2,4-Bis(hydroxymethyl)aniline as a building block for oligomers with self-eliminating and multiple release properties

Article abstract of DOI:10.1021/jo702484z

(Chemical Equation Presented) Linear self-eliminating (LSE) systems are oligomers of branched self-eliminating linkers that disassemble upon a single triggering event under complete degradation of the linear backbone, accompanied by the release of side-ch

Full text of DOI:10.1021/jo702484z

2,4-Bis(hydroxymethyl)aniline as a Building Block for Oligomers
with Self-Eliminating and Multiple Release Properties
Andre´ Warnecke* and Felix Kratz
Macromolecular Prodrugs, Tumor Biology Center, Breisacher Strasse 117, 79106 Freiburg, Germany
warnecke@tumorbio.uni-freiburg.de
ReceiVed NoVember 19, 2007
Linear self-eliminating (LSE) systems are oligomers of branched self-eliminating linkers that disassemble
upon a single triggering event under complete degradation of the linear backbone, accompanied by the
release of side-chain bound effector molecules. Enabling a controlled and almost simultaneous release of
different effectors (drugs) in defined ratios, LSE systems may gain importance for the development of
novel combination therapeutics. On the basis of the well-known self-eliminating p-aminobenzyloxycarbonyl
(PABC) linker, 2,4-bis(hydroxymethyl)aniline was considered a suitable branched linker for building
LSE systems that degrade by 1,6- and 1,4-benzyl elimination reactions. A first LSE model system based
on this linker was prepared in a simple procedure and was shown to release its effector payload efficiently
after activation. In addition, elimination model compounds were synthesized to study the release behavior
of LSE systems based on 2,4-bis(hydroxymethyl)aniline. It was found that chain degrading 1,6-benzyl
elimination occurs much faster than the effector releasing 1,4-elimination.
Introduction
cascade of elimination reactions which eventually lead to a
breakdown of the whole dendritic scaffold with a concomitant
release of the molecular payload. This simultaneous multiple
release of effector molecules upon a single activation step makes
these compounds attractive for a use as “intelligent” carriers
for drugs or imaging agents. Nevertheless, there are some present
drawbacks of this concept: The main problem is that only a
restricted number of drug molecules fit into the limited space
of the outer shell of the dendrimer. G3 dendrons with eight small
dye molecules5,6 and a G2 dendron with four molecules of the
bulky drug paclitaxel7 are the largest self-eliminating dendrimer
conjugates that could be synthesized up to now. Furthermore,
For developing more sophisticated prodrug technologies for
cancer chemotherapy, the interest in chemical architectures that
enable a site-specific release of carrier-bound drugs has grown
rapidly over the last years.1,2 Self-immolative dendrimers are
recent examples of complex molecular structures designed for
a controlled and multiple release of small molecules.3,4 Based
on self-eliminating linkers as branching units, self-immolative
dendrimers can be terminally loaded with various effector/
reporter molecules. Activation at the focal point initiates a
(1) Papot, S.; Tranoy, I.; Tillequin, F.; Florent, J.-C.; Gesson, J.-P. Curr.
Med. Chem. Anti-Cancer Agents 2002, 2, 155-185.
(2) Kratz, F.; Mu¨ller, I. A.; Ryppa, C.; Warnecke, A. ChemMedChem
2008, 3, 20-53.
(3) Meijer, E. W.; Van Genderen, M. H. Nature 2003, 426, 128-129.
(4) McGrath, D. V. Mol. Pharm. 2005, 2, 253-263.
(5) Amir, R. J.; Pessah, N.; Shamis, M.; Shabat, D. Angew. Chem., Int.
Ed. 2003, 42, 4494-4499.
(6) Szalai, M. L.; Kevwitch, R. M.; McGrath, D. V. J. Am. Chem. Soc.
2003, 125, 15688-15689.
10.1021/jo702484z CCC: $40.75 © 2008 American Chemical Society
Published on Web 01/19/2008
1546
J. Org. Chem. 2008, 73, 1546-1552

2,4-Bis(hydroxymethyl)aniline
system. Since respective o-aminobenzyl derivatives show an
analogous reaction (1,4-elimination),11 2,4-bis(hydroxymethyl)-
aniline (1) was chosen as an appropriate branched double release
linker for our approach. Scheme 1 shows one of the two possible
disassembly mechanisms for LSE systems based on 1: 1,6-
elimination leads to the breakdown of the linear chain followed
by release of the effectors through 1,4-elimination (in theory,
these reactions could also take place in reversed order).
Apparently, the rates for both benzyl elimination reactions as
well as for the regeneration of the aminobenzyl system from
the quinoidal intermediates are crucial for the double-release
efficiency.
In this work, we describe the synthesis of monoprotected 2,4-
bis(hydroxymethyl)aniline and its use as a self-eliminating
double release linker and building block for LSE systems. Model
compounds with tryptamine were prepared to study the elimina-
tion properties. Furthermore, by appending two linker units, a
first linear self-eliminating system was constructed. Elimination
kinetics of all model compounds were determined by using
HPLC.
FIGURE 1. Examples for a dendritic and a linear implementation of
a self-eliminating structure comprised of three linker units (L) that
disassemble upon activation to release four effector molecules (E).
dendritic structures are not suitable for conveniently combining
different drugs. For instance, 13 steps were necessary to
synthesize a G1 dendrimer that was loaded with one molecule
each of the anticancer drugs camptothecin, etoposide, and
doxorubicin.8
Results
As a straightforward approach for the construction of effector/
reporter-loaded, carbamate-linked oligomers of 1, we made use
of a simple two-step procedure: (1) conversion of the 4-hy-
droxybenzyl group into an activated 4-nitrophenyl (Np) carbon-
ate and (2) appending another linker through its amino group
by forming a carbamate bond. Unfortunately, it is not possible
to employ linker-effector derivatives as building blocks in this
strategy since effective linkers with an unprotected amino group
will immediately undergo elimination of the effector. Hence,
we planned to use monoprotected 1 for building up the linear
backbone followed by removal of the protective groups,
activation with 4-nitrophenyl chloroformate, and coupling with
the amine-containing reporter molecule. Monoprotected 2,4-bis-
(hydroxymethyl)aniline was readily prepared in a 5-step pro-
cedure starting with ethyl 3-methyl-4-nitrobenzoate (Scheme 2).
After radical bromination, conversion to the benzyl alcohol using
silver sulfate, TBS protection, and a successive reduction of
both nitro and ester group, linker 6 was obtained in an acceptable
overall yield.
For designing novel combination therapeutics with variable
drug ratios,9 nondendritic carrier topologies, i.e., linear self-
eliminating (LSE) systems, are likely to offer more potential.
Such linear systems are based on branched self-eliminating
linkers as monomer unitssthe same as for self-immolative
dendrimers (Figure 1). For n effector/reporter molecules, n -
1 linker units must be incorporated. Chemical or enzymatic
activation of the trigger causes the molecule to disassemble in
two directions: the bonds between two linkers that form the
linear backbone as well as the bonds between the linkers and
the effector molecules are cleaved by elimination reactions. For
drug delivery applications, this disassembly should occur
sufficiently fast in aqueous media to ensure high local drug
levels.
Very efficient self-eliminating structures are based on p-
aminobenzyloxycarbonyl (PABC) linkers first introduced by
Katzenellenbogen.10 With the amino group acylated or otherwise
protected, the system is considerably stable under physiologic
conditions whereas the linker with a free amino group undergoes
a rapid 1,6-elimination of unstable carbamic or carbonic acids
which in turn decarboxylate to the respective amines or alcohols.
To evaluate the potential of 1 as a building block for LSE
systems, we were interested in answering the following ques-
tions: (1) Do single 1,4- and 1,6-benzylelimination reactions
proceed with comparable reaction rates? (2) Do they also occur
when combined in a double elimination system? (3) Does chain
degradation in a LSE system occur with the same rate as the
release of the reporter molecules?
To clarify these issues we developed four different model
compounds. First, we synthesized the single 1,4- and 1,6-
benzylelimination model systems 8a and 8b (Scheme 3) that
are tryptamine-carbamoyl derivatives of 4- and 2-nitrobenzyl
alcohol, respectively. Since reduction of aromatic nitro groups
is a facile and fast reaction, 8a and 8b are suitable precursors
for the respective aminobenzyl compounds. Tryptamine serves
as a model for an amino-functionalized effector/drug and can
be readily quantified using HPLC. To study the double release
behavior of bis-carbamoyl derivatives of 1, model compound
12 bearing two molecules of tryptamine was synthesized from
4-nitrobenzyl alcohol and 6 in 5 steps (Scheme 3). Furthermore,
The intermediately formed iminoquinone methide is trapped
by nucleophiles (water) and thereby regenerates the aminobenzyl
(7) de Groot, F. M.; Albrecht, C.; Koekkoek, R.; Beusker, P. H.;
Scheeren, H. W. Angew. Chem., Int. Ed. 2003, 42, 4490-4494.
(8) Haba, K.; Popkov, M.; Shamis, M.; Lerner, R. A.; Barbas, C. F.,
3rd; Shabat, D. Angew. Chem., Int. Ed. 2005, 44, 716-720.
(9) Mayer, L. D.; Janoff, A. S. Mol. InterVentions 2007, 7, 216-223.
(10) Carl, P. L.; Chakravarty, P. K.; Katzenellenbogen, J. A. J. Med.
Chem. 1981, 24, 479-480.
(11) Wakselman, M. NouV. J. Chem. 1983, 7, 439-447.
J. Org. Chem, Vol. 73, No. 4, 2008 1547

Warnecke and Kratz
SCHEME 1. One of Two Possible Disassembly Pathways of 2,4-Bis(hydroxymethyl)aniline-Based Linear Self-Eliminating (LSE)
Systems in Aqueous Media
SCHEME 2. Synthesis of 6
by applying the above-mentioned strategy of two alternating
activation and coupling steps, followed by removal of all
protective groups, activation with 4-nitrophenyl chloroformate,
and reaction with tryptamine, compound 17 was obtained as a
first simple LSE system that enables the release of three effector
molecules (Scheme 3). In analogy to 8a and 8b, self-disassembly
of 12 and 17 was triggered by reduction of the 4-nitrobenzy-
loxycarbonyl group. The formed self-eliminating PABC linker
then initiates the cascade of elimination reactions.
Reduction of the nitro groups was accomplished by treating
the model compounds with an excess of zinc powder and 5%
AcOH in MeCN. Under these conditions, a reaction time of 1
min proved to be sufficient for complete conversion of the
starting material as shown by HPLC. Aqueous dilutions of the
samples (containing ∼0.3% AcOH, pH 3) were stored at 37 °C
and analyzed at appropriate intervals. For determination of the
disassembly kinetics the formed tryptamine was quantified.
Divided by the tryptamine concentrations c∞ for complete
conversion (determined after adequately long reaction times),
relative tryptamine concentrations c/c∞ were obtained. Figure
2 shows the time-dependent release of tryptamine from 8a, 8b,
12, and 17. To verify that the observed release of tryptamine is
due to benzyl elimination reactions and does not result from an
unspecific hydrolysis of the carbamates, analogous experiments
were carried out without the addition of zinc powder. For all
model compounds no substantial release of tryptamine could
be detected after one week storage at pH 3 and 37 °C.
strategy we employed 6 as a protected building block that was
activated, dimerized, and loaded with tryptamine as a reporter
to afford 17 as a first model LSE system. Compound 17 was
shown to disassemble upon reduction of the nitro group as
indicated by the formation of tryptamine with >50% of the
reporter being released after 4 h (Figure 2). To gain deeper
insights into the elimination mechanism of reduced 17, model
compounds 8a, 8b, and 12 were synthesized that release
tryptamine in a single 1,6-benzyl elimination, a single 1,4-benzyl
elimination, and a combined 1,6- and 1,4-benzyl elimination,
respectively. After reduction of the nitro groups, all model
compounds underwent self-elimination (Figure 2) albeit the
release of tryptamine from 8a (1,6-benzyl elimination) occurred
much more rapidly than from the other model compounds with
at least one 1,4-benzyl elimination step being involved. In a
logarithmic plot (Figure 3) the expected first-order elimination
kinetics of the reduced forms of 8a and 8b was found in a linear
correlation between ln(1 - c/c∞) and t. The calculated half-
lives (37 °C, pH 3) are ∼4 min for the 1,6-benzyl elimination
(8a) and ∼120 min for the 1,4-elimination reaction (8b), which
is a remarkable difference of at least 1 order of magnitude.
When analyzing the tryptamine release behavior of the double
release compound (reduced 12), three stages can be clearly
distinguished (Figures 2 and 3): a rapid release within the first
20 min followed by a phase of relative stagnation (20-45 min)
and a constant, but slower release (45-150 min). Comparing
the slope of the logarithmic plot of reduced 12 with the
linearized graphs of the single benzyl eliminations (Figure 3),
it can be assumed that the first stage is the 1,6-benzyl elimination
of the first tryptamine molecule and the last stage the respective
1,4-elimination of the second reporter molecule. The observed
Discussion
One major aim of our work was to develop a facile synthetic
approach to the first linear self-eliminating system. In our
1548 J. Org. Chem., Vol. 73, No. 4, 2008

2,4-Bis(hydroxymethyl)aniline
SCHEME 3. Synthesis of Model Compounds
stagnation phase could be explained by the required regeneration
of the aminobenzyl system from the chinoidal intermediate. It
is therefore most likely that disassembly of the LSE system (17)
mainly proceeds according to the mechanism depicted in
Scheme 1. Due to the more complex elimination mechanism
of reduced 17, however, an equally characteristic release pattern
as observed for reduced 12 could not be found.
Liberating >50% of the reporter molecule within 4 h, the
release efficiency of 17 proved to be satisfactory for biomedical
applications, even under acidic conditions (pH 3). Since it can
be assumed that only the aniline form of the linkers will undergo
elimination, a strong influence of the pH on the elimination
rate should be expected. Assuming further that the pKa of
protonated 1 is in the same range as for anilinium (pKa ) 4.6),
a high percentage of the amino groups are protonated at the
reaction conditions used (0.3% AcOH, pH 3) thus making the
reaction sufficiently slow to be analyzed by HPLC. For the
elimination of the reduced forms of 8a and 8b, a first-order
kinetics with a preceding acid-base equilibrium can be
described as:
With Ka(anilinium) ) 10-4.6 the overall rate constant is 0.025k
at pH 3 and 0.996k at pH 7, i.e., compared with our reaction
conditions the single benzyl elimination step should proceed
nearly 40 times faster at neutral pH. As the required regeneration
step for the disassembly of 12 and 17 is an addition of water
(or OH-) to the iminoquinone methide, it should likewise be
accelerated at higher pH. Thus, it can be expected that also the
multiple release properties of compounds 12 and 17 would be
significantly enhanced at pH 7. Unfortunately, all attempts to
determine the release kinetics of our model compounds at neutral
pH failed due to a spontaneous precipitation of the reduced
(amino) forms in aqueous solution. However, Shabat et al.
recently reported on a similar PABC-based triple release linker
that liberates three molecules of tryptophane in consecutive 1,6-
and 1,4-benzyl elimination steps.12 At pH 7.4 complete release
of the reporter molecule was observed within 40 min after
enzymatic activation supporting our assumption that the relevant
benzyl eliminations occur with considerably higher reaction rates
when carried out at neutral pH.
In summary, we have developed a first LSE model system
(17) that enables an effective release of reporter/effector
molecules upon activation of a trigger unit. By analyzing the
elimination kinetics of the employed double release linker (1),
we found the 1,6-benzyl elimination to occur much more rapidly
Ka
d[tryptamine]
) k
[reactant]
(2)
Ka + [H+]
dt
wherein k is the rate constant for the unprotonated form and Ka
describes the acid dissociation constant of the anilinium salt.
(12) Sagi, A.; Segal, E.; Satchi-Fainaro, R.; Shabat, D. Bioorg. Med.
Chem. 2007, 15, 3720-3727.
J. Org. Chem, Vol. 73, No. 4, 2008 1549

Warnecke and Kratz
FIGURE 2. Release of tryptamine from 8a, 8b, 12, and 17 after reduction with Zn/AcOH.
FIGURE 3. Release of tryptamine from 8a, 8b, and 12 after reduction with Zn/AcOH (logarithmic plot).
followed by 24 h of heating was repeated until TLC (1:4 EtOAc-
hexane) indicated nearly complete conversion of the starting
material. The solids were removed by filtration over celite then
washed with DCM, and the filtrate was concentrated in vacuo. The
dark brown residue was crystallized from EtOAc/hexane affording
44.29 g (64%) of pure 2 as a pale yellow solid: mp 105-106 °C
than the respective 1,4-elimination. Since both reactions proceed
sufficiently fast, 1 proved to be a suitable building block for
the construction of LSE systems. The synthetic strategy used
for the preparation of 17, however, is not feasible for the
construction of longer and more complex LSE systems,
especially, when aiming at combining different effector mol-
ecules or weakly reactive effectors such as aromatic amines or
alcohols. A novel synthetic approach employing polymerizable
linker-effector building blocks that circumvent the risk of any
premature decomposition is currently under investigation.
1
(lit.13 compound was obtained as an oil); H NMR (400 MHz,
CDCl3) δ 1.43 (t, J ) 7.2 Hz, 3H), 4.43 (q, J ) 7.1 Hz, 2H), 4.83
(s, 2H), 8.05 (d, J ) 8.6 Hz, 1H), 8.12 (dd, J1 ) 1.8 Hz, J2 ) 8.4
Hz, 1H), 8.22 (d, J ) 1.8 Hz, 1H); 13C NMR (100 MHz, CDCl3)
δ 14.2, 27.9, 62.1, 125.5, 130.6, 133.0, 133.6, 134.9, 150.4, 164.1;
+
MS (CI+ (NH3)) m/z 208.1 ([M - Br]+, 100), 305.1 (M + NH4
24).
,
Experimental Section
Ethyl 3-(Hydroxymethyl)-4-nitrobenzoate (3). A suspension
of 2 (10.0 g, 34.7 mmol) and Ag2SO4 (10.8 g, 34.7 mmol) in
dioxane/water (1:1, 300 mL) was heated under reflux in the dark
for 3 h. After cooling to room temperature, brine (150 mL) was
added and stirring was continued for 10 min. The mixture was
Ethyl 3-(Bromomethyl)-4-nitrobenzoate (2). A suspension of
ethyl 3-methyl-4-nitrobenzoate (50.0 g, 239 mmol) and NBS (49.3
g, 277 mmol) in CCl4 (600 mL) was heated for 30 min in a three-
necked flask equipped with a Dean-Stark apparatus to remove any
traces of water. After cooling and removal of the Dean-Stark
apparatus, a 20-mg portion of (BzO)2 was added and the reaction
mixture was heated under reflux for 24 h. Addition of (BzO)2
(13) Damen, E. W.; Nevalainen, T. J.; van den Bergh, T. J.; de Groot,
F. M.; Scheeren, H. W. Bioorg. Med. Chem. 2002, 10, 71-77.
1550 J. Org. Chem., Vol. 73, No. 4, 2008

2,4-Bis(hydroxymethyl)aniline
extracted with EtOAc (5 × 150 mL) and the combined organic
layers were dried over MgSO4. After removing the volatiles in
vacuo, the residue was purified by flash chromatography (1:4
EtOAc-hexane) to yield 6.26 g (80%) of 3 as a colorless solid:
4-Nitrobenzyl 2-(1H-Indol-3-yl)ethylcarbamate (1,6-Elimina-
tion Model Compound 8a). 8a was obtained from 7a and
tryptamine following the general procedure. Flash chromatography
of the crude product (1:2 EtOAc-hexane) afforded 8a (98%) as a
light yellow solid (purity (HPLC) 98%): mp 120 °C dec; 1H NMR
(400 MHz, DMSO-d6) δ 2.85 (t, J ) 7.5 Hz, 2H), 3.26-3.34 (m,
2H), 5.17 (s, 2H), 6.97 (t, J ) 7.8 Hz, 1H), 7.06 (t, J ) 8.1 Hz,
1H), 7.15 (d, J ) 2.3 Hz, 1H), 7.34 (d, J ) 8.1 Hz, 1H), 7.50-
7.55 (m, 2H), 7.57 (d, J ) 8.5 Hz, 2H), 8.23 (d, J ) 8.9 Hz, 2H),
10.82 (s, 1H); 13C NMR (100 MHz, DMSO-d6) δ 25.3, 41.2, 63.8,
111.3, 111.4, 118.10, 118.12, 120.8, 122.6, 123.4, 127.1, 127.9,
136.1, 145.3, 146.8, 155.7; MS (ESI+) m/z 362.1 (M + Na+, 100),
340.1 (M + H+, 49).
2-Nitrobenzyl 2-(1H-Indol-3-yl)ethylcarbamate (1,4-Elimina-
tion Model Compound 8b). 8b was obtained from 7b and
tryptamine following Procedure B. Flash chromatography of the
crude product (1:2 EtOAc-hexane) afforded 8b (86%) as a light
yellow solid (purity (HPLC) 97%): mp 100-101 °C dec; 1H NMR
(400 MHz, DMSO-d6) δ 2.85 (t, J ) 7.5 Hz, 2H), 3.25-3.32 (m,
2H), 5.37 (s, 2H), 6.97 (t, J ) 7.6 Hz, 1H), 7.06 (t, J ) 8.0 Hz,
1H), 7.15 (d, J ) 2.1 Hz, 1H), 7.33 (d, J ) 8.1 Hz, 1H), 7.50-
7.63 (m, 4H), 7.78 (t, J ) 7.6 Hz, 1H), 8.11 (d, J ) 8.4 Hz, 1H),
10.82 (s, 1H); 13C NMR (100 MHz, DMSO-d6) δ 25.3, 41.2, 61.8,
111.3, 111.4, 118.10, 118.13, 120.8, 122.6, 124.7, 127.1, 128.6,
128.9, 133.0, 134.0, 136.1, 147.0, 155.6; MS (ESI+) m/z 362.1 (M
+ Na+, 100), 340.1 (M + H+, 30).
Double Release Model Compound 12. 12 was obtained from
11 and tryptamine following the general procedure. Flash chroma-
tography of the crude product (1:1 EtOAc-hexane) afforded 12
(96%) as a yellow film (purity (HPLC) 97%,; purity (NMR) 93%
(contains approximately 7% ethyl acetate as the major contamina-
tion)): 1H NMR (400 MHz, DMSO-d6) δ 2.78-2.87 (m, 4H),
3.22-3.32 (m, 4H), 4.99 (s, 2H), 5.07 (s, 2H), 5.31 (s, 2H), 6.91-
6.99 (m, 2H), 7.01-7.08 (m, 2H), 7.10-7.15 (m, 2H), 7.28-7.39
(m, 5H), 7.50-7.55 (m, 2H), 7.44-7.54 (m, 4H), 7.65-7.71 (m,
2H), 8.22-8.28 (m, 2H), 9.38 (s, 1H), 10.79 (s, 2H); 13C NMR
(100 MHz, DMSO-d6) δ 25.35, 25.39, 41.14, 41.19, 61.6, 64.6,
64.7, 111.3, 111.44, 111.50, 118.06, 118.10, 118.11, 120.8, 122.5,
123.5, 127.09, 127.11, 127.9, 128.2, 128.5, 135.1, 136.1, 144.6,
146.9, 153.8, 155.9, 156.2; MS (ESI+) m/z 705.2 (M + H+, 100),
727.2 (M + Na+, 48).
LSE System Model Compound 17. 17 was obtained from 16
and tryptamine following the general procedure. Flash chromatog-
raphy of the crude product (3:2 EtOAc-hexane) afforded 17 (96%)
as a pale yellow foam (purity (HPLC) 95%; purity (NMR) 93%
(contains approximately 5% of ethyl acetate, 2% of DMF, and traces
of n-hexane)): 1H NMR (400 MHz, DMSO-d6) δ 2.78-2.86 (m,
6H), 3.22-3.32 (m, 6H), 4.98 (s, 2H), 5.05 (s, 2H), 5.09 (s, 2H),
5.12 (s, 2H), 5.30 (s, 2H), 6.91-6.99 (m, 3H), 7.02-7.08 (m, 3H),
7.10-7.14 (m, 3H), 7.26-7.55 (m, 15H), 7.66-7.70 (m, 2H),
8.22-8.27 (m, 2H), 9.23 (s, 1H), 9.40 (s, 1H), 10.79 (s, 3H); 13C
NMR (100 MHz, DMSO-d6) δ 25.44, 25.49, 41.3, 59.7, 61.68,
61.73, 64.7, 64.9, 111.3, 111.55, 111.60, 118.17, 118.21, 120.9,
122.6, 123.6, 127.2, 128.0, 128.3, 128.5, 129.0, 133.1, 133.5, 135.3,
135.5, 136.2, 144.6, 147.0, 153.9, 154.2, 156.0, 156.23, 156.27;
MS (ESI+) m/z 1070.3 (M + H+, 100), 1092.2 (M + Na+, 40),
1108.2 (M + K+, 11).
1
mp 62-63 °C (lit.13 compound was obtained as an oil); H NMR
(400 MHz, CDCl3) δ 1.41 (t, J ) 7.1 Hz, 3H), 2.78 (br s, 1H),
4.42 (q, J ) 7.0 Hz, 2H), 5.01 (s, 2H), 8.08 (s, 2H), 8.41 (s, 1H);
13C NMR (100 MHz, CDCl3) δ 14.2, 61.94, 61.96, 124.9, 129.4,
130.8, 135.1, 137.1, 149.8, 164.7; MS (CI+ (NH3)) m/z 243.2 (M
+ NH4+, 100).
Ethyl 3-((tert-Butyldimethylsilyloxy)methyl)-4-nitrobenzoate
(4). A solution of 3 (5.65 g, 25.1 mmol) in anhydrous DMF (10
mL) was added dropwise within 30 min to a stirred solution of
tert-butyldimethylchlorosilane (4.16 g, 27.6 mmol) and imidazole
(2.56 g, 37.7 mmol) in anhydrous DMF (40 mL). After stirring at
room temperature for 16 h, the reaction mixture was poured into
water (400 mL) and extracted with DCM (3 × 100 mL). The
combined organic layers were washed with water (2 × 100 mL)
and dried over MgSO4. After removing the volatiles in vacuo, the
residue was recrystallized from methanol/water (6:1) affording 7.91
g (93%) of pure 4 as colorless needles: mp 77 °C; 1H NMR (400
MHz, CDCl3) δ 0.15 (s, 6H), 0.99 (s, 9H), 1.42 (t, J ) 7.2 Hz,
3H), 4.42 (q, J ) 7.1 Hz, 2H), 5.10 (s, 2H), 8.05-8.08 (m, 1H),
8.12 (d, J ) 8.3 Hz, 1H), 8.59-8.61 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ -5.4, 14.2, 18.3, 25.8, 61.76, 61.81, 124.6, 128.6, 129.6,
134.9, 138.5, 148.8, 164.9; MS (CI+ (NH3)) m/z 326.2 (100), 340.2
(M + H+, 49), 357.3 (M + NH4+, 23).
Ethyl 4-Amino-3-((tert-butyldimethylsilyloxy)methyl)benzoate
(5). 4 (11.0 g, 32.4 mmol) and zinc powder (10 g) were suspended
in DCM (500 mL) in a double-necked flask equipped with a
condenser. To the stirred mixture was added acetic acid (18.4 mL,
0.32 mol) in DCM (100 mL) dropwise within 30 min and stirring
was continued for 1 h. Excessive zinc powder was filtered off and
the filtrate was washed with saturated NaHCO3 solution (3 × 150
mL), dried over MgSO4, and evaporated in vacuo to yield 10.0 g
1
(100%) of 5 as a colorless solid: mp 47 °C; H NMR (400 MHz,
CDCl3) δ 0.07 (s, 6H), 0.89 (s, 9H), 1.36 (t, J ) 7.2 Hz, 3H), 4.33
(q, J ) 7.2 Hz, 2H), 4.67 (br s, 2 H), 4.71 (s, 2H), 6.62 (d, J ) 8.3
Hz, 1H), 7.72 (d, J ) 1.9 Hz, 1H), 7.80 (dd, J1 ) 2.0 Hz, J2 ) 8.3
Hz, 1H); 13C NMR (100 MHz, CDCl3) δ -5.3, 14.4, 18.2, 25.8,
60.3, 64.8, 114.6, 119.2, 123.9, 130.3, 130.8, 150.7, 166.8; MS
(CI+ (NH3)) m/z 178.1 ([M - OTBS]+, 100), 195.1 ([M - OTBS
+ NH3]+, 33), 310.3 (M + H+, 17).
2-((tert-Butyldimethylsilyloxy)methyl)-4-(hydroxymethyl)-
aniline (6). A solution of 5 (6.90 g, 22.3 mmol) in anhydrous THF
(250 mL) was cooled to -78 °C and a 1 M solution of DIBALH
in DCM (66.9 mL) was added dropwise within 75 min. The reaction
mixture was stirred for 24 h at -78 °C and then poured into 500
mL of a saturated solution of Rochelle salt. Subsequently, the
mixture was extracted with diethyl ether (3 × 150 mL), the organic
layers were dried over Na2SO4, and the solvent was evaporated.
The residue was purified by flash chromatography (1:2 EtOAc-
hexane) to yield 4.63 g (75%) of 6 as a colorless solid: mp 49-50
°C; 1H NMR (400 MHz, CDCl3) δ 0.08 (s, 6H), 0.90 (s, 9H), 3.25
(br s, 3 H), 4.54 (s, 2H), 4.68 (s, 2H), 6.65 (d, J ) 8.0 Hz, 1H),
7.04 (d, J ) 2.0 Hz, 1H), 7.09 (dd, J1 ) 2.0 Hz, J2 ) 8.1 Hz, 1H);
13C NMR (100 MHz, CDCl3) δ -5.3, 18.2, 25.8, 64.8, 65.3, 115.7,
125.2, 127.9, 128.0, 130.3, 145.8; MS (ESI+) m/z 267.5 (M+, 100),
136.0 ([M - OTBS]+, 73).
HPLC. Chromatographic conditions: Nucleosil C8 column
(100-5, 250 × 4 mm) from Macherey-Nagel; flow: 1.0 mL/min;
mobile phase A: 30% MeCN, 70% H2O, 0.1% TFA; mobile phase
B: 80% MeCN, 20% H2O, 0.1% TFA; gradient: 0-5 min mobile
phase A isocrat., 5-15 min 0% to 100% mobile phase B, 15-20
min 100% mobile phase B isocrat., 20-25 min 0% to 100% mobile
phase A, 25-40 min mobile phase A isocrat.; injection volume:
50 µL; detection at 280 nm.
Cleavage Studies. Reduction with Zn/AcOH was performed by
adding approximately 10 mg of zinc powder and 25 µL of glacial
acetic acid to 475 µL of a 1 mM stock solution (acetonitrile) of
the respective model compound. The mixture was vortexed for 1
General Procedure for the Reaction of 4-Nitrophenoxycar-
bonyl-Activated Carbonates with Amines. To a mixture of
HOBt (1.5 mmol, 230 mg), finely ground molecular sieve 4 Å
(250 mg), and the amine (1.5 equiv per activated carbonate
group) in 5 mL of anhydrous DMF was added the activated alcohol
(1.5 mmol). After stirring for 16 h at room temperature, the
molecular sieve was filtered off and the solvent was removed in
vacuo. Further workup was performed as described for the
individual compounds.
J. Org. Chem, Vol. 73, No. 4, 2008 1551

Warnecke and Kratz
min, excessive zinc was immediately filtered off by using a 0.2
µm syringe filter, and the resulting clear solution was diluted with
tempered (37 °C) water/acetonitrile 9:1 to a final concentration of
60 µM. The samples were stored in a heating block (37 °C) and
analyzed with HPLC at appropriate intervals. The relative degree
of conversion was calculated by dividing the time-dependent
tryptamine concentration ct (from the peak area at the respective
time) by the tryptamine concentration c∞ at nearly complete
conversion (peak area determined after 1.5 h for 8a or 24 h for 8b,
12, and 17, respectively).
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft (DFG). We thank Tamara Huck and
Karin Scheuermann for their excellent technical assistance.
Supporting Information Available: Additional synthetic pro-
1
cedures, H and 13C NMR spectra of all compounds, as well as
chromatograms of the model compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO702484Z
1552 J. Org. Chem., Vol. 73, No. 4, 2008