A dimethyl ketal-protected benzoin-based linker suitable for photolytic release of unprotected peptides

Article abstract of DOI:10.1021/jo2017263

Photolabile 3′,5′-dimethoxybenzoin-based linkers are advantageous for a variety of solid-phase synthetic procedures and manipulations of biomolecules because UV irradiation in aqueous media provides fast and essentially quantitative release of tethered molecules, while generating unreactive side products. Practical applications of previously reported linkers are compromised to some extent by the 1,3-dithiane protection of the benzoin carbonyl group and the lengthy synthesis. We have extended the group of photocleavable 3′,5′-dimethoxybenzoin-based linkers by designing and synthesizing a linker in which the carbonyl group is protected as a dimethyl ketal. This protection is compatible with commonly used esterification and amide bond formation techniques, including the Fmoc/tBu strategy for solid phase peptide synthesis, is stable under mild acidic conditions, and can be quantitatively removed in <5 min by 3% TFA in dichloromethane. Irradiation of beads carrying peptides attached to the linker at 350 nm in aqueous or partially aqueous media affords >90% release after 30 min. The linker was synthesized from commercially available starting materials in five steps with an overall yield of 40% and without any column chromatography purification. Additionally, we developed a route to a dithiane-protected linker that requires only two steps and proceeds in 65% yield, a significant improvement over previous synthetic routes.

Full text of DOI:10.1021/jo2017263

Article
pubs.acs.org/joc
A Dimethyl Ketal-Protected Benzoin-Based Linker Suitable for
Photolytic Release of Unprotected Peptides
Nataliya Chumachenko,† Yehor Novikov,‡ Richard K. Shoemaker,‡ and Shelley D. Copley*,†
†Department of Molecular, Cellular and Developmental Biology and Cooperative Institute for Research in Environmental Sciences,
University of Colorado at Boulder, Boulder, Colorado 80309, United States
‡Department of Chemistry and Biochemistry, University of Colorado at Boulder, Boulder, Colorado 80309, United States
S
* Supporting Information
ABSTRACT: Photolabile 3′,5′-dimethoxybenzoin-based link-
ers are advantageous for a variety of solid-phase synthetic
procedures and manipulations of biomolecules because UV
irradiation in aqueous media provides fast and essentially
quantitative release of tethered molecules, while generating
unreactive side products. Practical applications of previously reported linkers are compromised to some extent by the 1,3-dithiane
protection of the benzoin carbonyl group and the lengthy synthesis. We have extended the group of photocleavable 3′,5′-
dimethoxybenzoin-based linkers by designing and synthesizing a linker in which the carbonyl group is protected as a dimethyl
ketal. This protection is compatible with commonly used esterification and amide bond formation techniques, including the
Fmoc/tBu strategy for solid phase peptide synthesis, is stable under mild acidic conditions, and can be quantitatively removed in
<5 min by 3% TFA in dichloromethane. Irradiation of beads carrying peptides attached to the linker at 350 nm in aqueous or
partially aqueous media affords >90% release after 30 min. The linker was synthesized from commercially available starting
materials in five steps with an overall yield of 40% and without any column chromatography purification. Additionally, we
developed a route to a dithiane-protected linker that requires only two steps and proceeds in 65% yield, a significant
improvement over previous synthetic routes.
Scheme 1. Photocleavage of Phenacyl Esters (a) and 2-
Nitrobenzyl Esters and Amides (b)
INTRODUCTION
■
Photolabile linkers are being used in an increasing number of
chemical and biological applications. Solid-phase organic
synthesis allows synthesis or transformation of substrates
while bound to a solid phase through a linker that can later
be degraded, allowing release of the product. If release is
achieved photochemically, the product does not require
purification from reagents necessary for chemical cleavage.
Biological applications of photolabile linkers have increased
dramatically within the past few years. Examples include
enrichment of specific proteins and metabolites,1 photo-
activatable probes for protein labeling,2 “caged” compounds
for light-induced release of biological activity,3 delivery of
activators/inhibitors to organelle surfaces,4 and construction of
polymeric vesicles for imaging and drug delivery.5
Our applications required synthesis and then efficient
release of short fully deprotected peptides and peptide libraries
from a solid support into neutral aqueous medium. The most
widely used photolabile linkers, which contain either a
phenacyl (2-oxo-2-phenylethyl) (Scheme 1a) or a 2-nitrobenzyl
(Scheme 1b) moiety, are not suitable for this purpose.
Phenacyl derivatives are not stable under the nucleophilic
conditions6 of the Fmoc/tBu strategy and are prone to
formation of piperazine-2,5-diones.7 Photolysis requires a
hydrogen donor, typically ethanol, which is converted to
acetaldehyde that can cause side reactions. Furthermore, the
efficiency of photocleavage is lower in the presence of water.8
We attempted to use two commercially available 2-nitrobenzyl
linkers3-amino-3-(2-nitrophenyl)propionic acid9 and 4-
(4-(1-aminoethyl)-2-methoxy-5-nitrophenoxy) butanoic acid
(the Holmes linker)10,11but observed <40% photorelease
for substrates containing free amino groups.
We turned our attention to the dithiane-protected benzoin-
based photolinkers 112 and 213,14 (Scheme 2). Unfortunately,
there is no reliable method for dithiane removal that is
compatible with the entire set of amino acid functional groups.
Rock and Chan used [bis(trifluoroacetoxy)iodo]benzene
(BTI)15 to remove the dithiane prior to release of a short
peptide,12 but this reagent promotes Hofmann rearrangement
of primary amides16 and oxidizes phenols,17 thiols,18 and
secondary amines.19 Further, 1 has only one meta-alkoxy
substituent on the nonconjugated phenyl ring; usually two such
substituents enable faster and cleaner photocleavage.20,21
Received: August 19, 2011
Published: September 27, 2011
© 2011 American Chemical Society
9409
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416

The Journal of Organic Chemistry
Article
Scheme 2. Previously Reported Dithiane-Protected Benzoin Linkers
Synthesis of 2 requires 3-hydroxy-5-methoxybenzaldehyde,
which is not widely commercially available.22
Here we report a new 3′,5′-dimethoxybenzoin-based linker 3
(Scheme 3) in which the carbonyl group is protected as a
immobilized peptide are used,26 the yields of compounds
containing free amino groups (glycinamide or 6-amino-
hexanamide) after photocleavage from these linkers were
below 40% in our hands, even on a 25 mg scale. The low
yields likely resulted from a combination of factors, including
formation of internal light filters and reactions of the free amino
groups on the substrate with nitrosoketone intermediates
formed during photocleavage. Amines, such as hydrazine,
hydroxylamine, or semicarbazide,24 or thiols, such as DTT,27
can be added during irradiation to scavenge nitrosoketone
intermediates. However, addition of scavengers negates
advantages of photocleavage for many biological applications.
Because of the difficulties we encountered with photo-
cleavage of 2-nitrobenzyl linkers, we turned our attention to
benzoin-based linkers. Facile release of esterified carboxylic
acids by photocyclization of benzoin esters was reported in
1964 by Sheehan and Wilson.28 They20 and others21 found
that, while electron-donating substituents on the benzoyl ring
interfered with photodeprotection, meta-alkoxy substitution on
the nonconjugated phenyl ring allows rapid and essentially
quantitative photorelease of the carboxylic acid and produces
relatively inert 2-phenyl-5,7-dimethoxybenzofuran (Scheme 4).
Scheme 3. Dimethyl Ketal-Protected Benzoin-Based
Photolinker 3
dimethyl ketal. The protection is stable under mild acidic
conditions. After simultaneous deprotection of the benzoin
carbonyl and amino acid side chains, unprotected peptides can
be rapidly released by photolysis. The linker can be synthesized
in five steps with an overall yield of 40%. All of the intermediates
are crystalline solids, and no intermediate requires purification by
column chromatography.
RESULTS AND DISCUSSION
■
Scheme 4. Products Formed by Photocleavage of 3,5-
Dimethoxybenzoin Esters
We began our work by exploring the utility of commercially
available N-Fmoc-protected 2-nitrobenzyl linkers3-amino-3-
(2-nitrophenyl)propionic acid9 and 4-(4-(1-aminoethyl)-2-
methoxy-5-nitrophenoxy) butanoic acid10,11 (the Holmes
linker)for photolytic release of unprotected peptides. The
linkers were quantitatively coupled to TentaGel S NH2 resin
using the standard PyAOP/HOAt/DIEA method. Fmoc
protection was removed by treatment with 20% piperidine in
DMF for 20 min at room temperature. The simple model
compounds N-Fmoc-glycine or N-Fmoc-6-aminohexanoic acid
were quantitatively attached to the amino groups of the linkers
and deprotected using the same protocols. The structures of
the products were confirmed on-resin using high-resolution
magic angle spinning (HRMAS) NMR spectroscopy. Sub-
sequently glycinamide or 6-aminohexanamide was released
from the resin (25 mg, containing 0.2−0.3 mmol of substrate
per g) by irradiation at 350 nm with stirring under argon. For
both linkers, the yield of released amide after irradiation in
water, methanol, or acetonitrile was <40%. After 20 min of
irradiation, the resins turned a deep brownish-red color, and the
reactions did not proceed further. We are not the first group to
encounter complications with the use of the Holmes linker for
release of compounds containing amino groups. Detrimental
side reactions that occur during photochemical cleavage of
unprotected peptides have been reported.23 Additionally, it has
long been known that deep red azo and azoxy compounds
arising from the nitroso photoproducts24,25 of linker decom-
position act as internal light filters, thus hindering complete
photorelease of substrates from o-nitrobenzyl moieties. While it
was reported that photocleavage of protected peptide fragments
from an o-nitrobenzyl linker proceeds in high yield when
degassed solvents and <300 mg of the resin containing
Efficient photolytic formation of the benzofuran was also
observed for 3,5-dimethoxybenzoin (DMB) derivatives bearing
other relatively good leaving groups, such as phosphate,29
phosphoester,21 carbonate,30 or carbamate.31 Photolysis can be
carried out in a variety of solvents (benzene, THF, dioxan,
acetonitrile, ethanol, water). Only two studies have reported
products other than inert benzofuran. Photolysis of water-
soluble 3′,5′-bis(carboxymethoxy)benzoin (BCMB) acetate in
aqueous media produced around 70% of BCMB,32 and
photolysis of DMB fluoride produced 20% of parent 3′5′-
dimethoxybenzoin on irradiation in acetonitrile−water (1:1 by
volume). In trifluoroethanol, DMB fluoride produced 35% of a
product in which the fluoride had been substituted by
trifluoroethoxide.33 Although the reaction mechanism of the
photodecomposition of benzoin derivatives is still debated,33,34
for our purposes it was germane that photolysis of 3′5′-
dimethoxybenzoin esters is rapid, results in nearly quantitative
release of carboxylic acids in aqueous media, and produces
nonreactive byproducts.
The stability of benzoin-based linkers is enhanced by
protection of the benzoin carbonyl to provide a “safety catch”
that prevents premature cleavage. Our goal was to design a
linker that possesses a carbonyl protecting group compatibile
9410
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416

The Journal of Organic Chemistry
Article
a
Scheme 5. Synthesis, Attachment to Beads, and Deprotection of Photolabile Linker 3
a
Key: (a) 1,3-propanedithiol, BF3*Et2O, AcOH, 75 °C, 1.5 h, 93%; (b) BuLi, −60 °C, 1 h 50 min, and then 3,5-dimethoxybenzaldehyde, −78 °C,
40 min, 70%; (c) TBSOTf, 2,6-lutidine, CH2Cl2, rt, 3 h, 88%; (d) 80% AcOH, 50 to 75 °C over 45 min, then 75 °C, 15 min, 86%; (e) BTI, MeOH,
rt, 30 min, 81%; (f) PyAOP, HOAt, DIEA, DMF, rt, 45 min; (g) 0.17 M TBAF in THF.
with the common Fmoc/tBu strategy of SPPS but that can be
removed without detrimental effects on common peptide
functional groups. Additionally, we aimed to design a less
laborious synthesis from commercially available starting
materials.
the dithiane ring. At 0 °C, BuLi slowly attacks the lithium
carboxylate, so it is important to cool the reaction to −60 °C,
use no more than 2.05 equiv of BuLi, and add the BuLi slowly
over at least 15 min while stirring efficiently. It is also important
to use no more than 1.06 equiv of 3,5-dimethoxybenzaldehyde
in the reaction to avoid formation of 3,5-dimethoxybenzoic
acid (which is extremely difficult to separate from 6),
probably via a base-induced disproportionation−Cannizzaro
reaction. Note that this strategy results in synthesis of the
dithiane-protected linker 6 in two steps in 65% overall
yield. This linker can be used in place of linkers 1 and 2 for
analogous applications, but its preparation is markedly more
efficient.12−14
Dithiane 6 could have been converted directly to the
corresponding dimethyl ketal. However, leaving the OH group
unprotected would have prevented applications requiring an
ester bond-based attachment of the linker to a solid support (or
some other moiety). Under esterification conditions, the free
hydroxyl of the linker will compete with the hydroxyl group of
the solid support, leading to polymerization of the linker. This
concern was based upon the report that dithiane-protected 3-
hydroxybenzoin immobilized on a resin was readily esterified
with Fmoc derivatives of Phe, Val, and Pro using the
diisopropylcarbodiimide protocol in 5 h (with yields of 100%,
65%, and 57%, respectively).13 Furthermore, the presence of a
free OH group could complicate loading of the linker onto
amino group-bearing moieties via common methods for
formation of amide bond. For example, sterically hindered
carboxylic acids were attached to the free hydroxyl group of a
resin-immobilized dithiane-protected 3-hydroxybenzoin in
significant (<30%) yields13 under standard HATU or PyBOP
activation-based protocols. Moreover, the PyAOP activation
procedure was used for acylation of hydroxymethyl resins with
Fmoc/Trt derivatives of Cys and His.36 To avoid any problems
with polymerization of the linker during loading onto a solid
phase under a broad range of conditions, we chose to protect
the free hydroxyl group before converting the dithiane to a
dimethyl ketal.
After synthesis of peptides on a solid phase using the Fmoc/
tBu strategy, deprotection of amino acid side chains is usually
achieved by treatment with concentrated TFA. We explored the
utility of protecting the benzoin carbonyl as an acid-labile ketal
to allow simultaneous deprotection of the benzoin carbonyl and
amino acid side chains prior to photocleavage of the peptide
from the support. The stability of ketals to aqueous and
nonaqueous bases and to nucleophiles allows for compatibility
with the Fmoc/tBu SPPS strategy. However, we had two major
concerns. First, the reactivity of carbonyl groups toward
ketalization is low for aromatic ketones. Indeed, our preliminary
attempts to convert the commercially available 2-hydroxy-2-
phenylacetophenone to the dimethyl ketal by the common
procedure ((MeO)3CH, anhydrous MeOH, TsOH, reflux)35
were unsuccessful. Second, the ketal protection might cleave
prematurely under mildly acidic (pH 4) conditions that might
be required during both synthesis of the linker and downstream
applications. Therefore, we chose to construct the benzoin
framework using a 1,3-dithiane-masked carbonyl and then to
transform the S,S-acetal directly into the dimethyl ketal by the
ingenious Stork and Zhao protocol.15 Here we describe a new
benzoin-based linker 3, which can be efficiently synthesized in
five steps with an overall yield of 40% (Scheme 5). The linker
possesses a ketal protecting group that remains intact on
heating of 3 to 95 °C in 80% aqueous acetic acid but can be
quantitatively removed in 3 min at room temperature by 3%
TFA in dichloromethane.
We chose to use a carboxylic acid as an anchoring group to
hydroxyl- or amine-functionalized resins and began the
synthesis from commercially available 3-formylbenzoic acid
(4). The transformation of 4 to the corresponding dithiane 5 is
straightforward and nearly quantitative. Conversion of 5 to 6
requires some care. The reaction begins with addition of 1
equiv of BuLi to generate the lithium salt of 5; a second
equivalent of BuLi is added to remove the acidic proton from
We could not find conditions for transformation of 6
into 7 using inexpensive tert-butyldimethylsilyl (TBS) chloride.
9411
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416

The Journal of Organic Chemistry
Article
a
Table 1. Yields of Dipeptides after Photorelease from Beads
yield (%) in the indicated solvent
CD3CN:D2O,
CD3OD:D2O,
6:4
dipeptide
irradn time, min
CD3CN
6:4
CD3OD
His-Ile
His-Ile
His-Ile
Val-Tyr
15
30
45
45
37
41
44
40
78
83
93
90
62
73
75
70
78
92
98
95
a
Yields were determined by 1H NMR in the presence of an added internal standard (p-toluic acid or sodium acrylate). Calculations are based on the
capacity of amino groups on the TentaGel (0.30 mmol/g).
However, reaction with TBSOTf (easily prepared from TBS
chloride and triflic acid)37 was clean and gave no side products.
The silyl ester 7 was readily purified by crystallization.
Complete hydrolysis of 7 to the corresponding acid 8 required
heating to 75 °C in 80% aqueous acetic acid for 1 h. As
expected, the S,S-acetal in 8 was smoothly converted to the
corresponding dimethyl ketal 3 in high yield (81%) by
treatment with BTI in anhydrous methanol by the method of
Stork and Zhao.15 Luckily, the ketal protection was entirely
stable to hydrolysis on prolonged heating of 3 to 95 °C in 80%
aqueous acetic acid.
amino acid side chains intact and found that removal of the
ketal is quantitative in 5 min at room temperature using 3% v/v
TFA in anhydrous dichloromethane. A higher concentration of
TFA (3%) than expected (1%) was necessary, most likely due
to buffering of the solution by the large number of ether
oxygens in the poly(ethylene glycol) chains grafted on the
polystyrene matrix of TentaGel.41
Photocleavage of peptides (either protected or unprotected)
from the resin was achieved by irradiation with a 100 W “street
light type” medium-pressure mercury lamp (GE H100-A4/T),
as described in the Experimental Section. Table 1 shows that
photorelease was complete in 45 min, with ∼80% of the
product being produced during the first 15 min of irradiation.
The efficiency of photocleavage of dipeptides His-Ile and
Val-Tyr depends on the nature of the solvent; yields were
nearly quantitative in the presence of water in acetonitrile or
methanol but considerably lower in anhydrous acetonitrile or
methanol.
Quantitative loading of 3 onto TentaGel NH2 beads to give 9
was achieved using the standard PyAOP/HOAt/DIEA method
in anhydrous DMF (2× excess of 3). Because the Kaiser test38
demonstrated the absence of unreacted amino groups,
treatment of the solid phase with acetic anhydride to block
residual amino groups was not required. Removal of the TBS
protecting group with TBAF in THF to give 10 occurred in
93% yield after 3 h at room temperature and was quantitative
1H NMR analysis revealed that each dipeptide was obtained
as a mixture of two diastereomers. When DCC/DMAP
coupling was used, the C-terminal amino acid underwent 6−
8% racemization when the first amino acid was Tyr (2 h
reaction) and up to 15% racemization when the first amino acid
was Ile (6 h reaction). Racemization of the C-terminal amino
acid due to use of DMAP as the reaction catalyst is a common
problem in SPPS.42 Less racemization is observed when
acylation of hydroxymethyl-based resins with Fmoc-protected
amino acids is carried out using TDO (4-hydroxy-3-oxo-2,5-
diphenyl-2,3-dihydrothiophene 1,1-dioxide) esters,43 mixed
anhydrides formed from the amino acid and 2,6-dichloroben-
zoyl chloride in pyridine,44 or acyl fluorides in a mixture of
pyridine and CH2Cl2.45 In our case, the TDO ester obtained
from Fmoc-Tyr(tBu)-OH was not reactive with 10, and use of
the corresponding acid fluoride resulted in only 20%
esterification in 24 h. Reaction of 10 with a 5-fold molar
excess of both Fmoc-Tyr(tBu)-OH and 2,6-dichlorobenzoyl
chloride and a 20-fold molar excess of pyridine occurred in
quantitative yield in 24 h but, unfortunately, resulted in 14%
racemization.
The partial loss of stereochemical integrity during the
esterification step, which is likely due to steric crowding around
hydroxyl reaction centers in benzoin-based photolinkers, has
not been previously recognized, although it would be expected
to occur with linkers 1 and 2. Previous workers did not report
this problem, because they either used Fmoc-Gly as the C-
terminal amino acid for peptide synthesis12 or limited their
trials to attachment following by release of a single amino acid
and therefore could not have detected racemization using the
techniques that were employed for characterization of the
product.13,14 Thus, this is an issue that likely needs to be
resolved for the entire class of benzoin-based linkers.
1
by 6−7 h. The reaction was monitored by HRMAS H NMR.
The TBS signals at δ 0.92, 0.18, and −0.18 ppm disappeared.
Concomitantly, the resonance of the proton at the carbon
carrying the OTBS group in 9 and OH in 10 shifted from
δ 4.99 to 5.08 ppm.
Esterification of ketal 10 proved to be more facile than
esterification of more sterically hindered 1,3-dithiane protected
benzoins such as 1 and 2 as well as others.12−14 In the case of
10, even the most sterically hindered amino acid, isoleucine,
was quantitatively39 added to the beads after 6 h at room
temperature using the carbodiimide protocol (DCC/DMAP),
the most common approach in the SPPS literature for the
esterification of solid-phase alcohol groups.40 For less-crowded
molecules such as Fmoc-Tyr(OtBu), Fmoc-6-aminohexanoic
acid, and propionic acid, esterification was complete in 2 h. In
contrast, only 65% loading of the sterically crowded Fmoc-Val
onto the solid-supported dithiane linker 2 was achieved after
extended reaction.13 After removal of the Fmoc protecting
group (20% piperidine in DMF, 15 min, rt), a second Fmoc-
protected amino acid was added (His to the Ile resin and Val to
the Tyr resin) using the PyAOP/HOAt/DIEA procedure.
As a next step we investigated whether the ester linkage
between 10 and a substrate is stable under conditions generally
employed for removal of protecting groups from amino acid
side chains. After treatment with a freshly prepared mixture of
TFA/H2O/phenol/thioanisole (85 mL/5 mL/5 g/5 g) for 6 h
at room temperature, we did not observe any loss of His-Ile or
Val-Tyr from the beads due to hydrolysis of the ester bond. As
expected, the dimethyl ketal on the linker was completely
removed under these conditions.
We also explored conditions that would allow deprotection
of the ketal of the linker while leaving protecting groups on the
9412
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416

The Journal of Organic Chemistry
Article
Nevertheless, our linker provides a significant improvement due
to the ease of the esterification of dimethyl ketal protected 10
in comparison with the dithiane-bearing 1 and 2. The observed
racemization will not be a problem for many molecular biology
applications because a spacer is often positioned between the
photolabile linker and an immobilized peptide to prevent steric
hindrance of binding of macromolecules to the resin.
crude peptide we obtained was higher than expected, even
though the standard SPPS procedure results in significant
contamination by salts and residual protecting groups. The H
NMR spectrum showed no visible traces of small molecules or
residual protecting groups (see Supporting Information). The
predicted monoisotopic mass for the peptide is 1837.1208; the
observed mass of the major peak is 1837.0793.
1
At present, the choice of photolable linkers for the most
efficient Fmoc/tBu strategy in SPPS is limited to o-nitrobenzyl-
based linkers. As discussed above, these linkers are restricted to
only relatively small amounts of peptide resin and might result
in low cleavage yields, and consequently, they are rarely used
for SPPS. Rather, acid-labile linkers are preferred. Deprotection
of amino acid side chains and release of the final peptide from
the solid support are accomplished simultaneously by acid
hydrolysis with concentrated TFA containing large (up to 5%)
quantities of phenol, thioanisole, and other scavengers of tBu
carbonium ions. The crude peptide is precipitated in diethyl or
methyl tert-butyl ether and contains over 50 wt % of salts, small
water-soluble molecules, protecting group residues, deletion
peptides, and cleavage-deteriorated peptides. Purification of the
peptide to >95% is complicated by the presence of this complex
mixture of salts and byproducts. A photocleavable linker should
allow deprotection of side chains on the resin and then
extensive washing of the resin-bound peptide before the final
cleavage, delivering the product contaminated only with the
inevitable deletion/truncation peptides and peptides damaged
by deprotection.
CONCLUSION
■
We have succeeded in synthesizing a photocleavable benzoin
linker 3 that has significant advantages over currently available
benzoin-based linkers due to the dimethyl ketal protection of
the benzoin carbonyl and the straightforward synthesis in 40%
overall yield from readily available starting materials via five
steps, all of which afford stable crystalline products. Linker 3
shares the previously recognized valuable photochemical
properties of 3′,5-dimethoxybenzoin-based protective groups
that allow rapid and essentially quantitative photorelease while
producing nonreactive and colorless 2-phenyl-5,7-dimethoxy-
benzofuran. Partial loss of stereochemical integrity of acids
possessing an α-stereogenic center occurs on esterification of
our linker and, most likely, other benzoin-based linkers under
common conditions. However, our linker provides considerable
advantages for a multitude of molecular biology applications in
which a spacer between the resin and the immobilized ligand or
peptide is commonly used and therefore racemization is not a
problem. Further, our linker may provide an improvement over
2-nitrobenzyl linkers in photoactivatable probes for in vivo
applications because photocleavage does not produce reactive
intermediates that can damage amino acid side chains.
To explore the usefulness of linker 3, we took on the
challenge of synthesizing a very difficult arginine-rich peptide
sequence (Lys-Tyr-Arg-Arg-Arg-Pro-Arg-Arg-Ser-Gln-Arg-Lys-
Arg-Gly). The synthesis was carried out on a 0.04 mmol scale.
A common problem with ester-based linkers is formation of
piperazine-2,5-diones (diketopiperazines), resulting in release
of the substrate during coupling of the third amino acid and re-
EXPERIMENTAL SECTION
■
General. Melting points were determined on a hot-stage apparatus
and are uncorrected. On-resin NMR experiments were performed on a
400 MHz spectrometer using with a 4 mm ghxNano probe. Magic
angle sample spinning (MAS) was performed at a spinning speed of
2.8−3.0 kHz. Prior to spinning, samples on polystyrene beads were
swelled in CDCl3 to allow sufficient line narrowing for high-resolution
NMR. High-resolution mass spectra were obtained on an ABI QSTAR
in ECI (−) mode. Air-sensitive reactions were performed in flame-
dried glassware under argon. Reagents and solvents were used as
purchased unless otherwise indicated.
1
formation of the bead-linked linker 10. However, HRMAS H
NMR of the beads carrying Lys-Arg-Gly after the third coupling
step shoved no signal at 5.08 ppm characteristic of the CHOH
proton of 10. Therefore, we conclude that formation of
piperazine-2,5-diones is not a concern.
According to the Proteomics Resource Center of the
Rockefeller University, a 0.04 mmol scale synthesis is expected
to produce 39−53 mg of a crude 14-mer peptide, and “HPLC
purification to >95% purity will reduce the yield by at least
50%”. Using our linker, after completion of the synthesis and
irradiation of the beads for 40 min in 5 mL of 30% aqueous
methanol, we obtained 62 mg of desalted peptide that was
>75% pure according to HPLC (Figure 1). Thus, the yield of
3-(1,3-Dithian-2-yl)benzoic Acid (5). 3-Formylbenzoic acid (4,
3.00 g, 20.0 mmol, 1.0 equiv) was dissolved in acetic acid (30 mL) at
75 °C in an oil bath. Neat 1,3-propanedithiol (2.40 mL, 2.58 g, 23.9
mmol, 1.2 equiv) was added, followed by boron trifluoride etherate
(1.00 mL, 1.15 g, 8.1 mmol, 0.41 equiv). The reaction mixture was
stirred in the bath for 1.5 h and then evaporated to dryness at reduced
pressure. The residue was dissolved by refluxing in methanol (30 mL).
The boiling solution was treated with 12 mL of water, which caused
the solution to become cloudy, and allowed to cool to room
temperature. Formation of white crystals of 5 was complete in 1 h at
room temperature. The crystals were filtered, washed with water,
dried, and washed with hexanes to remove residual dithiol. The white
solid 5 (4.70 g, 19.6 mmol, 98%) was pure by H and 13C NMR.
1
Recrystallization from methanol/water (30 mL/12 mL) was
performed to remove any traces of boron derivatives, yielding 5
(4.47 g, 18.6 mmol, 93%) as snow-white crystals; mp = 164−165 °C.
1H NMR (300 MHz, CDCl3): δ 8.23 (s, 1H), 8.06 (d, J = 7.8 Hz, 1H),
7.75 (d, J = 7.7 Hz, 1H), 7.47 (app. t, J = 7.75 Hz, 1H), 5.23 (s, 1H),
3.15−3.00 (m, 2H), 3.00−2.85 (m, 2H), 2.25−2.13 (m, 1H), 2.05−
1.86 (m, 1H). 13C NMR (300 MHz, CDCl3): δ 172.0, 139.9, 133.5,
130.5, 130.0, 129.3, 51.1, 51.0, 32.2, 25.2. Predicted exact mass for
C11H11O2S2 239.0206; observed mass 239.0193.
Figure 1. HPLC trace of crude Lys-Tyr-Arg-Arg-Arg-Pro-Arg-Arg-Ser-
Gln-Arg-Lys-Arg-Gly. The detector was set at 217 nm.
3-[2-(Hydroxy(3,5-dimethoxyphenyl)methyl)-1,3-dithian-2-yl]-
benzoic Acid (6). Preparation of 6 was done under argon with
9413
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416

The Journal of Organic Chemistry
Article
vigorous mechanical stirring. Acid 5 (3.00 g, 12.5 mmol, 1.00 equiv)
was dissolved in anhydrous THF (250 mL). The solution was cooled
to −60 °C, and BuLi (16 mL of 1.6 M solution in hexane, 25.6 mmol,
2.05 equiv) was added dropwise over 20 min. After addition of about
4.5 mL (7.2 mmol, 0.58 equiv) of BuLi, an increasing amount of white
precipitate appeared in the reaction mixture. After addition of about
8.5 mL (13.6 mmol, 1.09 equiv), the liquid became brown and the
precipitate began to disappear. By the end of the addition of BuLi, all
of the precipitate had dissolved. The very dark brown solution was
stirred at −60 °C for 1.5 h and then further cooled to −78 °C. A
solution of 3,5-dimethoxybenzaldehyde (2.20 g, 13.2 mmol, 1.06
equiv) in anhydrous THF (50 mL) was added dropwise over 10 min,
resulting in formation of a whitish precipitate. The reaction mixture
was warmed to room temperature over 30 min, stirred for an
additional 30 min, and treated with 1 M HCl (27 mL). The precipitate
immediately dissolved. The solvent was removed by rotary evaporation
(be careful: foams) to produce a yellow oil. The oil was extracted with
aqueous NaOH (20 mmol in 120 mL, 30 min stirring; then twice with
5 mmol in 30 mL). The combined aqueous extracts were centrifuged
at room temperature for 10 min. The pH of the aqueous extract was
adjusted to 2 with HCl, resulting in the separation of acid 6 as an oil.
The aqueous solution was treated with NaCl until saturation to ensure
the complete recovery of 6. The oil was extracted with diethyl ether (3
× 100 mL), and the combined extracts were dried over MgSO4,
filtered, and concentrated to produce a light yellow oil of crude 6.
Recrystallization from ethyl acetate gave 6 (3.55 g, 8.73 mmol, 70%) as
complete. The crystals were collected by filtration, washed with
pentane, and dried to give 8 (2.96 g, 5.68 mmol, 86%) as white
crystals; mp =135−137 °C. (Important note: do not crystallize the
product from hexanes, as it crystallizes as a complex with some hexane
1
isomers.) H NMR (300 MHz, CDCl3): δ 8.67 (br s, 1H), 8.04 (app
dt, Jd = 7.7 Hz, Jt = 1.4 Hz, 1H), 7.94 (br d, J = 7.2 Hz, 1H), 7.39 (app
t, J = 7.8 Hz, 1H), 6.30 (t, J = 2.3, 1H), 5.90 (very broad app s, 2H),
5.04 (s, 1H), 3.55 (s, 6H), 2.75−2.40 (m, 4H), 1.98−1.77 (m, 2H),
0.88 (9H), 0.10 (3H), −0.33 (3H). 13C NMR (300 MHz, CDCl3): δ
172.4, 159.3, 140.7, 137.8, 137.7, 134.3, 128.9, 128.8, 127.4, 106.4 (br),
100.7, 81.6, 66.8, 55.0, 27.4, 27.0, 25.7, 24.9, 18.3, −4.6, −5.1.
Predicted exact mass for C26H35O5S2Si 519.1701; observed mass
519.1636.
3-[1,1-Dimethoxy-2-tert-butyldimethylsilyloxy-2-(3,5-
dimethoxyphenyl)ethyl]benzoic Acid (3). Dithiane 8 (1.00 g, 1.92
mmol, 1.00 equiv) was dissolved in a mixture of anhydrous methanol
(20 mL) and dichloromethane (3 mL). A solution of [bis-
(trifluoroacetoxy)iodo]benzene (2.10 g, 4.88 mmol, 2.54 equiv) in
dichloromethane (8 mL) was added dropwise. The reaction mixture
was stirred for 30 min at room temperature and then poured into cold
aqueous 0.17 N NaOH (90 mL). The organic solvents were removed
by rotary evaporation (foams). The remaining aqueous solution
was filtered and treated with CO2 gas until the pH decreased to
6 and then with solid NaCl until saturation. Acid 3 separated as an oil.
To collect the oil, the mixture was extracted with dichloromethane.
The dichloromethane solution was dried over MgSO4, and the
dichloromethane was removed by rotary evaporation to leave a light-
yellow oil. The product was crystallized from a minimal volume
of hexanes (15−20 mL) to give 3 (0.74 g, 1.56 mmol, 81%) as white
crystals; mp = 129−130 °C. (Important note: If <2.5 equiv excess of
[bis(trifluoroacetoxy)iodo]benzene is used or if this reagent is old, the
final product can be contaminated with 5−10% of dithiane 8.
Recrystallization does not enrich the mixture in 3. A route to purify
3 from 8 is by additional treatment of the mixture with
1
white crystals; mp = 184−185 °C. H NMR (300 MHz, CDCl3): δ
8.55 (app t, J = 1.7 Hz, 1H), 8.04 (app dt, Jd = 7.7 Hz, Jt = 1.4 Hz,
1H), 7.92 (ddd, Jd = 8.0 Hz, Jd = 2.1 Hz, Jd = 1.2 Hz, 1H), 7.41 (app t,
J = 7.8 Hz, 1H), 6.31 (t, J = 2.3, 1H), 6.01 (d, J = 2.3, 2H), 4.98 (s,
1H), 3.58 (s, 6H), 2.85−2.51 (m, 4H), 2.07−1.83 (m, 2H). 13C NMR
(300 MHz, CDCl3): δ 172.1, 159.8, 139.2, 138.8, 136.8, 132.9, 129.6,
129.7, 128.5, 106.3, 101.2, 81.1, 65.8, 55.4, 27.5, 27.3, 24.9. Predicted
exact mass for C20H21O5S2 405.0836; observed mass 405.0844.
3-[2-(tert-Butyldimethylsilyloxy(3,5-dimethoxyphenyl)methyl)-
1,3-dithian-2-yl]benzoic Acid tert-Butyldimethylsilyl Ester (7). Acid
6 (2.30 g, 5.66 mmol, 1.00 equiv) was dried in vacuum over CaCl2
overnight, suspended with stirring in dichloromethane (35 mL), and
treated with 2,6-lutidine (3.50 mL, 3.23 g, 30.1 mmol, 5.33 equiv),
which resulted in complete dissolution. The solution was treated
dropwise with neat tert-butyldimethylsilyl triflate (4.40 mL, 5.08 g,
19.2 mmol, 3.39 equiv) and stirred at rt for 3 h until the starting
material had completely disappeared based upon TLC analysis
(toluene:AcOH, 9:1). The reaction mixture was poured into a mixture
of water (75 mL) and ethyl acetate (150 mL). The organic layer
(whose volume was kept at ca. 150 mL by the addition of extra ethyl
acetate when necessary) was washed with water (3 × 75 mL) and
brine (2 × 50 mL) and then dried over MgSO4, filtered, and
concentrated by rotary evaporation to produce light yellow crystals of
crude ester 7. Recrystallization from the minimal amount of hexanes
(rt for at least 5 h and then at 4 °C for 5 h) gave pure 7 (3.15 g, 4.96
1
[bis(trifluoroacetoxy)iodo]benzene as above.) H NMR (300 MHz,
CDCl3): δ 8.06 (br s, 1H overlaps with 8.06−7.98 (m, 1H), 7.29−7.32
(m, 2H), 6.27 (t, J = 2.3), 5.98 (d, J = 2.3 Hz, 2H), 4.97 (s, 1H), 3.56
(s, 6H), 3.42 (s, 3H), 3.25 (s, 3H), 0.89 (s, 9H), 0.10 (s, 3H), −0.20
(s, 3H). 13C NMR (300 MHz, CDCl3): δ 172.3, 159.2, 142.4, 137.7,
134.6, 131.4, 129.4, 128.0, 126.7, 106.3, 104.2, 99.7, 75.7, 55.0, 49.6,
49.4, 25.7, 18.2, −4.5, −5.0. Predicted exact mass for C25H35O7Si
475.2158; observed mass 475.2109.
Compound 9: Linker 3 Loaded onto TentaGel NH2 Beads. Tenta-
Gel NH2 beads (1.00 g, 300 μmol of NH2, 1 equiv) were swollen in
anhydrous DMF (5 mL) for 15 min, then washed three times with 5
mL of anhydrous DMF, and finally suspended in DMF (3 mL). Solid
linker 3 (287 mg, 0.602 μmol, 2 equiv) was added to the beads, and
the mixture was gently shaken for 1 min to allow the linker to dissolve.
Solid PyOAP (312 mg, 598 μmol, 2 equiv) and solid HOAt (81 mg,
595 μmol, 2 equiv) were added, followed by N,N-diisopropylethyl-
amine (210 μL, 156 mg; 1207 μmol, 4 equiv). The reaction mixture
was shaken gently for 40 min. The beads were drained, washed with
anhydrous DMF (5 times × 5 mL), anhydrous MeOH (5 times × 5
mL), and anhydrous dichloromethane (5 times × 5 mL), and dried.
Compound 10. One half of the TentaGel beads carrying the linker
(3) from the step above (562 mg) were swollen in THF (5 mL) for 15
min, washed three times with 5 mL of THF, and drained. The volume
of the swollen drained beads was ca. 2.7 mL. THF (3.1 mL) and TBAF
(1.20 mL, 1 M in THF, 1.20 mmol) were added, resulting in a final
TBAF concentration of ca. 0.17 M. The reaction mixture was gently
shaken for 7 h (93% deprotection occurred in 3 h). The beads were
drained, washed with THF (5 times × 5 mL), water (5 times × 5 mL),
anhydrous MeOH (5 times × 5 mL), and anhydrous dichloromethane
(5 times × 5 mL), and dried. Important note: Complete removal of
water and/or the increase of the final TBAF concentration over 0.2
mM slows or stops the reaction. Do not use anhydrous THF.
Acylation of 10. a. DCC/DMAP Protocol. The reaction was
accomplished using an 8-fold molar excess of Fmoc-N-protected
amino acid and a 4-fold molar excess of DCC relative to the amount of
linker on the beads. (The TentaGel NH2 beads carry 300 μmol of
1
mmol, 88%) as white crystals; mp = 153−154 °C. H NMR (300
MHz, CDCl3): δ 8.33 (br s, 1H), 8.01 (br d, J = 7.1 Hz), 7.95 (app dt,
Jd = 7.7 Hz, Jt = 1.4 Hz), 7.38 (app t, J = 7.8 Hz, 1H), 6.28 (t, J = 2.3,
1H), 5.94 (very broad app s, 2H), 5.00 (s, 1H), 3.54 (s, 6H), 2.70−
2.40 (m, 4H), 1.90−1.75 (m, 2H), 0.96 (s, 9H), 0.84 (s, 9H), 0.35 (s,
6H), 0.09 (s, 3H), −0.34 (s, 3H). 13C NMR (300 MHz, CDCl3): δ
166.8, 159.4, 140.9, 137.5, 137.0, 134.5, 131.1, 129.1, 127.6, 107.5−
106.0 (br), 101.0, 81.8, 65.3, 55.2, 27.5, 27.1, 26.0, 25.8, 25.1, 18.5,
18.0, −4.4, −4.6, −4.7, −4.9.
3-[2-(tert-Butyldimethylsilyloxy(3,5-dimethoxyphenyl)methyl)-
1,3-dithian-2-yl]benzoic Acid (8). Ester 7 (4.20 g, 6.61 mmol, 1.0
equiv) was suspended in 80% aqueous acetic acid (160 mL) and
heated with stirring from 50 to 75 °C over 45 min. The solution was
kept at 75 °C for an additional 15 min and then cooled. The solvent
was removed by rotary evaporation. The oily residue was coevaporated
with toluene (3 × 50 mL) to produce an off-white foam. The foam was
dissolved in a minimal amount (10−15 mL) of dichloromethane and
diluted with pentane (ca. 120 mL). After standing for 2 h at room
temperature followed by 2 h at 4 °C, crystallization of the product was
9414
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416

The Journal of Organic Chemistry
Article
NH2/g; quantitative loading results in 275 μmol of NH-Linker-OH/g
of beads due to the increased mass of the loaded beads.) The molar
ratio of DMAP/DCC was 1/10. The beads were swollen in
dichloromethane and drained. The Fmoc-N-protected amino acid to
be added to the linker was suspended/dissolved in anhydrous
dichloromethane (1 mL of CH2Cl2 for 0.2 mmol of the acid). If
necessary, anhydrous DMF was added dropwise to dissolve the amino
acid. The solution was cooled to 4 °C, and DCC was added as a dry
powder that dissolved immediately. After stirring for 15 min at 4 °C, a
white precipitate formed in the reaction mixture, which was removed
by filtration. The filtrate was added to the beads, followed by DMAP,
and the reaction was stirred overnight at room temperature. The beads
were drained, washed eight times with alternating dichloromethane
and methanol (5 mL/g beads), and dried. Fmoc protecting groups
were removed by treatment with 20% piperidine in DMF (5 mL/g
beads; two treatments: 5 and 15 min). The beads were drained,
washed with anhydrous DMF (3 times, 5 mL/g beads) and
dichloromethane (3 times, 5 mL/g beads), and dried.
b. 2,6-Dichlorobenzoyl Chloride Protocol. The reaction was
accomplished using a 5-fold molar excess of both Fmoc-N-protected
amino acid and 2,6-dichlorobenzoyl chloride and a 20-fold molar
excess of pyridine relative to the concentration of linker on the beads
(275 μmol of NH-Linker-OH/g of beads). The beads were swollen in
anhydrous DMF and then drained. An additional amount of DMF (ca.
1 mL per 0.2 g of dry resin) was added, followed by the solid Fmoc-N-
protected amino acid. The mixture was shaken for 10−15 min until all
the amino acid dissolved. 2,6-Dichlorobenzoyl chloride and pyridine
were added, and the reaction was stirred for 24 h at room temperature.
The beads were drained, washed eight times with alternating
dichloromethane and methanol, and dried. Fmoc protecting groups
were removed by treatment with 20% piperidine in DMF (5 mL/g
beads; two treatments: 5 and 15 min). The beads were drained,
washed with anhydrous DMF (3 times, 5 mL/g beads) and
dichloromethane (3 times, 5 mL/g beads), and dried.
35 mm) that was not removed. The lamp was placed at one focus
of an elliptical mirror while a water-jacketed Pyrex reactor was placed
at the second focus of the same mirror. The mirror was made by
simply bending a band of thin polished stainless steel plate over two
aluminum elliptical guides, which were made on an in-house computer
numerical controlled (CNC) mill (mirror dimensions: small diameter
50 mm, large diameter 100 mm, height 60 mmsame as the height of
the lamp active zone). Solvents were not degassed to remove O2.
Argon (Scientific grade, grade 6, oxygen content 20−50 ppm) was
slowly bubbled through the suspension of beads (20−100 mg) in the
solvents listed in Table 1 for cleavage of dipeptides (0.5−2.5 mL) or in
30% aqueous methanol (5 mL for a 0.04 mmol scale synthesis) for
cleavage of larger peptides to provide stirring during irradiation. Yields
1
of products released by photocleavage were determined by H NMR
in the presence of an added internal standard (p-toluic acid, sodium
acrylate, or pyridine hydrochloride) or by weight. The concentration
of amino groups on the TentaGel as provided by the manufacturer was
0.30 mmol/g. This concentration was confirmed to be accurate by UV
analysis of fulvene−piperidine adducts46 released during the acylation
of the beads with the first Fmoc-protected amino acid. The yields in
Table 1 of the text are calculated based on the capacity of amino
groups on the TentaGel beads.
After synthesis of Lys-Tyr-Arg-Arg-Arg-Pro-Arg-Arg-Ser-Gln-Arg-
Lys-Arg-Gly on a 0.04 mmol scale, the beads were irradiated for
40 min in 5 mL of 30% aqueous methanol. The solvent was drained,
and the beads were washed three times with 0.05% TFA in water. The
drained solvent and the aqueous washings were combined, filtered
through a 0.2 μm membrane filter, and concentrated by rotary
evaporation to ca. 0.3 mL. The peptide was precipitated by addition of
methanol (ca. 3 mL). The precipitate was collected by centrifugation,
washed with anhydrous methanol, and dried before NMR and HPLC
analysis. HPLC analysis was carried out using an Eclipse XDB-C18
column (4.6 × 150 mm). Samples were injected in water, isocratic
elution, 1 mL/min, 95% solvent A (0.1% aqueous TFA), 5% solvent B
(0.085% TFA/acetonitrile). Compounds were detected at 217 nm.
Predicted monoisotopic mass for the peptide: 1837.1208; observed
mass (major peak) 1837.0793.
Peptide Synthesis. Amino acids for peptide synthesis were
protected as follows: Fmoc-Tyr(Trt)-OH, Fmoc-Ser(tBu)-OH,
Fmoc-Thr(tBu)-OH, Fmoc-Asp(tBu)-OH, Fmoc-Gln(Trt)-OH,
Fmoc-Lys(Boc)-OH, Fmoc-Arg(Pbf)-OH, Fmoc-His(Boc)-OH,
Fmoc-Pro-OH, Fmoc-Val-OH, Fmoc-Gly-OH.
ASSOCIATED CONTENT
■
Beads carrying the first amino acid (attached as described above)
were swollen in anhydrous DMF (5 mL/1 g beads) for 15 min, then
washed three times with anhydrous DMF (5 mL/1 g beads each
washing), and finally suspended in the minimal amount of anhydrous
DMF required to achieve a homogeneous solution of the reagents. A
4-fold molar excess of solid Fmoc-protected amino acid was added to
the beads, and the mixture was gently shaken for 1−2 min for the acid
to dissolve. Solid PyOAP (4-fold molar excess) and solid HOAt (4-
fold molar excess) were added, followed by N,N-diisopropylethylamine
(8-fold molar excess). The reaction mixture was gently shaken for 25
min. The beads were drained and washed with anhydrous DMF (5
mL/g beads), and the coupling step was repeated a second time to
ensure complete derivatization of the beads. The beads were drained,
washed with anhydrous DMF (7 times, 5 mL/g beads) and anhydrous
dichloromethane (5 times, 5 mL/g beads), and dried.
Between coupling steps, Fmoc protecting groups were removed by
treatment with 20% piperidine in DMF (5 mL/g beads; two
treatments: 5 and 15 min). The beads were drained, washed with
anhydrous DMF (7 times, 5 mL/g beads), and used for the next step
of peptide synthesis using the protocol described above.
After completion of peptide synthesis, side chain protecting groups
and the dimethylketal protecting group on the linker were removed by
treatment with TFA/water/phenol/thioanisole (85 mL/5 mL/5 g/5
g) at room temperature for 4−6 h. The beads were washed with
dichloromethane (10 × 1 mL per 200 mg beads) and dried. Important
note: Once the dimethyl ketal protecting group on the linker is
removed, the beads should not be exposed to pH > 7, as side-reactions
and/or decomposition might occur.
S
* Supporting Information
NMR spectra for all synthesized materials, assignment of the
NMR signals for 6−10 and Lys-Tyr-Arg-Arg-Arg-Pro-Arg-Arg-
Ser-Gln-Arg-Lys-Arg-Gly, and a figure detailing the photo-
reactor assembly. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: shelley.copley@colorado.edu.
ACKNOWLEDGMENTS
■
This work was supported by the National Science Foundation
under Grant 0526747.
REFERENCES
■
(1) Orth, R.; Sieber, A. J. Org. Chem. 2009, 74, 8476−8479.
(2) Maurel, D.; Banala, S.; Laroche, T.; Johnsson, K. ACS Chem. Biol.
2010, 5, 507−516.
(3) Lee, H.-M.; Larson, D. R.; Lawrence, D. S. ACS Chem. Biol. 2009,
4, 409−427.
(4) Lee, H.-M.; Priestman, M. A.; Lawrence, D. S. J. Am. Chem. Soc.
2010, 132, 1446−1447.
(5) Katz, J. S.; Zhong, S.; Ricart, B. G.; Pochan, D. J.; Hammer, D. A.;
Burdick, J. A. J. Am. Chem. Soc. 2010, 132, 3654−3655.
(6) Sheehan, J. C.; Daves, G. D. Jr. J. Org. Chem. 1964, 29, 2006−
2008.
UV Irradiation. Irradiation was carried out using a 100 W “street
light type” medium-pressure mercury lamp (GE H100-A4/T). The
lamp had a transparent glass envelope (diameter about
9415
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416

The Journal of Organic Chemistry
Article
(7) Tjoeng, F. S.; Tam, J. P.; Merrifield, R. B. Int. J . Pept. Protein Res.
1979, 14, 262−274.
(41) Riniker, B.; Florsheimer, A.; Fretz, H.; Sieber, P.; Kamber, B.
Tetrahedron 1993, 49, 9307−9320.
̈
(42) Atherton, E.; Benoiton, N. L.; Brown, E.; Sheppard, R. C.;
Williams, B. J. J. Chem. Soc., Chem. Commun. 1981, 336−337.
(43) (a) Kirstgen, R.; Sheppard, R. C.; Steglich, W. J. Chem. Soc.,
Chem. Commun. 1987, 1870−1871. (b) Kirstgen, R.; Olbrich, A.;
Rehwinkel, H.; Steglich, W. Liebigs Ann. Chem. 1988, 5, 437−340.
(44) Sieber, P. Tetrahedron Lett. 1987, 28, 6147−6150.
(45) Carpino, L. A., Beyermann, M., Wenschuh, H., Bienert, M. Acc.
Chem. Res. 1996, 29, 268−274, and references therein.
(46) Ramage, R.; Green, J.; Ogunjobi, O. M. Tetrahedron Lett. 1989,
30, 2149−52.
(8) Sheehan, J. C.; Kazuo, U. K. J. Org. Chem. 1973, 38, 3771−3774.
(9) Brown, B. B.; Wagner, D. S.; Geysen, H. M. Mol. Diversity 1995,
1, 4−12.
(10) Holmes, C. P.; Jones, D. G. J. Org. Chem. 1995, 60, 2318−2319.
(11) Holmes, C. P. J. Org. Chem. 1997, 62, 2370−2380.
(12) Rock, R. S.; Chan, S. I. J. Org. Chem. 1996, 61, 1526−1529.
(13) Lee, H. B.; Balasubramanian, S. J. Org. Chem. 1999, 64, 3454−
3460.
(14) Cano, M.; Ladlow, M.; Balasubramanian, S. J. Org. Chem. 2002,
67, 129−135.
(15) Stork, G.; Zhao, K. Tetrahedron Lett. 1989, 30, 287−290.
(16) Loudon, G. M.; Radhakrishna, A. S.; Almond, M. R.; Blodgett, J.
K.; Boutin, R. H. J. Org. Chem. 1984, 49, 4272−4276.
(17) Barret, R.; Daudon, M. Tetrahedron Lett. 1990, 31, 4871−4872.
(18) Birk, C.; Voss, J. Tetrahedron 1996, 52, 12745−12760.
(19) Haidle, A. M.; Myers, A. G. Proc. Natl. Acad. Sci. U. S. A. 2004,
101, 12048−12053.
(20) Sheehan, J. C.; Wilson, R. M.; Oxford, A. W. J. Am. Chem. Soc.
1971, 93, 7222−7228.
(21) Pirrung, M. C.; Shuey, S. W. J. Org. Chem. 1994, 59, 3890−
3897.
(22) According to SciFinder, m-hydroxy-m-methoxybenzaldehyde is
available from only two suppliers: (a) Eurolabs Limited, United
Kingdom, semibulk or bulk quantities; (b) Pecchem Ltd., People’s
Republic of China.
(23) Rinnova,
2000, 6, 355−365.
(24) Patchornik, A.; Amit, B.; Woodward, R. B. J. Am. Chem. Soc.
1970, 92, 6333−6335.
́ ́ ́ ̌ ́
M.; Novakova, M.; Kasicka, V.; Jiracek, J. J. Pept. Sci.
(25) Pillai, V. N. R. Synthesis 1980, 1−26.
(26) Lloyd-Williams, P.; Gairi, M.; Albericio, F.; Giralt, E.
Tetrahedron 1991, 47, 9867−9880.
(27) Walker, J. W.; Reid, G. P.; McCray, J. A.; Trentham, D. R. J. Am.
Chem. Soc. 1988, 110, 7170−7177.
(28) Sheehan, J. C.; Wilson, R. M. J. Am. Chem. Soc. 1964, 86, 5277−
5281.
(29) Baldwin, J. E.; McConnaughie, A. W.; Moloney, M. G.; Pratt, A.
J.; Shin, S. B. Tetrahedron 1990, 46, 6879−6884.
(30) Pirrung, M. C.; Bradley, J. C. J. Org. Chem. 1995, 60, 1116−
1117.
́
(31) Cameron, J. F.; Willson, C. G.; Frechet, J. M. J. J. Am. Chem. Soc.
1996, 118, 12925−12937.
(32) Rock, R. S.; Chan, S. I. J. Am. Chem. Soc. 1998, 120, 10766−
10767.
̌
̌
rlj, B.; Sket, B.; Wirz, J. J. Phys. Chem. A
(33) Boudebous, H.; Kosm
2007, 111, 2811−2813.
(34) (a) Ma, K.; Kwok, W. M.; An, H.-Y.; Guan, X.; Fu, M. Y.; Toy,
P. H.; Phillips, D. L. Chem.Eur. J. 2010, 16, 5102−5118. (b) Pirrung,
M. C.; Ye, T.; Zhou, Z.; Simon, J. D. Phohochem. Photobiol. 2006, 82,
1258−1264.
(35) Wenkert, E.; Goodwin, T. E. Synth. Commun. 1977, 7, 409−415.
(36) Kates, S. A.; Diekmann, E.; El-Faham, A.; Ionescu, D.;
McGuiness, B. F.; Triolo, S. A.; Albericio, F.; Carpino, L. A. In
Techniques in Protein Chemistry VII; Marshak, D. R., Ed.; Academic:
New York, 1996; pp 513−523.
(37) Corey, E. J.; Cho, H.; Rucker, C.; Hua, D. H. Tetrahedron Lett.
̈
1981, 22, 3455−3458.
(38) Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. I. Anal.
Biochem. 1970, 34, 595−598.
(39) Upon acetylation of the hydroxyl of 12, the signal of the CHOH
proton at 5.08 ppm shifts downfield 5.9 ppm and becomes obscured
by the absorption of the polystyrene beads. We assume that the
complete disappearance of the signal d 5.08 ppm indicates quantitative
aminoacylation.
(40) Albericio, F.; Giralt, E. In Synthesis of Peptides and
Peptidomimetics; Goodman, M., Ed.; Thieme: Stuttgard, 2004; Vol.
E22a, pp 710−725, and references therein.
9416
dx.doi.org/10.1021/jo2017263|J. Org. Chem. 2011, 76, 9409−9416