warm to rt. After 1 h, the reaction was poured
into H2O (1 L) with rapid stirring for 30 min, the CH2Cl2 layer
was separated and the aqueous layer extracted with CH2Cl2
(2 × 250 mL). The combined extracts were dried and evapo-
rated to give a light brown solid, 80.09 g (>100% yield). This
was dissolved in the minimum amount of CH2Cl2 and passed
down a column of alumina (ICN Alumina N, Ak. I), 175 mm
bed on a 70 mm diameter column. Elution with 5% EtOAc/
isohexane, followed by evaporation, gave a yellow solid, wt
The very viscous polyphosphoric acid mixture is difficult to
stir on a large scale and may be substituted by Eaton’s reagent
(10% P2O5 in MeSO3H) using 40 mL/g of catechol and stirring
for 3 h at ambient temperature. A similar work up gives a
rather oily product which is best purified by flash chromatog-
raphy using CH2Cl2 to give a 35-40% yield of product.
2-ter t-Bu toxyp h en yl Ben zoa te (2). The phenol 1 (21.4 g,
0.10 mol) was dissolved in 200 mL of CH2Cl2 and cooled to
-40 °C under an argon atmosphere. Isobutene (56 g, 1.0 mol)
was added over 1 h and the mixture cooled to -50 °C.
Trifluoromethanesulfonic acid (0.5 mL) was added carefully
(mild exotherm) and the reaction stirred for 2.5 h at -45 to
-50 °C. Triethylamine (Et3N, 0.8 mL) was then added to
quench the reaction, and it was allowed to warm to rt
overnight. Evaporation of the solvent gave the product 2 as
an oil (27 g, 100%), which was not purified. A small sample
was purified on silica gel using 10% EtOAc/isohexane to give
pure 2 as a colorless oil that crystallized on standing. Mp: 63
°C. Anal. Calcd for C17H18O3: C, 75.53; H, 6.71. Found: C,
75.01; H, 6.66. MS: 271 (M + H)+. 1H NMR (DMSO): δ 1.2 (s,
9H), 7.1-7.3 (m, 4H), 7.6-7.8 (m, 3H), 8.1 (d, 2H) ppm.
2-ter t-Bu toxyp h en ol (3). The ester 2 (27 g, 0.10 mol) was
dissolved in EtOH (240 mL), and aqueous NaOH (40%, 14 mL)
was added (slight exotherm). The solution was refluxed 45 min,
15 mL of water added, and then the mixture was refluxed a
further 5 h. After cooling, the suspension was diluted with 30
mL of water and the solution evaporated to a dark oil. This
was shaken with water (220 mL) and isohexane (220 mL) with
the pH adjusted to 9 with AcOH. The extracts were dried
(MgSO4) and evaporated. The resulting oil was purified on a
70 g silica gel column using 5-10% EtOAc/isohexane, giving
15.3 g (92%) of pure 3 (alternatively the product may be
distilled, bp 60 °C/0.4 mmHg). Overall yield was 33% from
catechol. Anal. Calcd for C10H14O2: C, 72.26; H, 8.49. Found:
C, 72.22; H, 8.53. MS: 167 (M + H)+. 1H NMR (DMSO): δ 1.3
(s, 9H), 6.7 (m, 1H), 6.8-7.8 (m, 3H), 8.6 (s, 1H) ppm.
4-Am in o-2-ter t-bu toxyp h en ol, Ben za ld eh yd e Im in e (5).
A mixture of sulfanilic acid (57.15 g, 0.33 mol), Na2CO3 (17.49
g, 0.165 mol), and H2O (600 mL) was stirred and cooled to 0
°C. NaNO2 (22.77 g, 0.33 mol) was added, followed by 5.5 M
HCl (165 mL), added below the liquid surface over 30 min at
4-9 °C. After being stirred for a further 30 min at 0 °C, the
reaction mixture was poured into a (previously prepared and
cooled to 0 °C) solution of 2-tert-butoxyphenol (50.0 g, 0.30 mol)
in 10% NaOH (375 mL). After 5 min at 0 °C, the deep red
solution was stirred for a further 1 h and allowed to warm to
1
61.15 g (78% yield). MS: 374 (M + H)+. H NMR (CDCl3): δ
1.30 (s, 9H), 6.97 (dd, 1H), 7.04 (d, 1H), 7,20 (d, 1H), 7.47 (m,
5H), 7.62 (m, 1H), 7.92 (m, 2H), 8.22 (m, 2H), 8.46 (s, 1H) ppm.
4-Am in o-2-ter t-bu toxyp h en yl Ben zoa te (7). The imine
6 (61.08 g, 0.164 mol) was dissolved in CH2Cl2 (620 mL) and
the mixture stirred vigorously with a paddle stirrer. HCl (2
M, 620 mL) was added and the stirring continued for 1.25 h,
adding a further 400 mL of CH2Cl2 after 45 min. Vigorous
mixing of the two layers was essential. The white solid that
separated was collected, washed with EtOAc (3 × 100 mL),
sucked as dry as possible, and then dried under vacuum at rt
to give 7 as an almost white solid, wt 38.63 g (73% yield). MS:
286 (M + H)+. 1H NMR (CDCl3): δ 1.24 (s, 9H), 7.03 (dd, 1H),
7.23 (s, 1H), 7.34 (d, 1H), 7.61 (t, 2H), 7.77 (t, 1H), 8.12 (d,
2H) ppm. The product did not analyze well, and so a portion
was converted to the free base using NaHCO3/EtOAc extrac-
tion, and the resulting solid was analyzed. Anal. Calcd for
C
17H19NO3: C, 71.56; H, 6.71; N, 4.91. Found: C, 71.35; H,
6.61; N, 4.92.
4-(Ca r boxyeth yl)p h en oxym eth yl P olystyr en e (8). Chlo-
romethylpolystyrene resin (Polymer Labs, 150-300 µm, 3.47
mmol/g, 20 g, 69.4 mmol) was shaken in DMF (200 mL), and
ethyl 3-(4-hydroxyphenyl)propionate (16.85 g, 87 mmol) was
added, followed by NaOMe (5.06 g, 94 mmol). The mixture was
shaken 24 h at 50 °C, cooled, and filtered. The resin was
washed with DMF (3 × 200 mL), THF (1 × 200 mL), 50%
aqueous THF (3 × 200 mL), THF (3 × 200 mL), MeOH (3 ×
200 mL), and Et2O (3 × 200 mL) before drying under vacuum.
The product weighed 29.54 g (95% theoretical) with no
detectable Cl by microanalysis.
This resin was suspended in THF (300 mL), 10% aqueous
NaOH (40 mL, 0.10 mol) was added, and the mixture was
shaken 18 h at 40 °C. After cooling, the mixture was filtered,
and the resin was washed with THF (3 × 300 mL), 50%
aqueous THF (2 × 300 mL), 1:1 THF/2 M HCl (2 × 300 mL),
50% aqueous THF (2 × 300 mL), THF (3 × 300 mL), MeOH
(3 × 300 mL), and Et2O (3 × 300 mL) before drying under
vacuum to give 29.18 g resin (theory 29.02 g). FT-IR (CHCl3
gel): CdO 1713 cm-1 13C NMR (CDCl3): δ 29.79, 35.79, 69.84,
.
157.38, 179.06 ppm.
4-(P olystyr ylm eth oxy)p h en ylp r op ion ic Acid (4-Ben -
zoyloxy-3-ter t-bu toxy)a n ilid e (9). The resin 8 (50 g, 0.09
mol) was swollen in dry N-methylpyrrolidone (NMP, 400 mL)
under an argon atmosphere, HBTU (37.5 g, 0.10 mol), HOBT
(13.4 g, 0.10 mol), and DIPEA (33.9 mL, 25.15 g, 0.20 mol)

2244 J . Org. Chem., Vol. 66, No. 7, 2001
Beech et al.
were added, and the mixture was shaken for 1 h at 50 °C. The
aniline hydrochloride 7 (31.8 g, 0.10 mol) was then added and
shaking continued for 20 h at 50 °C. After cooling, the reaction
was filtered and the product washed with NMP/H2O (90/10, 2
× 400 mL), THF (3 × 400 mL), 50% aqueous THF (3 × 400
mL), THF (3 × 400 mL), MeOH (3 × 400 mL), and Et2O (3 ×
400 mL) before drying under vacuum. The whole procedure
was then repeated (double coupling) and resulted in a pale
beige resin, wt 73.9 g (theory 73.9 g). FT-IR: CdO 1686 and
1732 cm-1. Anal. N calcd as 2.04% for 1.46 mmol/g, found
1.91%, equals 1.37 mmol/g. 13C NMR (CDCl3): δ 28.5, 30.4,
35.7, 68.8, 80.3, 114.5, 115.0, 122.3, 136.2, 140.6, 146.9, 157.2,
165.1, 171.1 ppm.
44.3, 113.9, 115.4, 123.1, 126.3, 129.4, 130.0, 130.7, 133.7,
134.3, 136.3, 138.5, 140.3, 146.1, 148.4, 158.4, 170.5, 171.2
ppm.
2-(2-Ch lor op h en ylth io)p r op ion ic Acid Mor p h olid e 15g
(R1 ) CHMe, X ) S, R2 ) 2-Cl-P h , R3 ) (CH2)2O(CH2)2, R4
) H). The resin 13 (R1 ) CHMe, R2XH ) 2-Cl-C6H4SH, 1.20
g, 1.27 mmol) was treated with 10% trifluoroacetic acid (TFA)
in CH2Cl2 (25.0 mL) and triisopropylsilane (TIPS, 600 µL) for
0.5 h. The resin was filtered and washed with CH2Cl2 (2 × 15
mL), 10% Et3N in THF (2 × 15 mL), and THF (3 × 15 mL),
suspended in THF (15 mL), and morpholine (1.0 mL, 11.5
mmol) was added. After shaking for 18 h, the suspension was
filtered and evaporated to give 365 mg of crude product (100%
yield), 90% pure by HPLC, purified by flash chromatography
(CH2Cl2) to give 319 mg (88%) of pure product. A portion of
this was triturated with isohexane, when white crystals were
formed; they were collected, washed, and dried. Anal. Calcd
for C13H16ClNO2S: C, 54.64; H, 5.64; N, 4.90; S, 11.22.
Found: C, 54.53; H, 5.65; N, 4.75; S, 11.20. MS: 286 (M +
H)+. 1H NMR: δ 1.4 (d, 3H), 3.3-3.7 (m, 8H), 4.1 (q, 1H), 7.15
(m, 2H), 7.4 (dd, 1H), 7.5 (dd, 1H) ppm.
4-(P olyst yr ylm et h oxy)p h en ylp r op ion ic Acid (4-H y-
d r oxy-3-ter t-bu toxy)a n ilid e (10) “Bu P h e” Resin . The resin
9 (73.9 g, 0.09 mol) was shaken with THF (1200 mL), and
saturated NaOMe in MeOH (76 mL) was added over 0.5 h.
The reaction was stirred for 4 h and filtered, and the product
was washed with 50% aqueous THF (3 × 400 mL), THF/1 M
HCl (50/50, 3 × 400 mL), 50% aqueous THF (3 × 400 mL),
THF (3 × 400 mL), MeOH (3 × 400 mL), and Et2O (3 × 400
mL) before drying under vacuum. The light brown resin
weighed 65.45 g (theory 64.64 g). FT-IR (CHCl3 gel): CdO
1660 cm-1, OH 3520 cm-1. Anal. for N calcd as 2.41% for 1.72
mmol/g, found, 2.12%, equals 1.55 mmol/g. 13C NMR (CDCl3):
δ 29.4, 30.6, 35.7, 69.8, 80.6, 114.4, 114.6, 115.5, 116.0, 129.1,
141.7, 145.0, 146.4, 157.2, 171.1 ppm.
2-(4-Meth ylth io-3-m eth ylp h en oxy)p r op ion ic Acid (3-
P h en ylp r op yl)a m id e (15h , R1 ) CHMe, X ) O, R2
)
4-MeS-3-Me-P h , R3 ) P h (CH2)3, R4 ) H). Under argon,
cesium carbonate (1.076 g, 3.3 mmol) and 4-(methylthio)-3-
methylphenol (254 mg, 1.65 mmol) were shaken in dry NMP
(10 mL) for 1 h. Resin 13 (R ) CHMe) (500 mg, 0.55 mmol)
was added and the mixture shaken for 5 h. The resin was
filtered, washed with H2O (3 × 20 mL), THF (3 × 20 mL),
and MeOH (3 × 20 mL), and dried to give a brown resin, wt
445 mg. Anal. Calcd for 1.02 mmol/g: N, 1.42; Br, absent; S,
3.26. Found: N, 1.81; Br, <0.3; S, 1.41; corresponding to 0.44
mmol/g (some cleavage from resin).
Gen er a l P r oced u r e for Cou p lin g of Bu P h e Resin w ith
Br om oa cid s. A typical procedure is as follows using 2-bro-
mopropionic acid:
(a ) Via th e Acid Ch lor id e. The resin 10 (2.8 g, 5.0 mmol)
was stirred in CH2Cl2 (35 mL) under an argon atmosphere,
and Et3N (2.78 mL, 2.02 g, 20 mmol) was added, followed by
2-bromopropionyl chloride (2.02 mL, 3.43 g, 20 mmol), in CH2-
Cl2 (15 mL) added over 30 min (exotherm), keeping the temp
below 25 °C. The mixture was shaken for 6 h at rt, filtered,
and washed with CH2Cl2 (3 × 30 mL), THF (3 × 30 mL), 50%
aqueous THF (3 × 30 mL), THF (3 × 30 mL), MeOH (3 × 30
mL), and Et2O (3 × 30 mL), before drying under vacuum to
give a light brown free flowing resin, wt 3.7 g. Microanalysis
gave N, 1.8%, Cl, trace, Br, 10.8%, from which the loading was
This resin (365 mg, ca. 0.16 mmol) was shaken with TFA/
CH2Cl2 (5 mL of 10%) and TIPS (150 µL) for 30 min and
washed with CH2Cl2 (3 × 10 mL), 1 M Et3N/CH2Cl2 (3 × 10
mL) and THF (3 × 10 mL). DMSO (5 mL) was added, followed
by 3-phenylpropylamine (217 µL, ca. 3 equiv based on original
resin), and the reaction was shaken for 18 h at 50 °C. The
mixture was filtered, washed with DMSO (10 mL), and
evaporated to give 86 mg (39%) of crude product, 73% pure by
HPLC. This was purified by flash chromatography (gradient
of 2-10% EtOAc/isohexane) to give 26 mg of product (15%) as
a brown oil. MS: 344 (M + H)+. HRMS: calcd 344.1684, found
344.1696 (+3.4 ppm). 1H NMR (CDCl3): δ 1.5 (d, 3H), 1.85
(m, 2H), 2.32 (s, 3H), 2.4 (s, 3H), 2.55 (t, 2H), 3.3 (m, 2H), 4.65
(m, 1H), 6.4 (br s, 1H), 6.75 (m, 2H), 7.1-7.3 (m, 6H) ppm.
2-[1-(4-Ben zyl)p ip er a zin yl]p r op ion ic Acid (3-P h en yl-
p r op yl)a m id e 15i (R1 ) CHMe, X ) 1-(4-Ben zyl)p ip er a -
zin e, R3 ) P h (CH2)3, R4 ) H). Under argon, DIPEA (144 uL,
0.825 mmol) was added to a mixture of resin 13 (R ) CHMe)
(500 mg, 0.55 mmol), KI (137 mg, 0.825 mmol), and 1-ben-
zylpiperazine (144 µL, 0.825 mmol) in dry NMP (5 mL). The
reaction mixture was shaken for 18 h at rt, washed with NMP
(3 × 10 mL), 50% aqueous THF (3 × 10 mL), THF (3 × 10
mL), and MeOH (3 × 10 mL), and dried to give 520 mg of resin.
Anal. Calcd for 1.0 mmol/g: N, 4.20; Br absent. Found: N,
4.51; Br, <0.3.
1.35 mmol/g (theory 1.35). FT-IR: OH absent, CdO 1804 cm-1
.
13C NMR (THF-d8): δ 20.8, 29.1, 31.4, 36.4, 40.7, 70.6, 80.6,
113.9, 115.4, 122.9, 130.0, 134.4, 138.0, 138.8, 148.3, 158.4,
168.6, 171.2 ppm.
(b) Via th e Acid . 2-Bromopropionic acid (25.56 mL, 43.45
g, 0.284 mol) was stirred in CH2Cl2 (400 mL) under an argon
atmosphere. Diisopropylcarbodiimide (3.88 mL, 35.9 g, 0.284
mol) was added slowly, keeping the temperature below 10 °C,
and the mixture was then shaken 2 h. The resin 10 (40 g, 56.8
mmol) was added to the mixture, followed by 4-(dimethylami-
no)pyridine (DMAP, 1.73 g, 0.14 mmol), and the reaction was
shaken for 24 h at rt. The resin was filtered and then washed
with DMF (3 × 300 mL), THF (3 × 300 mL), 50% aqueous
THF (3 × 300 mL), THF (3 × 300 mL), MeOH (3 × 300 mL),
and Et2O (3 × 300 mL) before drying under vacuum to give a
light brown free-flowing resin, wt 47.67 g. Microanalysis gave
N, 1.59%, Br, 8.89%, from which the loading was 1.12 mmol/g
(theory 1.19). FT-IR, 13C NMR both as (a ).
This resin (300 mg, ca. 0.30 mmol) was shaken in 10% TFA/
CH2Cl2 (5 mL) and TIPS (150 µL) for 30 min. The resin was
filtered and then washed with CH2Cl2 (3 × 10 mL), 1 M Et3N
in CH2Cl2 (3 × 10 mL), and THF (3 × 10 mL). THF (5 mL)
was added to the resin, followed by 3-phenylpropylamine (150
µL, ca. 2 equiv), and the reaction mixture was shaken at 50
°C for 18 h. The resin was filtered and then washed with THF
(10 mL), and the solution was evaporated to give 165 mg of
crude product, 52% pure by HPLC. Purification by flash
chromatography (20% EtOAc/isohexane) gave 80 mg of the
desired product (69% yield) as a colorless oil. MS: 366 (M +
Bu P h e Resin , Ester w ith 2-(2-Ch lor op h en ylth io)p r o-
p ion ic Acid 13 (R1 ) CHMe, R2XH ) 2-Cl-C6H4SH). The
resin 12 (R1 ) CHMe, 1.0 g, 1.24 mmol) was added to a
prestirred mixture of 2-chlorothiophenol (2.89 g, 2.0 mmol),
diisopropylethylamine (320 µL, 240 mg, 1.86 mol), and KI (206
mg, 1.86 mmol) in NMP (10 mL) under an argon atmosphere
and the reaction shaken for 24 h at rt. The resin was filtered,
washed with NMP (3 × 15 mL), THF (3 × 15 mL), 50%
aqueous THF (3 × 15 mL), THF (3 × 15 mL), MeOH (3 × 15
mL), and Et2O (3 × 15 mL) before drying under vacuum to
give a light yellow-brown free-flowing resin, wt 1.20 g. Mi-
croanalysis gave N, 1.6%, Br, 0%, S, 3.2%, Cl, 3.7% from which
the mean loading was 1.06 mmol/g (theory 1.07). FT-IR (CH2-
1
H+). HRMS: calcd 366.2525, found 366.2553 (+2.1 ppm). H
NMR (CDCl3): δ 1.2 (d, 3H), 1.85 (m, 2H), 2.5-2.7 (m, 10H),
3.0 (q, 1H), 3.3 (m, 2H), 3.5 (s, 2H), 7.1-7.4 (m, 11H) ppm.
Com p ou n d s 15a ,c-f,j (Ta ble 2). These compounds were
all made by the method described for 15g, from resins prepared
as described for 13, in order to define the general applicability
Cl2 gel): CdO 1737 cm-1 13C NMR (CDCl3): δ 15.8, 29.1, 41.5,
.

“Safety Catch” Ester Linker for Solid-Phase Synthesis
J . Org. Chem., Vol. 66, No. 7, 2001 2245
of the reaction. 15d and 15f were essentially pure after the
reaction and were only characterized by MS; 15a , 15c, 15d ,
and 15j were purified and characterized by NMR and HRMS.
15a . Mp: 59-60 °C. HRMS: calcd 362.1345, found 362.1316
(-8.1 ppm). 1H NMR (CDCl3): δ 1.6-1.9 (m, 6H), 2.15 (m, 2H),
2.6 (m, 2H), 2.95 (m, 2H), 3.3 (m, 2H), 5.4 (br, 1H), 7.05-7.4
(m, 9H).
mol). The mixture was stirred for 18 h at 55 °C under argon,
cooled, and filtered. The product was washed with 50%
aqueous DMF (4 × 200 mL), DMF (2 × 200 mL), THF (4 ×
400 mL), and Et2O (4 × 200 mL) before drying under vacuum
to give 111.5 g of resin. Anal.: Cl, <0.3%. 13C NMR (CDCl3):
δ 18.9, 25.2, 30.4, 61.9, 70.9, 97.2, 115.4, 117.6, 151.1, 153.8
ppm.
15c. Mp: 98-100 °C. HRMS: calcd 396.1189, found 396.1188
The above resin (109 g) was stirred for 24 h in a mixture of
THF (750 mL), MeOH (330 mL), and p-toluenesulfonic acid
(2 g) and then filtered. The product was washed with THF (3
× 500 mL), MeOH (3 × 500 mL), and Et2O (3 × 300 mL) before
drying under vacuum. Yield: 76 g. 13C NMR (THF-d8): δ 70.1,
115.3, 151.5, 152.0 ppm.
Hyd r oxyp h en ylth iom eth yl P olystyr en e Resin . Chlo-
romethyl polystyrene (Polymer Labs, 3.6 mmol/g, 150-300 µm,
1% DVB: 395 g) was stirred with DMF (5 L) and added to a
solution of 4-mercaptothiophenol (474 g, 3.76 mol) in DMF (2.2
L). Et3N (430 mL, 3.1 mol) was added (exotherm to 35 °C) and
the mixture stirred for 72 h at rt. The product was filtered,
washed with DMF (4 × 2 L), THF (4 × 1.5 L), 50% aqueous
THF (4 × 1.5 L), THF (4 × 2.5 L), and MeOH (4 × 2.5 L), and
dried under vacuum. Yield: 520 g. Anal.: S, 9.1; Cl, 0;
corresponds to 2.56 mm/g. 13C NMR (THF-d8): δ 40.0, 115.5,
157.3 ppm.
Note: A similar resin is now available from Novabiochem.
Hyd r oxyp h en ylsu lfon ylm eth yl P olystyr en e Resin . The
above resin (200 g) was stirred in CH2Cl2 (2.5 L) for 15 min at
15 °C, and then peracetic acid (39%, 437 mL) was added
dropwise, keeping the temperature below 25 °C. After the
addition the reaction was filtered and washed with AcOH (3
× 2.5 L), CH2Cl2/AcOH (4:1, 3 × 1.5 L), CH2Cl2 (3 × 1.5 L),
THF (5 × 1.5 L), and t-BuOMe (3 × 1.5 L) before drying under
vacuum. Yield 197 g. Anal.: S, 7.4%; corresponds to 2.08 mm/
g. 13C NMR (THF-d8): δ 40.0, 62.0, 114.7, 161.5 ppm.
1
(-0.2 ppm). H NMR (CDCl3): δ 1.95 (m, 2H), 2.75 (m, 2H),
3.5 (m, 2H), 4.15 (s, 2H), 6.05 (br, 1H), 7.1-7.6 (m, 13H).
15e. Mp: 71-72 °C. HRMS: calcd 348.1189, found 348.1198
(+2.6 ppm). 1H NMR (CDCl3): δ 1.1 (m, 3H), 1.75 (m, 2H), 1.9
(m, 1H), 2.05 (m, 1H), 2.5 (t, 2H), 3.2 (m, 2H), 3.75 (m, 1H),
6.7 (br, 1H), 7.05-7.4 (m, 9H).
15j. Waxy solid. HRMS: calcd 396.1189, found 396.1198
(+2.3 ppm). 1H NMR (CDCl3): d 1.75 (m, 2H), 2.5 (m, 2H), 3.3
(m, 2H), 5.05 (s, 1H), 6.9 (br s, 1H), 7.0-7.5 (m, 14H).
Kin etic Exp er im en ts. Esters of the various resins with
4-methoxyphenylacetic acid were prepared by the following
generic method:
4-Methoxyphenylacetic acid (2.66 g, 16.0 mmol) in DMF (15
mL) and CH2Cl2 (40 mL) was treated, over 10 min at 0-5 °C,
with a solution of diisopropylcarbodiimide (1.21 mL, 2.66 g,
8.0 mmol) in CH2Cl2 (5 mL). After 30 min, 4-dimethylami-
nopyridine (49 mg, 0.40 mmol) and N-methylmorpholine (440
µL, 404 mg, 4.0 mmol) were added, and the solution wasadded
to the resin (4.0 mmol), pre-swelled in CH2Cl2. The mixture
was shaken for 16 h and then filtered and washed with DMF
(2 × 50 mL), MeOH (2 × 40 mL), and Et2O (2 × 40 mL), before
drying under vacuum.
Kin etic An a lysis. The ester resin (40 µmol) was allowed
to swell in THF (1 mL), and piperidine (175 µL, 2 mmol, 50
equiv) was added. The mixture was shaken under an argon
atmosphere at rt, and aliquots of 20 µL were taken at intervals.
These were added to an internal standard (140 µL of a 0.55%
solution of 4-benzyloxybenzyl alcohol in MeOH), and an HPLC
analysis was performed using MeCN (+0.1% TFA) and water
(+0.1% TFA) in a gradient from 25 to 62% organic phase over
5 min (C18 column: Spherisorb ODS2) and detection at 275
nm. The degree of reaction was calculated and half-life
estimated using the Enzfitter program13 (or this may be done
manually), assuming “pseudo-first-order” kinetics due to the
large excess of piperidine employed.
Ack n ow led gm en t. We thank H. Beeley for expert
technical assistance and interpretation of gel-phase
“magic angle” NMR spectra and J . Hammond for sup-
plying the accurate mass data.
Su p p or tin g In for m a tion Ava ila ble: 13C NMR spectra for
compounds 8-10, 12, 13, 15i, and the hydroxy resins; 1H NMR
spectra for compounds 1-3, 5-7, and 15a ,c,e,g-j. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Hyd r oxyp h en oxym eth yl P olystyr en e Resin . Chlorom-
ethyl polystyrene (Polymer Labs, 3.6 mmol/g, 150-300 µm,
1% DVB: 70 g) was stirred in dry DMF (350 mL) and treated
with quinol monotetrahydropyranyl ether (108.5 g, 0.56 mol)
followed by careful addition (5 min) of NaOMe (30.1 g, 0.56
J O0012086