Synthesis of a photocaged tamoxifen for light-dependent activation of Cre-ER recombinase-driven gene modification

Article abstract of DOI:10.1039/c3cc42179a

We report the design of a water-soluble, quaternized tamoxifen photoprobe and demonstrate its application in light-controlled induction of green fluorescent protein expression via a Cre-ER recombinase system. The Royal Society of Chemistry.

Full text of DOI:10.1039/c3cc42179a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Synthesis of a photocaged tamoxifen for light-dependent
activation of Cre-ER recombinase-driven gene
modification†
Matthew A. Inlay,*a Veronica Choe,b Sophia Bharathi,bc Nathaniel B. Fernhoff,a
James R. Baker, Jr.,bc Irving L. Weissmana and Seok Ki Choi*bc
Cite this: Chem. Commun., 2013,
49, 4971
Received 26th March 2013,
Accepted 10th April 2013
DOI: 10.1039/c3cc42179a
www.rsc.org/chemcomm
We report the design of a water-soluble, quaternized tamoxifen
photoprobe and demonstrate its application in light-controlled
induction of green fluorescent protein expression via a Cre-ER
recombinase system.
In vivo lineage tracing has been revolutionized by the Cre–loxP system,
whereby tissue-specific promoter activation of the Cre-recombinase
leads to permanent activation of a reporter gene in a cell and all its
downstream progeny.1 Fusion of the Cre enzyme with the estrogen
receptor (ER), which prevents nuclear localization of the Cre-ER
protein unless bound by an estrogen analog (e.g. tamoxifen), allows
an additional level of temporal control of reporter activation.2–4
Tamoxifen (TAM) and its active metabolite form 4-hydroxytamoxifen
(4-OHT) have brief half-lives upon injection into mice, thus restricting
reporter induction to a 24–48 h window when tamoxifen is active.5
Despite the utility of this system, tamoxifen diffuses rapidly in vivo,
preventing precise spatial activation of a specific location or region
within an organism (e.g. femur vs. tibia; left vs. right).
Fig. 1 Schematic for the release of tamoxifen from its photocaged form by UV
light, and control of Cre-ER recombinase-mediated GFP expression in mouse
embryonic fibroblast (MEF) cells.
control of small molecule mediated biological processes such
as ion channel gating,7,8 protein–protein interactions,9,10 protein
phosphorylation,11 transcription,12 and payload release in targeted
drug delivery.13–16 In addition, tamoxifen-related molecular
photoprobes have previously been utilized for studying gene
expression with Cre-ER fusion proteins.2,17,18
In designing our photocaged tamoxifen molecule, we performed
the N-quaternization reaction of tamoxifen at its tertiary amine
nitrogen with an o-nitrobenzyl (ONB)-based alkylating mole-
cule15,19,20 2 (Scheme 1). This reaction afforded TAM-ONB 3: ESI
HRMS (m/z): calcd for C43H53N4O8 753.3858 [M]+, found 753.3856;
purity Z99% (anal. HPLC); UV-vis absorption peaks at 350,
270 nm.1 After deprotection of its N-Boc, TAM-ONB 4 was obtained
as a HCl salt (see ESI† for details). This design is notable in two
aspects. First, it allows us to photocage tamoxifen directly without
In this study, we investigated a photochemical approach6 for the
controlled expression of mutant genes using a mammalian Cre-ER
recombinase in combination with light-triggered release of
tamoxifen, an estrogen receptor antagonist. The mechanism for
the drug release is based on the design of photocaged tamoxifen in
which the tamoxifen molecule is temporarily inactivated through
its covalent attachment to a photocleavable ortho-nitrobenzyl (ONB)
group (Fig. 1). Upon exposure to UV irradiation, the ONB group
of the caged molecule (TAM-ONB) is irreversibly cleaved, leading to
release of free tamoxifen. This photocaging strategy has been
previously employed in other systems for the precise spatio-temporal
a Institute for Stem Cell Biology and Regenerative Medicine, Stanford University
School of Medicine, Stanford, CA 94305, USA. E-mail: minlay@stanford.edu;
Fax: +1 650 498-6255; Tel: +1 650 725-5808
b Michigan Nanotechnology Institute for Medicine and Biological Sciences,
University of Michigan, Ann Arbor, MI 48109, USA
c Department of Internal Medicine, University of Michigan, Ann Arbor, MI 48109,
USA. E-mail: skchoi@umich.edu; Fax: +1 734 615-0621; Tel: +1 734 615-0618
† Electronic supplementary information (ESI) available: Experimental details for
synthesis and characterization of 3 and 4; details for photocleavage experiments
and cell studies (PDF). See DOI: 10.1039/c3cc42179a
Scheme 1 Synthesis of photocaged tamoxifen. Reagents and conditions: (i) ethyl
bromoacetate, K2CO3, DMF, RT; (ii) NaOH, THF, MeOH, H2O, RT; (iii) conc. HNO3, AcOH,
0 1C to RT; (iv) N-Boc-1,2-diaminoethane, DCC, DMAP, DMF, 0 1C to RT; (v) NaBH4, THF,
MeOH, 0 1C to RT; (vi) methanesulfonyl chloride (MsCl), Et3N, CH2Cl2, 0 1C, 88%;
(vii) tamoxifen, acetone, 37 1C, 3 days, 27%; (viii) 6 M HCl, MeOH, rt, 6 h, 96%.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 4971--4973 4971

View Article Online
Communication
ChemComm
making any structural modifications or derivatizations to the drug indicate that TAM can be rendered inactive by photocaging with
molecule. It therefore differs from earlier methods reported ONB, and this caging can be released by UV exposure and recombi-
for preparing tamoxifen photoprobes which are made using nation activation.
tamoxifen derivatives or analogs.17,18 Second, the aqueous solubility
We next compared the length of UV exposure to recombination
of TAM-ONB 4 (Z10 mg mLÀ1) was greatly improved compared to efficiency (Fig. 3a). We exposed either regular tamoxifen (TAM) or
practically insoluble tamoxifen (E0.0002 mg mLÀ1 water21). We TAM-ONB 4 to UV light for 1 min up to 1 h, then treated UT MEFs
believe that its quaternary salt form improves the solubility as with the uncaged products at 8 mM. One min UV exposure was
does the hydrophilic amine-terminated side chain tethered to sufficient to activate low level reporter expression, and 5 min was
the ONB group.
sufficient for near peak recombination efficiency. There was a mild
Generation of a tamoxifen-inducible reporter MEF line: to validate increase in recombination efficiency between 5 min and 20 min,
TAM-ONB 4, we generated tamoxifen-inducible reporter mouse though beyond 20 min, there was no improvement. TAM induction
embryonic fibroblasts (MEFs) by crossing a transgenic mouse line remained around 35% regardless of the length of UV exposure.
that expresses Cre-ERT2 under the control of the constitutively- Thus, 5 min UV exposure is sufficient for robust uncaging of 4 and
expressed human ubiquitin C (UBC) promoter22 to a transgenic reporter activation.
reporter line with a reporter cassette (mTmG) inserted into the
Time-dependent control of GFP expression: we next tested different
constitutive Rosa26 locus.23 This cassette initially expresses the red concentrations of TAM-ONB 4 to optimize the dose and UV exposure
fluorescent protein TdTomato (Tomato), but will permanently time for TAM release (Fig. 3b). We found that the concentration of 4
recombine and switch expression to green fluorescent protein had a significant impact on recombination frequency, with 32 mM
(GFP) when Cre-recombinase is active. We first tested UBC-Cre- and 64 mM treatments approaching the recombination efficiency of
ERT2 Â mT/mG (UT) MEFs for reporter induction upon tamoxifen unmodified tamoxifen (35%). With each concentration, there was
exposure (Fig. 2, Fig. S5, ESI†). One day after 8 mM TAM treatment, only a modest increase in recombination efficiency by increasing the
we examined MEFs using flow cytometry for GFP and Tomato UV exposure from 5 min to 20 min. We conclude the ideal
expression. After treatment, 35% of UT MEFs had recombined and concentration to be 16–32 mM with 5–20 min UV exposure.
activated GFP expression, although our timepoint was too early to
Pre-treatment with TAM-ONB 4 prior to UV uncaging: In order
observe Tomato fluorescence diminish after its excision. Thus UT to allow precise labeling in vivo, TAM-ONB 4 will have to be
MEFs are a highly sensitive tool for detecting TAM-dependent loaded efficiently into cells in its caged, inactive state prior to
recombination due to its extremely low levels of spontaneous UV release. We next tested the loading capacity of 4 in UT-MEFs
recombination.
prior to UV uncaging. Preliminary experiments revealed that 1 h
Validation of TAM-ONB 4 release: we next tested whether the exposure to TAM was sufficient for near maximum recombina-
photocaged tamoxifen molecule (TAM-ONB, 4) is released by a tion frequency (Fig. S6, ESI†). However, the presence of serum
photochemical mechanism, causing Cre activation and GFP
induction in UT MEFs (Fig. 2b). Here, the MEF cells were treated
with 4 (8 mM) that was either unexposed, or exposed for 20 min to
UV-A light with a mean wavelength of 365 nm which is strongly
absorbed by the ONB 1 (lmax = 340 nm, e340 = 2750 MÀ1 cmÀ1).1 24 h
after adding TAM-ONB 4, MEFs treated with unexposed 4 showed
less than 1% GFP induction. In contrast, 20 min UV exposure was
sufficient to activate recombination and induce GFP expression in
21% of UT MEFs. While this is somewhat less than the 35% of GFP+
UT MEFs we observed with 24 h of TAM treatment, our data clearly
Fig. 2 Tamoxifen (TAM)-controlled GFP expression in mouse embryonic fibro-
blast (MEF) cells that constitutively express UBC-promoter Cre-ER mTmG fusion
protein (UT MEFs). Flow cytometry analysis of MEF cells treated with free TAM
(À, +) (a), or with TAM-ONB 4 pre-exposed to 0 or 20 min of UV irradiation at
365 nm (b). Fluorescent microscopic images of the same cells treated as above
are shown below respectively: green fluorescence protein (GFP), and Tomato
fluorescence protein (Tomato, inset).
Fig. 3 Time-dependent, UV light-controlled activation of TAM-ONB 4 and
induction of GFP-positive MEF cells. (a and b) Flow cytometry analysis of MEF
cells exposed to UV light as a function of time after treatment with either free
TAM or TAM-ONB 4. Plots for induction of GFP-positive cells following UV
exposure after treatment with TAM-ONB 4 at a single concentration of 14 mM
(a, bottom), or at four different concentrations (b). Induction of GFP-positive cells
quantified in each of the plots is based on flow cytometry data.
c
4972 Chem. Commun., 2013, 49, 4971--4973
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
It should be noted that tamoxifen must be converted to
4-hydroxytamoxifen (4-OHT) in order to bind Cre-ERT2 effi-
ciently.26 MEFs are a heterogeneous mixture of cells, and likely
only 35% of our UT MEF line is able to mediate this conversion.
Our future efforts will focus on the generation of a caged 4-OHT
molecule for in vivo applications of 4-OHT-ONB in a transgenic
mouse model such as real-time tracking of GFP-expressing
Cre-ER(+) cells activated locally by irradiation of a specific tissue.
This work was supported by MNIMBS, University of Michigan
Medical School, the Ellison Medical Foundation, the California
Institute for Regenerative Medicine (grant RT2-02060); and the
NIH (grants R01HL058220 and R01CA86065). We thank Mr
Ankur Desai for his contribution in HPLC analysis and Ivan
Dimov, Shah Ali, and Humberto Contreras-Trujillo for technical
discussion.
Fig. 4 Pre-treatment with TAM-ONB 4 prior to UV uncaging. UT MEFs were
treated with TAM-ONB 4 for 24 h (left) or 1 h (middle) after 5 min UV uncaging
(‘‘post’’), or pre-treated with caged TAM-ONB 4 for 1 h (‘‘pre’’, right) prior to
5 min UV exposure. After 1 h treatments (either pre- or post-UV exposure),
UT MEFs were washed to remove unbound TAM-ONB 4.
Notes and references
1 R. Feil, Handb. Exp. Pharmacol., 2007, 3–28.
2 D. Metzger, J. Clifford, H. Chiba and P. Chambon, Proc. Natl. Acad.
Sci. U. S. A., 1995, 92, 6991–6995.
3 R. Feil, J. Brocard, B. Mascrez, M. LeMeur, D. Metzger and
P. Chambon, Proc. Natl. Acad. Sci. U. S. A., 1996, 93, 10887–10890.
4 R. Feil, J. Wagner, D. Metzger and P. Chambon, Biochem. Biophys.
Res. Commun., 1997, 237, 752–757.
5 E. Nakamura, M. T. Nguyen and S. Mackem, Dev. Dyn., 2006, 235,
2603–2612.
6 G. Mayer and A. Heckel, Angew. Chem., Int. Ed., 2006, 45, 4900–4921.
7 R. Wieboldt, K. R. Gee, L. Niu, D. Ramesh, B. K. Carpenter and
G. P. Hess, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 8752–8756.
8 J. W. Walker, J. A. McCray and G. P. Hess, Biochemistry, 1986, 25,
1799–1805.
9 P. G. S. Sonia and K. Pollitt, Angew. Chem., Int. Ed., 1998, 37, 2104–2107.
10 A. V. Karginov, Y. Zou, D. Shirvanyants, P. Kota, N. V. Dokholyan,
D. D. Young, K. M. Hahn and A. Deiters, J. Am. Chem. Soc., 2010, 133,
420–423.
dramatically reduced TAM-mediated recombination, requiring
treatments to be in serum-free media (Fig. S6, ESI†). UV had little
effect on MEF viability or background recombination of UT MEFs
(Fig. S7, ESI†), and we could expose 4 to UV directly on MEFs with
no loss in recombination efficiency or cell viability (Fig. S8, ESI†).
Uncaging also worked whether the UV source was above the cells,
or when cells (in plastic dishes) were placed directly onto the
UV source (Fig. S9, ESI†). We then compared recombination
frequency in UT MEFs treated for 1 h with 4 either prior to (pre)
or after (post) UV exposure (Fig. 4). In both cases, UT MEFs were
washed after 1 h treatment to remove unbound 4. Compared to
24 h treatment (24 h post) with uncaged 4, we observed a slight
reduction in recombination efficiency when UT MEFs were
treated for 1 h (1 h post) with uncaged 4. For UT MEFs treated 11 E. A. Lemke, D. Summerer, B. H. Geierstanger, S. M. Brittain and
P. G. Schultz, Nat. Chem. Biol., 2007, 3, 769–772.
12 A. V. Pinheiro, P. Baptista and J. C. Lima, Nucleic Acids Res., 2008,
with caged 4 for 1 h prior to UV uncaging (1 h pre), there was
clear evidence of recombination and GFP activation, though
36, e90.
substantially less than 1 h treatment with uncaged 4, and 13 N. K. Mal, M. Fujiwara and Y. Tanaka, Nature, 2003, 421, 350–353.
14 S. S. Agasti, A. Chompoosor, C.-C. You, P. Ghosh, C. K. Kim and
efficiency depended critically on the absence of serum during
V. M. Rotello, J. Am. Chem. Soc., 2009, 131, 5728–5729.
treatment (Fig. S10, ESI†). Nevertheless, our data clearly indicate
15 S. K. Choi, T. Thomas, M. Li, A. Kotlyar, A. Desai and J. R. Baker Jr,
that the caged, inactive form of 4 can bind to cells and mediate
recombination once released upon exposure to UV light.
In summary, we developed an efficient method for the
synthesis of a water soluble photocaged tamoxifen by using an
N-quaternization approach with an ONB photolabile moiety. Its
water solubility may also facilitate applications where standard
Chem. Commun., 2010, 46, 2632–2634.
16 Y. Shamay, L. Adar, G. Ashkenasy and A. David, Biomaterials, 2011,
32, 1377–1386.
17 K. H. Link, Y. Shi and J. T. Koh, J. Am. Chem. Soc., 2005, 127,
13088–13089.
18 Y. Shi and J. T. Koh, ChemBioChem, 2004, 5, 788–796.
19 S. K. Choi, T. P. Thomas, M.-H. Li, A. Desai, A. Kotlyar and
J. R. Baker, Photochem. Photobiol. Sci., 2012, 11, 653–660.
tamoxifen solvents (EtOH, DMSO, oil) may be too harsh for 20 S. K. Choi, M. Verma, J. Silpe, R. E. Moody, K. Tang, J. J. Hanson and
J. R. Baker Jr, Bioorg. Med. Chem., 2012, 20, 1281–1290.
21 W. M. Meylan, P. H. Howard and R. S. Boethling, Environ. Toxicol.
sensitive cells or tissues. The tamoxifen release was achieved by a
single exposure to UV light in a time- and dose-dependent
Chem., 1996, 15, 100–106.
manner. This tamoxifen photoprobe proved to be a valuable 22 Y. Ruzankina, C. Pinzon-Guzman, A. Asare, T. Ong, L. Pontano,
G. Cotsarelis, V. P. Zediak, M. Velez, A. Bhandoola and E. J. Brown,
Cell Stem Cell, 2007, 1, 113–126.
23 M. D. Muzumdar, B. Tasic, K. Miyamichi, L. Li and L. Luo, Genesis,
molecular tool for controlling GFP reporter induction via Cre-ER
activation in a transgenic cell model. This is the first report of
photo controlled GFP expression using a photocaged tamoxifen.
Unlike other enzyme-based reporter systems,17,18 the expression
of fluorescent proteins allows direct detection, quantification
2007, 45, 593–605.
24 A. Schroeder, M. S. Goldberg, C. Kastrup, Y. Wang, S. Jiang,
B. J. Joseph, C. G. Levins, S. T. Kannan, R. Langer and
D. G. Anderson, Nano Lett., 2012, 12, 2685–2689.
and imaging of intact biological tissues and live cells. We believe 25 A. Deiters, R. A. Garner, H. Lusic, J. M. Govan, M. Dush,
N. M. Nascone-Yoder and J. A. Yoder, J. Am. Chem. Soc., 2010, 132,
15644–15650.
26 J. N. Beverage, T. M. Sissung, A. M. Sion, R. Danesi and W. D. Figg,
that this photochemical release strategy is of potential value for
in vitro and in vivo applications that require spatiotemporal and
non-invasive gene expression or deletion.24,25
J. Pharm. Sci., 2007, 96, 2224–2231.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 4971--4973 4973