Angewandte
Chemie
untreated cells. The dehydroxy estrone 2 was used as
a negative control and indeed showed only a basal level of
firefly luciferase expression. Moreover, luciferase induction
by both estrone (1) and dehydroxy estrone (2) was not
significantly affected by addition of H O .
activity for the boronate estrone 4 and a fivefold increase for
the diboronate estrone 5. The lower activation from 5 is
presumably the result of an incomplete conversion into 1 by
intracellularly generated H O . Further increase in the
2
2
concentration of EGF added to the cell culture media led to
a linear increase in luciferase signal, because cellular H O
2
2
The estrone derivative 3, with a boronic acid ester at the 3-
position, showed a slightly higher background level of
luciferase expression than the negative control 2, but is still
mostly inactive compared to estrone (1). Upon addition of
H O , the boronate group of 3 is oxidized and the resulting
2
2
[25]
production increases with increasing EGF exposure (see
Supporting Information).
Thus, the developed H O reporter provides a substan-
2
2
tially greater dynamic range than previously reported H O
2
2
2
2
estrone is able to bind to the ER inducing a fivefold increase
in luciferase expression, a level that approximates that of
treatment with 1. The estrone boronated at the 17-position
sensors. The higher signal-to-background ratio (up to 33-fold)
of this system may be the result of two linked catalytic
processes: gene transcription induced by H O and subse-
2
2
(
4), displayed a lower background level of luciferase expres-
quent bioluminescence through conversion of luciferin into
oxyluciferin catalyzed by luciferase. Intracellular detection of
sion than 3 and exposure to H O2 resulted in a 28-fold
2
increase in gene expression. Even though the level of
luciferase expression was only 48% of the natural estrone
H O2 through fluorescence measurements have been
2
[
13]
reported using boronate fluorophores and by a genetically
encoded protein that emits fluorescence when oxidized by
(1), the signal-to-background ratio was excellent. Similar
[
26,27]
results were found with the diboronated estrone 5, which
displayed a low background level of luciferase activity before
exposure to H O and an eightfold increase in gene expression
H O .
However, only two- to sixfold changes in fluores-
2
2
cence were measured with these systems. Importantly, since
any coding or non-coding genetic sequence can be cloned
downstream of the UAS, the developed system reported
herein can also be used as a transcriptional switch for the
activation of any gene of interest by H O .
2
2
after H O addition.
2
2
The ability of the sensor to detect endogenously produced
H O in mammalian cells was tested. Cells produce H O
2
2
2
2
2
2
when stimulated with external cytokines such as transforming
One challenge in creating a cellular H O reporter is to
2 2
growth factor-b1, interleukin-1, or epidermal growth factor
ensure that it is sensitive and selective to H O over other
2 2
[24]
(
EGF). Here, EGF was used to stimulate H O production
competing cellular ROS, such as hydroxyl radicals and
2
2
[
27,28]
in A431 cells. These assays were conducted as described
above but instead of adding H O to the media, the cells were
hypochlorite ions.
In this regard, the selectivity of the
gene activation system was tested in cell culture by treatment
2
2
À1
treated with EGF (1 mgmL ). As seen in Figure 2, the
with several ROS: H O , tert-butyl hydroperoxide (TBHP),
2
2
À
intracellular generation of H O can be detected as efficiently
hypochlorite (OCl ), hydroxyl radical (COH, generated from
H O and FeSO ), and tert-butoxy radical (COtBu, generated
2
2
as when it is added externally. In comparison to the positive
and negative controls (1 and 2, respectively), the boronate
estrone 3 showed only a moderate level of background
luciferase expression before addition of EGF and a fivefold
increase in luciferase after EGF addition. Importantly, the
boronate estrone 4 and the diboronate estrone 5 showed
further reduced levels of background activity, and addition of
EGF resulted in a dramatic 33-fold increase in luciferase
2
2
4
from TBHP and FeSO ). The boronate estrone 3 was used
4
because it showed the highest recovery of gene expression
after addition of H O , relative to native estrone (see
2
2
Figure 1). No reporter gene expression was detected in
response to any ROS except H O2 (Figure 3). Following
2
addition of H O to cells treated with the boronate estrone 3,
2
2
Figure 3. The boronated estrone derivative 3 is selective for H O .
2
2
Figure 2. Intracellular detection of hydrogen peroxide. A431 cells
A431 cells (10000) were transfected with pBind-ERa and pGL4.35,
treated with estrone (1, 50 nm), boronate estrone 3 (50 nm), or DMSO
only, then exposed to the ROS (100 nm) shown under the graph, and
a luciferase assay was performed. Reporter gene activity is normalized
to treatment with estrone (1). All experiments were performed in
triplicate and error bars represent the standard deviation.
(
10000) were transfected with pBind-ERa and pGL4.35, and treated
À1
with estrone analogues 2–5 (50 nm) and EGF (1 mgmL ). A firefly
luciferase assay was performed and reporter gene activity was normal-
ized to exposure to estrone (1). All experiments were performed in
triplicate and error bars represent the standard deviation.
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!