Disassembly kinetics of quinone-methide-based self-immolative spacers that contain aromatic nitrogen heterocycles

Article abstract of DOI:10.1002/asia.201400051

We prepared several pyridine- and pyrimidine-based self-immolative spacer groups to evaluate the significance of the resonance energy of the spacer aromatic ring on the kinetics of 1,4- and 1,6-elimination reactions, which govern spacer disassembly. Subsequently, we relied on a photoactivation procedure to accurately analyze the disassembly kinetics. Beyond providing new results that are relevant for deriving quantitative structure-property relationships, herein, we demonstrate that pH value can be used as an efficient parameter to finely control the disassembly time of a self-immolative spacer after an initial activation. Burn rubber: Kinetic analysis of the pH-dependent disassembly of self-immolative spacers that contain aromatic nitrogen heterocycles was performed. Electron-poor pyrimidine cores exhibited the longest disassembly times. This study confirms the trend that electron-rich aryl cores accelerate self-immolation.

Full text of DOI:10.1002/asia.201400051

FULL PAPER
DOI: 10.1002/asia.201400051
Disassembly Kinetics of Quinone-Methide-Based Self-Immolative Spacers
that Contain Aromatic Nitrogen Heterocycles
Ahmed Alouane,[a, b, c] Raphaꢀl Labruꢁre,*[a, b, c, e] Katherine J. Silvestre,[a, b]
Thomas Le Saux,[c, d] Frꢂdꢂric Schmidt,*[a, b] and Ludovic Jullien*[c, d]
Abstract: We prepared several pyri-
dine- and pyrimidine-based self-immo-
lative spacer groups to evaluate the sig-
nificance of the resonance energy of
the spacer aromatic ring on the kinetics
of 1,4- and 1,6-elimination reactions,
which govern spacer disassembly. Sub-
sequently, we relied on a photoactiva-
tion procedure to accurately analyze
the disassembly kinetics. Beyond pro-
viding new results that are relevant for
deriving quantitative structure–proper-
ty relationships, herein, we demon-
strate that pH value can be used as an
efficient parameter to finely control
the disassembly time of a self-immola-
tive spacer after an initial activation.
Keywords: kinetics
heterocycles photochemistry
pyrimidines · self-immolation
·
nitrogen
·
·
Introduction
the location of the activation position with substrate release.
Hence, accurate kinetics analyses of the disassembly of self-
immolative spacer groups have been critical in determining
their scope of application. Useful structure–kinetics relation-
ships have been established for the prediction of significant
features, such as the stability of the covalent precursor and
the rate of substrate release.[2a,5,6]
First described by Katzenellenbogen and co-workers in
1981,[1] self-immolative spacer groups combine the cleavage
of a covalent bond with the spontaneous release of one or
several reporter groups.[2] More precisely, in these stimuli-re-
sponsive covalent chemical structures, a primary intramolec-
ular reaction drives a second reaction, which releases a de-
sired substrate (Scheme 1).[1]
Self-immolative spacer groups have now been extensively
used in the design of prodrugs[3] and in bioanalysis.[4] Useful-
ly, upon controlling the kinetics of substrate release after ac-
tivation, these groups make it possible to spatially correlate
We have recently developed several procedures that rely
on light activation to analyze the disassembly kinetics of
self-immolative spacer groups down to a temporal resolution
of 20 ms.[6] Our procedures have permitted us to analyze at
different pH values and temperatures the disassembly kinet-
ics of quinone-methide-based self-immolative spacers with
1) various electron-donating or -withdrawing substituents
and 2) various links between the benzenic core and the sub-
strate. For the purpose of comparison, we were interested in
applying the same procedures to the investigation of pyridi-
none-methide elimination, which was previously described
as effective.[7] In particular, we wondered whether fast disas-
sembly could originate from the heteroaromatic ring[8] of
the spacer and whether further acceleration could result
from a further lowering of the resonance energy of the aro-
matic ring.[5d–e] Herein, we prepared several pyridine- and
pyrimidine-based self-immolative spacer groups and ana-
lyzed the pH-dependent disassembly kinetics that were asso-
ciated with their 1,4- and 1,6-eliminations.
[a] A. Alouane, Dr. R. Labruꢀre, K. J. Silvestre, Dr. F. Schmidt
Institut Curie, Centre de Recherche
26 rue d’Ulm, Paris F-75248 (France)
Fax : (+33)1-56-24-66-31
E-mail: frederic.schmidt@curie.fr
[b] A. Alouane, Dr. R. Labruꢀre, K. J. Silvestre, Dr. F. Schmidt
CNRS, UMR 176
26 rue d’Ulm, Paris F-75248 (France)
[c] A. Alouane, Dr. R. Labruꢀre, Dr. T. Le Saux, Prof. Dr. L. Jullien
Ecole Normale Supꢁrieure, Dꢁpartement de Chimie
UMR CNRS-ENS-UPMC 8640 PASTEUR
24 rue Lhomond, 75231 Paris (France)
E-mail: ludovic.jullien@ens.fr
[d] Dr. T. Le Saux, Prof. Dr. L. Jullien
UPMC Univ. Paris 6
4 Place Jussieu, 75232 Paris (France)
Results and discussion
[e] Dr. R. Labruꢀre
Current address:
Institut de Chimie Molꢁculaire et
des Matꢁriaux d’Orsay
UMR CNRS 8182, Universitꢁ Paris Sud
Orsay, 91405 Cedex (France)
E-mail: raphael.labruere@u-psud.fr
Experimental Design
Herein, we synthesized a series of substituted-phenol-, pyri-
dinol-, or pyrimidinol-containing caged spacers (cSXO;PF) that
could undergo ortho or para liberation through the cleavage
of a carbamate bond. This functional group has been shown
to be resistant to spontaneous hydrolysis (Scheme 2).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201400051.
1
&
&
Chem. Asian J. 2014, 00, 0 – 0
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
Ludovic Jullien et al.
the reporting amino-coumarin
through their corresponding
isocyanate (Scheme 3).
Photoactivated Self-Immolation
All of the experiments were
performed in MeCN/0.1m Brit-
ton–Robinson buffer (1:1, v/v).
From the absence of temporal
evolution of the absorption
spectrum on the 24 h timescale,
we first concluded that the
cSXO;PF molecules were stable in
the dark under our experimen-
tal conditions. To analyze the
self-immolation kinetics of the
photoactivated cSXO;PF inter-
mediates, we subsequently re-
corded the temporal evolution
of the fluorescence emission
from the reporting fluorophore
upon submitting solutions of
cSXO;PF to continuous 365 nm il-
lumination to both photorelease
and excite the reporting fluoro-
phore. Our kinetic experiments
were performed at various pH
values (pH 4–10) to study the
effect of the ionization state of
the phenol spacer and the pro-
tonation of ring-nitrogen atoms
on the rate of disassembly.
Given the relevance of self-im-
molative spacers for biological
applications, these assays were
performed at 310 K.
Scheme 1. In a quinone-methide-based self-immolative spacer, triggered cleavage spontaneously causes reor-
ganization of the electron density within the aromatic core, which leads to the release of a substrate through
1,4- or 1,6-elimination. The quinone methide subsequently re-aromatizes in the presence of water.
First, we analyzed the behav-
ior of phenyl reference com-
pounds cSOF and cSPF. From
To “lock” the precursor and permit its further photoacti-
vation, the phenol oxygen atom was protected with the
widely used 4,5-dimethoxy-2-nitrobenzyl caging group. Such
2-nitrobenzyl caging groups release phenols on the millisec-
ond timescale, thereby making it possible to analyze self-im-
molation processes slower than 10 ms.[9] To analyze the stoi-
chiometry and kinetics of self-immolation of the photoacti-
vated spacers, we chose 7-amino-4-(trifluoromethyl)coumar-
in (F), a latent fluorophore for which the emission is
quenched and blue-shifted in the caged precursor with re-
spect to its strongly fluorescent free state.[10]
the temporal evolution of the fluorescence emission on the
release of coumarin F upon the continuous illumination of
5 mm solutions of the samples at pH 8 (see the Supporting
Information, Figure 2S), we confirmed the quantitative pho-
torelease of F from cSOF and cSPF (see the Supporting In-
formation).
A cascade of reactions leads from the photoactivated pre-
cursors (cSXO;PF) to the final reporting fluorophore (F). As in
our previous work, we adopted the kinetic model shown in
Scheme 1, which contains steps that typically occur more
slowly than on the millisecond timescale: First, illumination
of the caged precursors yields phenol intermediates (rate
constant k1, associated with photoactivation), which subse-
quently disassemble into a benzenic core and a carbamic
ester, which further decomposes to afford carbon dioxide
and the reporting fluorophore in the third step (rate con-
stants k2 and k3 refer to the self-immolation and decarboxy-
Synthesis
Caged self-immolative spacers cSXO;PF were synthesized in
fair (30–46%; see the Supporting Information) yields by
coupling of the caged benzyl-alcohol modules (cSXO;POH) to
&
&
2
Chem. Asian J. 2014, 00, 0 – 0
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
Ludovic Jullien et al.
lation steps, respectively). Be-
cause the final step in this case
was faster than the temporal
resolution of our fluorescence-
acquisition setup,[6b,11] we could
analyze our data by using a sim-
plified kinetic model that fo-
cused on the first two steps
shown in Scheme 1 (see the
Supporting Information).
The Supporting Information,
Figure 2S, shows that our kinet-
ic model was appropriate for
obtaining
satisfactory
fits.
Hence, for cSPF, we could re-
trieve 1) the photoactivation
quantum yield,
fACTHNGUTERN(UNG 365 nm)=
160(ꢀ1)ꢃ10ꢁ4, which was in ac-
cordance with reported values
for uncaging with the ortho-ni-
trobenzyl group;[12] 2) the rela-
tive brightness of the corre-
sponding phenol intermediate
with respect to that of coumarin
F, 0.4, which was in line with
our previous results;[6] 3) k2 =
0.6
corresponding
(365 nm)=23.3
k2 =0.4
ACHTUNGTRENNUNG
values
(f-
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
have been reported previous-
ly.[6b] Table 1 summarizes the as-
sociated disassembly times, td =
1/k2, for cSOF and cSPF.
With these satisfactory results
in hand, we analyzed the self-
immolation kinetics of spacers
that contained a pyridine core
Scheme 2. Structures of caged self-immolative spacers cSXO;P F.
at pH 4 and 8. Figure 1 shows the temporal evolution of
fluorescence emission from the continuous illumination of
the three caged precursors cS1NF at 310 K. As anticipated,
Table 1. Disassembly times (td, in s) of the activated self-immolative
spacers at several pH values. MeCN/0.1m Britton–Robinson buffer (1:1,
v/v), T=310 K.
cSXO;P
F
pH 4
pH 8
2.5
1.7
pH 9.5
cSOF
cSP F
–
–
G
–
–
–
–
E
cS1ON
cS1PN
cS1ON0
cS2PN
F
F
F
F
2702
8330
N
149(ꢀ1)
294(ꢀ2)
1667(ꢀ5)
1515(ꢀ3)
–
–
7690
N
1429(ꢀ7)[b]
1250ACHTUNGTRENNUNG
cS2ON
F
–
1587(ꢀ2)
250(ꢀ2)[b]
[a]
204(ꢀ3)[c]
[a] No discernible release of
F under the experimental conditions;
[b] values extracted from the experiments at constant pH value;
[c] values extracted from the two-stage (pH 4 to 9.5) experiments.
Scheme 3. Synthetic pathway for the synthesis of cSXO;P F.
3
&
&
Chem. Asian J. 2014, 00, 0 – 0
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
Ludovic Jullien et al.
an increase in fluorescence emission was observed, which
was associated with the photorelease of the strongly fluores-
cent coumarin F, but also possibly of the intermediates
shown in Scheme 1. In particular, by using a solution of F
for the fluorescence-emission calibration, the final signal no-
tably compared with the one expected from the quantitative
photorelease of F (Figure 1). In that series, the caged com-
pounds again led to satisfactory fits and demonstrated ex-
tracted values of the photoactivation quantum yields, f-
The corresponding experiments and analyses were also
performed for the caged self-immolative spacers that con-
tained a pyrimidine core. At pH 8 and 9.5, illumination of
solutions of cS2PNF and cSO2NF led to the complete liberation
of F (Figure 2a,b). However, in contrast to the phenyl and
pyridine series, the release of coumarin F did not occur at
pH 4. Indeed, the emission spectra that was recorded after
the longest investigated times did not match the emission
spectrum of coumarin F, but rather was in agreement with
the expected blue-shifted ones of carbamic intermediates
S2PNF and S2ONF (Figure 2a,b). This assumption was confirmed
by HPLC with mass spectrometry detection: At pH 4, illu-
mination of cS2PNF led to the disappearance of the signal for
cS2PNF and the emergence of a peak when detection was per-
formed for the S2PNF molecular weight (see the Supporting
Information).
ACHTUNGTRENNUNG(365 nm), that were in line with literature expectations at
both investigated pH values;[12] k2 values were within the
range 1–60ꢃ10ꢁ4 sꢁ1, with the smallest values observed at
pH 4. The disassembly times, td =1/k2, are shown in Table 1.
Figure 1. Temporal evolution of the fluorescence emission, IFF ðtÞ (lem
=
500 nm), at pH 4 (a) and pH 8 (b) upon illumination at lex =365
(ꢀ25) nm
a solution of cS1ONF (4(ꢀ1) mm; gray), cS1PNF (3
ACHTUNGTRENNUNG
cS1ON0
F
(2.5
(ꢀ0.5) mm; black) at
a light intensity of 18.9ACHTUGTNENRNUG
Figure 2. Final emission spectra that result from photoactivation of caged
self-immolative spacers cSP2NF (a) and cS2ONF (b). The fluorescence emis-
10ꢁ9 Einsꢁ1. Markers denote experimental data; solid lines indicate fits to
[Eq. (33)]. From the signal calibration with the fluorophore (F), we ex-
tracted final coumarin concentrations (F1) of 2(ꢀ1), 1.5(ꢀ1), and
1.7(ꢀ1) mm (a) and 3(ꢀ1), 2(ꢀ1), and 1.5(ꢀ1) mm (b) for cS1ONF, cSP1NF,
and cS1ON0 F, respectively. From these fits, we also obtained: cS1ONF: f=78-
sion spectra (lex =365 nm) were recorded for 5
A
identical shape at pH 4 and 9.5) or 5(ꢀ1) mm solutions of cS F after
O;P
complete photoactivation at pH 4 (disks) and 9.5 (circles). Solvent:
MeCN/0.1m Britton–Robinson buffer (1:1, v/v), T=310 K. In line with
the deactivation of the nitrogen atom of the carbamic group, the final
emission spectrum at pH 4 is blue-shifted with respect to the F emission
G
R
ACHTUNGTRENNUNG
G
,
k2 =0.00012
(ꢀ0.00007) sꢁ1 (pH 4),
spectrum at the same pH value. In contrast, except for a shoulder that or-
G
(ꢀ0.5)ꢃ10ꢁ4, k2 =0.00066-
2N
O;P
iginated from the remaining intermediate, S F, the final emission spec-
R
ACHTUNGTRENNUNG
trum at pH 9.5 matched the F spectrum.
0.1m Britton–Robinson buffer (pH 4, pH 8; 1:1, v/v), T=310 K.
Chem. Asian J. 2014, 00, 0 – 0
&
&
4
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
Ludovic Jullien et al.
2N
O;P
2N
O;P
To further test the hypothesis that cS F could not self-
ed solutions of cS F at constant pH value. At pH 4, the ex-
immolate at pH 4, we devised a complementary illumination
experiment: We recorded the temporal evolution of the
fluorescence emission at pH 4 and subsequently increased
the pH value of the solution to 9.5 by adding concentrated
sodium hydroxide (Figure 3a,b). Before the pH increase, we
observed a monoexponential relaxation, in line with uncag-
ing leading to the formation of S2ON;PF. We extracted f-
perimental curves were successfully fitted by using a one-
step kinetic model (see the Supporting Information, Fig-
ure 5S), thereby yielding f(365 nm)=16.2
U
18.3
AHCTUNGTRENNUNG
cS2PNF and cS2ONF, respectively, which were in good agree-
ment with the values obtained in the pH-jump experiments.
In contrast, at pH 8 and 9.5, we employed the kinetic model
that involved the two first steps shown in Scheme 1 to fit the
experimental data. Hence, by using 0.5 for the relative
brightness of the corresponding phenol intermediate with
respect to that of coumarin F for both compounds, we ob-
A
N
ACHTUNGTRENNUNG
photoactivation quantum yields of cS2PNF and cSO2NF, respec-
tively, in line with the values observed above. After the pH
increase, we also observed a monoexponential relaxation,
which was expected for the decomposition of S2ON;PF into the
aromatic core, coumarin F, and carbon dioxide, assuming
that the decomposition of the carbamic intermediate was
fast.[6b,11] Moreover, the final emission spectrum was consis-
tent with the quantitative liberation of coumarin F.
tained f
(365 nm)=20.9
N
ACHTUNGTRENNUNG
and 0.7
A
ACHTUNGTRENNUNG
tivation quantum yields and k2 values at 310 K for cSP2N
F
and cS2ONF, respectively. In particular, the latter values were
in line with those extracted from the pH-jump experiments.
Table 1 lists the extracted disassembly times.
Following the above interpretation, we analyzed the tem-
poral evolution of the fluorescence emission from illuminat-
Discussion
This investigation was conceived to evaluate the relevance
of heteroaromatic rings for designing faster self-immolative
spacers that relied on quinone-methide-like elimination. The
rationale behind the choice of the pyridine and pyrimidine
rings was their significantly lower resonance energies (142
and 132 kJmolꢁ1) than 150 kJmolꢁ1 for the benzene ring.[16]
Indeed, disassembly of the phenol intermediates (SXO;PF) re-
quires loss of aromaticity and the corresponding energy bar-
rier should be lowered by decreasing the resonance energy
so as to accelerate spacer disassembly.
For data analysis, it is first essential to compare behaviors
at a same ionization state. Indeed, we have previously re-
ported that phenol ionization significantly decreases the dis-
assembly time in the phenyl series.[6b] The pKa values of the
phenol group in the phenyl cores of cSOF and cSPF are
above 10;[6b] thus, that only the phenol group is present at
the investigated pH value (pH 8). In 3-hydroxypyridine, the
reported pKa values that are associated to the ring-nitrogen
atom and the phenol group can be estimated to be about 5.0
and 8.5.[7,13] Correspondingly, experiments at pH 4, 8, and
9.5 would involve phenol pyridinium, phenol pyridine, and
phenolate pyridine cores, respectively. The pKa values for
ring-nitrogen atoms in the unsubstituted pyrimidinol core
have been reported to be about ꢁ7.0 and 1.5,[14] whereas the
phenol pKa values were estimated to be 7.0[13] and 8.6[14b,15]
in the bare pyrimidinol cores S2PN and S2ON, respectively. Thus,
the pyrimidinol core should always contain unprotonated ni-
trogen atoms under our experimental conditions. In con-
trast, it would involve the phenol group at pH 4 but the phe-
nolate from pH 8 and above.
Figure 3. Inhibition of total disassembly after photoactivation of caged
self-immolative spacers cSP2NF (a) and cS2ONF (b) at pH 4. Temporal evolu-
tion of the fluorescence emission at lem =500 nm upon illumination of
First, we will focus on the effect of the structure of the ar-
omatic ring on the disassembly time of the neutral self-im-
molative spacer. Thus, we will compare the data for the
phenyl and pyridine series at pH 8 with the data for the pyr-
imidine series at pH 4. Self-immolative spacers with a ben-
zene core displayed faster release than those with a pyridine
a) 5(ꢀ1) mm·cS2PN
F
or b) 2.5
G
F solutions at lex =365-
(ꢀ25) nm, 18.9ꢃ10ꢁ9 Einsꢁ1. After reaching a steady-state at pH 4, the
pH value was raised to pH 9.5 by the addition of NaOH. Markers denote
the experimental data; solid lines indicate fits with [Eq. (34)] at pH 4 and
with [Eq. (34)] at pH 9.5 (see the Supporting Information). Solvent:
MeCN/0.1m Britton–Robinson buffer (1:1, v/v), T=310 K.
5
&
&
Chem. Asian J. 2014, 00, 0 – 0
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
Ludovic Jullien et al.
core; the pyrimidinic spacers did not self-immolate on the
investigated timescale under such conditions. Beyond the
considerations based on resonance energy (which would
have led us to expect the reverse order), we hypothesize
that this result originates from differences in the electron
density on the carbon atom bearing the benzylic substituent
that is involved in the elimination step. The electron-with-
drawing nitrogen atom is expected to decrease this density
in cS1ONF and cSO1N0 F with respect to cSOF and in cS1PNF with
respect to cSPF. This feature should lead to longer disassem-
bly times. In particular, such an explanation could be
strengthened by the observation that 1) nitrogen protona-
tion leads to further slowdown of the disassembly in the pyr-
idine series at pH 8; 2) the presence of two nitrogen atoms
on the aromatic ring effectively suppresses disassembly, as
observed in the pyrimidine series at pH 4.
observed pH-dependence of the disassembly time for this
series of self-immolative spacers could be significant for var-
ious applications, ranging from prodrugs to the programmed
degradation of materials.
Experimental Section
General Synthesis
Commercially available chemicals were used without further purification.
Anhydrous solvents were freshly distilled prior to use. Low-actinic glass-
ware was used for all experiments with compounds that contained nitro-
veratryl moieties. Column chromatography was performed on Merck
silica gel 60 (0.040–0.063 mm). Analytical and thin layer chromatography
(TLC) were performed on plates that were precoated with Merck silica
gel 60 F-254; detection was performed by using UV light (l=254 nm).
NMR spectra were recorded on an AC Bruker spectrometer at 300 MHz
1
(for H nuclei) and 75 MHz (for 13C nuclei). Coupling constants (J) are in
Next, we will analyze the effect of spacer ionization on
the disassembly rate. As we previously observed for the
phenyl series,[6b] phenol ionization significantly accelerated
disassembly for the pyrimidine series; on increasing the pH
value from 8 to 9.5, we typically observed a decrease in the
disassembly time by a factor of 5. This observation con-
firmed the trend that electron-rich aryl cores accelerated
self-immolation.[5,6b–c]
Hz. HPLC analysis and purification of the final caged species were per-
formed on a Waters system with a Wdelta 600 pump and a PDA 996 UV
detector at l=245 nm (analytical HPLC: X-Terra Waters MS C18 5 mm
(particle size) column, 150 mm (length)ꢃ4.6 mm (diameter), flow rate=
1 mLminꢁ1; preparative HPLC: X-Terra Waters Prep MS C18 5 mm (par-
ticle size) column, 150 mm (length)ꢃ19 mm (diameter), flow rate=
10 mLminꢁ1; elution with MeCN/water mixtures). For the synthesis of
the intermediate benzylic alcohols, cSXO;P OH, see the Supporting Informa-
tion.
Finally, we will examine the role of ortho/para substitution
on the elimination rate. In agreement with previous observa-
tions,[17] ortho- and para-quinone-methide elimination oc-
curred on similar timescales (cf. cSOF and cSPF). We arrived
at the same conclusion regarding pyridinone-methide elimi-
nation (cf. cS1ONF and cSP1NF). However, one should note that
we observed that the disassembly times of two ortho-substi-
7-Isocyanato-4-(trifluoromethyl)-2H-chromen-2-one (F’)
To a suspension of 7-amino-4-(trifluoromethyl)-2H-chromen-2-one (F,
100 mg, 0.4 mmol) under an Ar atmosphere was dropwise added a 20%
solution of phosgene in toluene (4 mL, 8 mmol) with a syringe. The reac-
tion mixture was heated at reflux for 15 h under an Ar atmosphere,
during which time a white solid precipitated out of the solution. After
cooling, Ar gas was bubbled through the solution for 15 min to remove
any unreacted phosgene. Then, the reaction mixture was evaporated
under reduced pressure to afford the product as a white solid (100 mg,
90% yield), which was used directly in the next step.
tuted derivatives differed by a factor of about 10 (cf. cSO1N
F
1N
and cS F), which revealed the importance of the position
O0
of the nitrogen atom on the aromatic core. Unlike cS1ONF,
cS1ON0 F showed a relatively slow release that was not influ-
enced by protonation. In that case, the exocyclic methylene
group that contained the leaving group was at the para posi-
tion relative to the ring-nitrogen (Scheme 2). Thus, we hy-
pothesized that the neutral pyridinic nitrogen moiety of
cS1ONF at the ortho position relative to the benzylic position
could assist the liberation of the reporter because the nitro-
gen lone pair could attack the electrophilic carbonyl group
of the carbamate linkage.
General Procedure for the Preparation of Caged Self-Immolative Spacers
To a solution of the desired benzyl alcohol intermediate (cSXO;P OH,
0.03 mmol) in anhydrous THF (5 mL) at RT under an Ar atmosphere
was dropwise added
a suspension of sodium tert-butoxide (4 mg,
0.045 mmol) in anhydrous THF (1 mL). The mixture was cooled to 08C
and a solution of F’ (8 mg, 0.03 mmol) was added dropwise. Then, the
mixture was stirred at RT until the reaction had gone to completion. The
resulting suspension was diluted with CH2Cl2 (10 mL) and water (10 mL)
was added. The organic phase was separated, dried over MgSO4, and
evaporated under reduced pressure. The crude product was recovered by
either preparative HPLC or by precipitation with acetone (2 mL).
cSP F: Prepared by preparative HPLC (A/B, 80:20), yellow solid (8 mg,
46% yield); 1H NMR (300 MHz, CDCl3, 258C): d=7.77 (s, 1H), 7.64 (d,
J=2 Hz, 1H), 7.63 (d, J=9 Hz, 1H), 7.37 (d, J=8 Hz, 2H), 7.33 (s, 1H),
7.32 (dd, J=9, 2 Hz, 1H), 7.03 (d, J=8 Hz, 1H), 6.67 (s, 1H), 5.50 (s,
2H), 5.19 (s, 2H), 3.97 (s, 3H), 3.96 ppm (s, 3H); MS (ES+): m/z: 597.1
[M+Na]+; HRMS (ES+): m/z calcd for C27H21F3N2NaO9 [M+Na]+:
597.1097, found 597.1091.
cS1ONF: Prepared by precipitation from acetone; yellow solid (5 mg, 30%
yield); 1H NMR (300 MHz, [D7]DMF, 258C): d=10.54 (s, 1H), 8.23 (s,
1H), 7.75–7.69 (m, 4H), 7.54–7.45 (m, 3H), 6.84 (s, 1H), 5.58 (s, 2H),
5.45 (s, 2H), 3.98 (s, 3H), 3.93 ppm (s, 3H); MS (ES+): m/z 576.14
[M+H]+; HRMS (ES+): m/z calcd for C26H21F3N3O9: 576.1230 [M+H]+;
found: 576.1224.
Conclusions
Pyridine- and pyrimidine-based self-immolative spacer
groups have been shown to disassemble through 1,4- and
1,6-elimination pathways. A photoactivation procedure has
been used to show that these self-immolative spacers exhib-
ited slower kinetics for spacer disassembly than their corre-
sponding benzenic analogues. These results are relevant for
deriving quantitative structure–property relationships. In
particular, they suggest that electron density plays a more
important role than resonance energy in governing the dis-
assembly rate of an aromatic ring spacer. Interestingly, the
cS1PNF: Prepared by precipitation from acetone; yellow solid (7 mg, 40%
yield); 1H NMR (300 MHz, [D7]DMF, 258C): d=10.54 (s, 1H), 8.43 (s,
1H), 7.80–7.70 (m, 4H), 7.60–7.40 (m, 3H), 6.87 (s, 1H), 5.57 (s, 2H),
&
&
6
Chem. Asian J. 2014, 00, 0 – 0
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
Ludovic Jullien et al.
5.26 (s, 2H), 3.98 ppm (s, 6H); MS (ES+): m/z: 576.2 [M+H]+; HRMS
(ES+): m/z calcd for C26H21F3N3O9: 576.1230 [M+H]+; found: 576.1224.
nahary, S. Michel, M. Koch, F. Tillequin, M. Gerken, J. Czech, R.
Straub, K. Bosslet, J. Med. Chem. 1998, 41, 3572–3581; b) B. E.
Toki, C. G. Cerveny, A. F. Wahl, P. D. Senter, J. Org. Chem. 2002,
67, 1866–1872; c) M. Shamis, H. N. Lode, D. Shabat, J. Am. Chem.
Soc. 2004, 126, 1726–1731; d) R. J. Amir, M. Popkov, R. A. Lerner,
C. F. Barbas III., D. Shabat, Angew. Chem. Int. Ed. 2005, 44, 4378–
4381; Angew. Chem. 2005, 117, 4452–4455; e) A. Gopin, S. Ebner,
B. Attali, D. Shabat, Bioconjugate Chem. 2006, 17, 1432–1440; f) Y.
Zhang, Q. Yin, L. Yin, L. Ma, L. Tang, J. Cheng, Angew. Chem. Int.
Ed. 2013, 52, 6435–6439; Angew. Chem. 2013, 125, 6563–6567.
[4] a) N. H. Ho, R. Weissleder, C. H. Tung, ChemBioChem 2007, 8,
560–566; b) T. Chauvin, P. Durand, M. Bernier, H. Meudal, B. T.
Doan, F. Noury, B. Badet, J. C. Beloeil, E. Tꢄth, Angew. Chem. Int.
Ed. 2008, 47, 4370–4372; Angew. Chem. 2008, 120, 4442–4444; c) E.
Sella, D. Shabat, J. Am. Chem. Soc. 2009, 131, 9934–9936; d) G. C.
van de Bittner, E. A. Dubikovskaya, C. R. Bertozzi, C. J. Chang,
Proc. Natl. Acad. Sci. USA 2010, 107, 21316–21321; e) M. S. Baker,
S. T. Phillips, J. Am. Chem. Soc. 2011, 133, 5170–5173; f) G. G.
Lewis, M. J. Ditucci, S. T. Phillips, Angew. Chem. Int. Ed. 2012, 51,
12707–12710; Angew. Chem. 2012, 124, 12879–12882; g) H. Xie, J.
Mire, Y. Kong, M. Chang, H. A. Hassounah, C. N. Thornton, J. C.
Sacchettini, J. D. Cirillo, J. Rao, Nat. Chem. 2012, 4, 802–809; h) O.
Redy, D. Shabat, J. Controlled Release 2012, 164, 276–282; i) F.
Touti, P. Maurin, J. Hasserodt, Angew. Chem. Int. Ed. 2013, 52,
4654–4658; Angew. Chem. 2013, 125, 4752–4756; j) J. Sloniec, U.
Resch-Genger, A. Hennig, J. Phys. Chem. B 2013, 117, 14336–
14344.
cS1ON0 F: Prepared by precipitation from acetone; yellow solid (5 mg, 30%
yield); 1H NMR (300 MHz, [D7]DMF, 258C): d=10.59 (s, 1H), 8.57 (s,
1H), 8.33 (s, 1H), 7.77–7.70 (m, 3H), 7.57–7.51 (m, 3H), 6.87 (s, 1H),
5.68 (s, 2H), 5.40 (s, 2H), 4.00 (s, 3H), 3.96 ppm (s, 3H); MS (ES+): m/
z: 576.09 [M+H]+; HRMS (ES+): m/z calcd for C26H21F3N3O9: 576.1230
[M+H]+; found: 576.1224.
cS2PNF: Prepared by preparative HPLC (A/B, 70:30); yellow solid (5 mg,
30% yield); 1H NMR (300 MHz, CDCl3, 258C): d=8.52 (s, 2H), 7.78 (s,
1H), 7.67 (d, J=2 Hz, 1H), 7.63 (d, J=9 Hz, 1H), 7.32 (dd, J=9, 2 Hz,
1H), 7.23 (s, 1H), 6.67 (s, 1H), 5.50 (s, 2H), 5.18 (s, 2H), 3.97 (s, 3H),
3.96 ppm (s, 3H); MS (ES+): m/z: 577.14 [M+H]+; HRMS (ES+): m/z
calcd for C25H20F3N4O9: 577.1182 [M+H]+; found: 577.1177.
cS2ONF: Prepared by preparative HPLC (A/B, 70:30); yellow solid (6 mg,
35% yield); 1H NMR (300 MHz, CDCl3, 258C): d=8.41 (s, 1H), 7.69 (s,
1H), 7.67 (d, J=2 Hz 1H), 7.63 (d, J=9 Hz, 1H), 7.32 (dd, J=9, 2 Hz,
1H), 7.11 (s, 1H), 6.67 (s, 1H), 5.79 (s, 2H), 5.16 (s, 2H), 3.99 (s, 3H),
3.96 ppm (s, 3H); MS (ES+): m/z: 591.1 [M+H]+; HRMS (ES+): m/z
calcd for C26H22F3N4O9: 591.1339 [M+H]+; found: 591.1333.
Analytical Solutions
All of the experiments were performed in MeCN/0.1m Britton–Robinson
buffer (1:1, v/v).[18] All of the solutions were prepared by using water that
was purified through a Direct-Q 5 (Millipore, Billerica, MA) system.
UV/Vis Absorption
[5] a) M. P. Hay, B. M. Sykes, W. A. Denny, C. J. OꢅConnor, J. Chem.
Soc. Perkin Trans. 1 1999, 2759–2770; b) E. E. Weinert, R. Dondi, S.
Colloredo-Melz, K. N. Frankenfield, C. H. Mitchell, M. Freccero,
S. E. Rokita, J. Am. Chem. Soc. 2006, 128, 11940–11947; c) K. M.
Schmid, L. Jensen, S. T. Phillips, J. Org. Chem. 2012, 77, 4363–4374;
d) J. S. Robbins, K. M. Schmid, S. T. Phillips, J. Org. Chem. 2013, 78,
3159–3169; e) K. M. Schmid, S. T. Phillips, J. Phys. Org. Chem.
2013, 26, 608–610.
[6] a) R. Labruꢀre, A. Alouane, T. Le Saux, I. Aujard, P. Pelupessy, A.
Gautier, S. Dubruille, F. Schmidt, L. Jullien, Angew. Chem. Int. Ed.
2012, 51, 9344–9347; Angew. Chem. 2012, 124, 9478–9481; b) A.
Alouane, R. Labruꢀre, T. Le Saux, I. Aujard, S. Dubruille, F.
Schmidt, L. Jullien, Chem. Eur. J. 2013, 19, 11717–11724; c) Record
of records, Chem. Rec. 2013, 13, 456–479.
UV/Vis absorption spectra were recorded in quartz cuvettes (1 cmꢃ1 cm,
Hellma) on a diode-array UV/Vis spectrophotometer (Cary 300, Agilent,
Thermo Scientific) at 298 K. Molar absorption coefficients were extracted
whilst checking the validity of the Beer–Lambert law.
Steady-State Fluorescence Emission
Corrected fluorescence spectra upon one-photon excitation were record-
ed on a Photon Technology International QuantaMaster QM-1 spectro-
fluorimeter (PTI, Monmouth Junction, NJ) that was equipped with a Pelt-
ier cell holder (TLC50, Quantum Northwest, Shoreline, WA). Solutions
for the fluorescence measurements were adjusted to concentrations such
that the absorption maximum was about 0.15 at the excitation wave-
length.
[7] R. Perry-Feigenbaum, P. S. Baran, D. Shabat, Org. Biomol. Chem.
2009, 7, 4825–4828.
Irradiation Experiments
[8] Heterocyclic self-immolative linkers and conjugates, PCT WO 2005/
082023A2 (09.09.2005); heterocyclic self-immolative linkers and
conjugates, US 2009/0041791A1 (12.02.2009); heterocyclic self-im-
molative linkers and conjugates, US 7,754,681 B2 (13.07.2010).
[9] R. Jasuja, J. Keyoung, G. P. Reid, D. R. Trentham, S. Khan, Biophys.
J. 1999, 76, 1706–1719.
[10] H.-J. Jin, J. Lu, X. Wu, Bioorg. Med. Chem. 2012, 20, 3465–3469.
[11] a) M. Caplow, J. Am. Chem. Soc. 1968, 90, 6795–6803; b) S. L. John-
son, D. L. Morrison, J. Am. Chem. Soc. 1972, 94, 1323–1334.
[12] I. Aujard, C. Benbrahim, M. Gouget, O. Ruel, J.-B. Baudin, P.
Neveu, L. Jullien, Chem. Eur. J. 2006, 12, 6865–6879.
One-photon-irradiation experiments were performed on the spectro-
fluorimeter with or without pH variation. Irradiations were performed by
using a filtered 75 W Xe lamp at several slit widths on 400 mL samples in
quartz fluorescence cuvettes (0.2 cmꢃ1 cm, Hellma) under constant stir-
ring. The incident-light intensities were calibrated with a-(4-dimethylami-
nophenyl)-N-phenylnitrone actinometer according to a literature proce-
dure.[19] Typical integral incident-light intensities were within the range
1ꢃ10ꢁ9 Einsꢁ1
.
Acknowledgements
[13] S. J. Nara, L. Valgimigli, G. F. Pedulli, D. A. Pratt, J. Am. Chem.
Soc. 2010, 132, 863–872.
This work was supported by France Bioimaging and by the ANR Blanc
2010 (Kituse).
[14] a) S. F. Mason, J. Chem. Soc. 1958, 674–685; b) D. J. Brown, Pyrimi-
dines and their Benzo Derivatives in Comprehensive Heterocyclic
Chemistry (Eds.: A. R. Katritzky, C. W. Rees), Pergamon Press,
Oxford, 1984.
[15] D. J. Brown, L. N. Short, J. Chem. Soc. 1953, 331–337.
[16] A. R. Katritzky, A. F. Pozharskii, Handbook of Heterocyclic Chemis-
try, 2nd ed., Pergamon Press, Oxford, 2000.
[17] H. Y. Lee, X. Jiang, D. Lee, Org. Lett. 2009, 11, 2065–2068.
[18] C. Frugoni, Gazz. Chim. Ital. 1957, 87, 403–407.
[1] P. L. Carl, P. K. Chakravarty, J. A. Katzenellenbogen, J. Med. Chem.
1981, 24, 479–480.
[2] a) C. A. Blencowe, A. T. Russell, F. Greco, W. Hayes, D. W.
Thornthwaite, Polym. Chem. 2011, 2, 773–790; b) A. D. Wong,
M. A. DeWit, E. R. Gillies, Adv. Drug Delivery Rev. 2012, 64, 1031–
1045; c) R. E. Wang, F. Costanza, Y. Niu, H. Wu, Y. Hu, W. Hang,
Y. Sun, J. Cai, J. Controlled Release 2012, 159, 154–163.
[19] P. F. Wang, L. Jullien, B. Valeur, J.-S. Filhol, J. Canceill, J.-M. Lehn,
New J. Chem. 1996, 20, 895–907.
[3] a) J. C. Florent, X. Dong, G. Gaudel, S. Mitaku, C. Monneret, J. P.
Gesson, J. C. Jacquesy, M. Mondon, B. Renoux, S. Andrianomenja-
Received: January 9, 2014
Published online: && &&, 0000
7
&
&
Chem. Asian J. 2014, 00, 0 – 0
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

FULL PAPER
Photoactivation
Ahmed Alouane, Raphaꢀl Labruꢁre,*
Katherine J. Silvestre, Thomas Le Saux,
Frꢂdꢂric Schmidt,*
Ludovic Jullien*
&&&& — &&&&
Disassembly Kinetics of Quinone-
Methide-Based Self-Immolative
Spacers that Contain Aromatic
Nitrogen Heterocycles
Burn rubber: Kinetic analysis of the
pH-dependent disassembly of self-
immolative spacers that contain aro-
matic nitrogen heterocycles was per-
formed. Electron-poor pyrimidine
cores exhibited the longest disassembly
times. This study confirms the trend
that electron-rich aryl cores accelerate
self-immolation.
&
&
8
Chem. Asian J. 2014, 00, 0 – 0
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!