Synthetic self-localizing ligands that control the spatial location of proteins in living cells

Article abstract of DOI:10.1021/ja4046907

Small-molecule ligands that control the spatial location of proteins in living cells would be valuable tools for regulating biological systems. However, the creation of such molecules remains almost unexplored because of the lack of a design methodology. Here we introduce a conceptually new type of synthetic ligands, self-localizing ligands (SLLs), which spontaneously localize to specific subcellular regions in mammalian cells. We show that SLLs bind their target proteins and relocate (tether) them rapidly from the cytoplasm to their targeting sites, thus serving as synthetic protein translocators. SLL-induced protein translocation enables us to manipulate diverse synthetic/endogenous signaling pathways. The method is also applicable to reversible protein translocation and allows control of multiple proteins at different times and locations in the same cell. These results demonstrate the usefulness of SLLs in the spatial (and temporal) control of intracellular protein distribution and biological processes, opening a new direction in the design of small-molecule tools or drugs for cell regulation.

Full text of DOI:10.1021/ja4046907

Article
pubs.acs.org/JACS
Synthetic Self-Localizing Ligands That Control the Spatial Location of
Proteins in Living Cells
Manabu Ishida,† Hideaki Watanabe,‡ Kazumasa Takigawa,‡ Yasutaka Kurishita,⊥ Choji Oki,‡
Akinobu Nakamura,§ Itaru Hamachi,*,⊥,¶ and Shinya Tsukiji*,†
‡
§
†Top Runner Incubation Center for Academia-Industry Fusion, Department of Bioengineering, Department of Electrical
Engineering, Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan
⊥Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto University, Katsura,
Nishikyo-ku, Kyoto 615-8510, Japan
¶Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Agency, 5 Sanbancho,
Chiyoda-ku, Tokyo 102-0075, Japan
S
* Supporting Information
ABSTRACT: Small-molecule ligands that control the spatial
location of proteins in living cells would be valuable tools for
regulating biological systems. However, the creation of such
molecules remains almost unexplored because of the lack of a
design methodology. Here we introduce a conceptually new
type of synthetic ligands, self-localizing ligands (SLLs), which
spontaneously localize to specific subcellular regions in
mammalian cells. We show that SLLs bind their target
proteins and relocate (tether) them rapidly from the cytoplasm
to their targeting sites, thus serving as synthetic protein
translocators. SLL-induced protein translocation enables us to
manipulate diverse synthetic/endogenous signaling pathways.
The method is also applicable to reversible protein translocation and allows control of multiple proteins at different times and
locations in the same cell. These results demonstrate the usefulness of SLLs in the spatial (and temporal) control of intracellular
protein distribution and biological processes, opening a new direction in the design of small-molecule tools or drugs for cell
regulation.
rapamycin addition, the POI is moved to the target site through
the formation of a ternary FKBP12-rapamycin-FRB complex.
The CID system is versatile and has been successfully used for
various biological applications.3c,d However, it is also associated
with several limitations. First, this method requires the
introduction of at least two fusion proteins into a cell to
control a single POI. Second, although new abscisic acid4 and
gibberellin5-based dimerizers have recently been developed,
expanding the repertoire of orthogonal chemical dimerizers is
quite difficult. In experimental biology, there is a growing need
for the ability to control multiple molecular processes at
different times and locations in the same cell. Third, induced
protein translocation is essentially irreversible with current CID
systems.3c,d,6,7 Finally, perhaps the most crucial limitation is its
inapplicability to control endogenous proteins: POIs must
always be conjugated with a dimerization domain.
INTRODUCTION
■
Cells are spatially organized molecular systems that contain
various organelles, cytoskeletal structures, and other membrane
domains inside the plasma membrane.1 The location of
proteins in the cell is critical to their cellular functions. In
addition, upon cell stimulation, many proteins change their
location dynamically, for example, from the cytoplasm to the
surface of a particular organelle. Such protein translocation is a
fundamental mechanism underlying the spatial regulation of
cell signaling processes.2
Controlling the intracellular location of proteins with
synthetic small molecules would be a powerful chemical
approach for artificially regulating cell function. Yet, methods
for this purpose are very limited. Chemically inducible
dimerization (CID) is the only general approach currently
available.3 The best-characterized CID system uses the small-
molecule rapamycin, which mediates dimerization of FK506-
binding protein 12 (FKBP12) and the FKBP-rapamycin
binding domain (FRB). In this technique, FRB (or FKBP12)
is localized to a specific cellular site using a targeting sequence,
whereas the partner FKBP12 (or FRB) is expressed in the
cytoplasm as a fusion with a protein of interest (POI). Upon
From the viewpoint of chemistry, it is of great importance to
establish a molecular approach to manipulate the spatial
location of a single POI with a single ligand without relying
Received: May 10, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja4046907 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Spatial control of intracellular protein location by SLLs. (a) Schematic illustration of the strategy (in the case of iPM targeting). SLLs are
synthetic ligands modified with small-molecule localization motifs that determine the site of localization. The SLL molecule can relocate its target
protein from the cytoplasm to the assigned region after entering the cell. (b) Structures of SLLs used in this study: 1, 3, and 4 for eDHFR; 5 for
FKBP12 (F36V) mutant; 2, control compound for 1.
on the expression of a second protein. Such a technique would
overcome many of the limitations of the CID approach and
may also provide a new direction for small molecule-based cell
regulation. This goal requires the development of synthetic
small molecules capable of controlling intracellular protein
location in their own right. However, despite its potential, the
creation of such molecules remains almost unexplored because
of the lack of a design methodology.
on the mechanism of phorbol myristate acetate (PMA), a small-
molecule activator of protein kinase C (PKC).8 PMA is a
unique molecule with two functionalities. The phorbol moiety
serves as a specific ligand to PKC, whereas the myristate group
enables localization of the PMA molecule in the plasma
membrane. As such, PMA can bind PKC and tether it from the
cytoplasm to the iPM, resulting in PKC translocation and
activation. Inspired by this, we reasoned that if a POI-selective
ligand has the ability to localize at a specific intracellular region9
(in a way that the ligand is still accessible by POI), the ligand
itself would spatially relocate the POI from the cytoplasm to its
targeting site in a “single ligand-single protein” manner (Figure
1a).10 We adopt a modular approach11 for designing such SLLs
(Figure 1b). A ligand of interest is linked to a small-molecule
localization motif that can bind to a specific component in a
targeted organelle or membrane region. The modular design
allows us to construct a variety of SLLs by interchanging the
ligand molecule and the localization motif. To test this idea, we
first chose trimethoprim (TMP) and eDHFR as a candidate
ligand and protein pair.12 Neither of the molecules has any
intracellular localization property, and thus, both diffuse readily
in the cytoplasm. TMP binds to eDHFR with high (nano-
molar) affinity and exhibits >1000-fold higher selectivity to this
protein over mammalian counterparts.12 Therefore, TMP-based
SLLs allow specific control of eDHFR-fusion proteins with
minimal background binding to endogenous DHFR in
mammalian cells.
Here we introduce a novel class of synthetic ligands, self-
localizing ligands (SLLs), which spontaneously localize to
specific organelles or subcellular regions in living mammalian
cells. In particular, in this work we developed several SLLs for
the inner leaflet of the plasma membrane (iPM), the nucleus,
and the cytoskeleton based on small-molecule ligands for
Escherichia coli dihydrofolate reductase (eDHFR) and (en-
gineered and native) human FKBP12. We show that SLLs bind
their target proteins and relocate (tether) them rapidly from
the cytoplasm to their targeting sites, thus acting as synthetic
protein translocators. The SLL tools were able to induce
protein translocation in a reversible manner, manipulate diverse
synthetic and natural signaling pathways, and simultaneously
control the location of distinct proteins at different times and
locations in the same cell. We also show that the SLL method
can be used to translocate an endogenous protein in intact cells.
These results demonstrate that conferring a self-localizing
ability on small-molecule ligands is a promising strategy to
generate synthetic small molecules that allow the spatial
regulation of protein distribution and biological processes in
living cells.
Inducible eDHFR Translocation by iPM-Localizing
TMP. The iPM is an important domain for cell signaling. We
thus targeted this region first and sought to generate a TMP-
based SLL that induces iPM translocation of an eDHFR fusion
protein in living cells. A previous study demonstrated that a
myristoyl-Gly-Cys (myrGC) lipopeptide derived from the N-
RESULTS AND DISCUSSION
Basic Strategy and SLL Design. To develop a general
design strategy for synthetic protein translocators, we focused
■
B
dx.doi.org/10.1021/ja4046907 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. mgcTMP-mediated protein translocation and activation of the synthetic Akt pathway. (a) Inducible translocation of eDHFR-GFP by
mgcTMP. Confocal fluorescence images of HeLa cells expressing eDHFR-GFP were obtained before and 20 min after the addition of 5 μM
mgcTMP. Scale bar, 10 μm. (b) Time course and dose dependence of eDHFR-GFP translocation. eDHFR-GFP translocation was induced by 5 μM
(red), 10 μM (green), or 20 μM (blue) mgcTMP or by 5 μM free TMP (black). Relative fluorescence signal intensities (F/F0) in the iPM (solid
line) and cytoplasm (dashed line) of cells were plotted as a function of time after the ligand addition. Data are represented as mean s.d. (n = 10
cells). (c) Immunoblot analysis of the synthetic Akt activity. NIH3T3 cells expressing YFP-eDHFR-AktKD were serum-starved and incubated for 15
min under the following conditions: lane 1, none; lane 2, 5 μM mgcTMP; lane 3, 5 μM mgcTMP and 50 μM TMP; lane 4, 10% fetal bovine serum
(FBS). Cell lysates were immunoblotted with the indicated antibodies. (d−g) Quantification of phosphorylation. Lane numbers correspond to the
conditions given in panel c. Phosphorylation of each protein was normalized to its corresponding total protein. Data are represented as mean values
relative to the mean of the no-treatment condition (lane 1) from three independent experiments. Error bars represent s.d. For GSK3β and Erk2, see
Supporting Information, Figure S2d,e.
terminus of Src-family proteins is membrane-permeant and
localizes to the iPM (and partially at the Golgi) in cells.13
Accordingly, we designed an iPM-localizing TMP in a modular
manner by connecting a myrGC localization motif to the TMP
ligand through a flexible spacer (mgcTMP, 1) (Figure 1b and
see Supporting Information for synthesis). To assess its
localization and translocation properties, we used human
epithelial HeLa cells expressing eDHFR fused to green
fluorescent protein (GFP) (eDHFR-GFP). Confocal fluores-
cence imaging showed that eDHFR-GFP was distributed
throughout the cytoplasm at rest and moved to the iPM
within 20 min upon addition of 5 μM mgcTMP (Figure 2a,b
and Figure S1a,b). The partial accumulation at the Golgi
(Figure S1c) is consistent with the localization properties of the
myrGC motif.13 The addition of free TMP instead of mgcTMP
had no effect on the spatial distribution of eDHFR-GFP
(Figure 2b and Figure S1d), indicating that the myrGC motif is
responsible for the induced iPM translocation. The presence of
excess free TMP blocked the mgcTMP-induced translocation
(Figure S1e). Also, no translocation occurred when an
mgcTMP analogue lacking the TMP ligand (mgcAc, 2, Figure
1b) was used (Figure S1f,g). These results verify that eDHFR-
GFP is recruited to the iPM by its direct binding to the ligand
moiety of mgcTMP. The eDHFR-GFP translocation was
accelerated in a dose-dependent manner and was completed
in 8 min at 20 μM mgcTMP (Figure 2b). Thus, the mgcTMP-
based system induces rapid protein translocation on a time
scale of minutes. Overall, we demonstrated that a synthetic
ligand with self-localization ability can control the spatial
location of its target protein in living cells.
mgcTMP-Mediated Control of Synthetic and Endog-
enous Signaling Pathways. Cells induce various types of
signaling by directing specific proteins to the iPM.1a,2 We
therefore investigated whether mgcTMP-induced protein
translocation can be used to artificially manipulate biological
signaling in living cells. We began by targeting the Akt signaling
pathway. Akt is a protein kinase involved in diverse cellular
processes such as cell survival, proliferation, metabolism, and
migration.14 Upon growth factor stimulation, Akt is recruited to
the iPM through binding of its N-terminal pleckstrin homology
(PH) domain to phosphatidylinositol 3,4,5-triphosphate (PIP3)
produced by phosphoinositide 3-kinase (PI3K). Subsequently,
Akt is activated by phosphorylation at Thr308 and Ser473 by
PDK1 and the mTOR-rictor complex, respectively.15 To build a
synthetic signaling system in which the Akt pathway is
selectively activated by mgcTMP, we linked yellow fluorescent
protein (YFP)-tagged eDHFR (YFP-eDHFR) to an Akt kinase
domain lacking the PH domain16 (YFP-eDHFR-AktKD). We
expressed this protein in mouse fibroblast NIH3T3 cells.
Treatment of cells with 5 μM mgcTMP for 20 min led to
efficient translocation of YFP-eDHFR-AktKD from the
cytoplasm to the iPM (Figure S2a,b). The translocation rate
(Figure S2c) was comparable to that observed with eDHFR-
GFP, indicating that both the N- and C-termini of eDHFR can
be fused to various (and relatively large) proteins without
affecting mgcTMP-mediated translocation. Furthermore, im-
munoblot analysis revealed that YFP-eDHFR-AktKD underwent
C
dx.doi.org/10.1021/ja4046907 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
phosphorylation at both Thr308 and Ser473 upon addition of
mgcTMP (Figure 2c,d, lane 2). The phosphorylation levels of
endogenous GSK3α and β, downstream substrates of Akt, also
significantly increased (Figure 2c,f and Figure S2d, lane 2).
Excess free TMP blocked the translocation and phosphor-
ylation of YFP-eDHFR-AktKD (Figure 2c,d, lane 3, and Figure
S2f). These results demonstrate that mgcTMP stimulates the
Akt pathway by recruiting YFP-eDHFR-AktKD to the iPM.
Moreover, serum treatment induced phosphorylation of
endogenous Akt but not YFP-eDHFR-AktKD because of the
lack of the PH domain (Figure 2c−e, lane 4). Addition of
mgcTMP did not activate endogenous Akt or other signaling
proteins, such as Erk1/2 (Figure 2c,g and Figure S2e, lane 2).
Therefore, this system allows the specific, inducible activation
of the synthetic Akt pathway in living cells using mgcTMP as an
input, independent of extracellular (upstream) stimuli.
To further test the usefulness of the mgcTMP-induced
spatial protein translocation system, we applied this method to
two other signaling molecules: the Rho-family small GTPase
Rac17 and the lipid kinase PI3K.18 First, we attempted to
activate endogenous Rac by recruiting Tiam1, a guanine
nucleotide exchange factor for Rac, to the iPM.19 We fused
YFP-eDHFR to the DH-PH domain of Tiam1 (YFP-eDHFR-
Tiam1DH‑PH) and expressed this protein in NIH3T3 cells. The
addition of mgcTMP translocated YFP-eDHFR-Tiam1DH‑PH to
the iPM and induced lamellipodia in most transfected cells (ca.
70%) (Figure S3a−d). No obvious changes in cell morphology
were observed when the translocation was inhibited by free
TMP or when YFP-eDHFR (lacking the Tiam1DH‑PH domain)
was recruited to the iPM (Figure S3d). The activation of
endogenous Rac was confirmed using a fluorescent biosensor,
PakGBD-mCherry,20 which binds to the active form of Rac
(Figure S3e,f). Next, we constructed a synthetic PI3K
(mCherry-eDHFR-p85iSH) by combining mCherry-tagged
eDHFR (mCherry-eDHFR) with the inter-Src homology 2
domain from p85 (p85iSH), which forms a complex with the
endogenous p110 subunit of PI3K in cells.21 The addition of
mgcTMP resulted in the production of PIP3 in mCherry-
eDHFR-p85iSH-expressing HeLa cells but not in mCherry-
eDHFR-expressing cells, as monitored by the PIP3 indicator
AktPH-GFP22 (Figure S4a,b). Immunoblotting revealed the
subsequent activation of endogenous Akt (Figure S4c,d).
Therefore, the mgcTMP-based eDHFR translocation system
is applicable to the conditional control of various iPM-
organized signaling pathways involving proteins and lipids.
Self-localizing TMPs for Other Organelles and
Reversible Protein Translocation. To demonstrate the
generality of the SLL approach to other cellular regions, we
next designed and synthesized nucleus- and cytoskeleton-
localizing TMPs. The former was generated using the DNA-
binding Hoechst dye23 as the localization motif (hoeTMP, 3)
(Figure 1b). The latter was based on the tubulin-binding drug
taxol24 (taxTMP, 4) (Figure 1b). Addition of hoeTMP
efficiently relocalized eDHFR-GFP from the cytoplasm to the
interior of the nucleus (Figure 3a and Figure S5a−e). Likewise,
eDHFR-GFP was assembled on the microtubule structure upon
addition of taxTMP (Figure 3b and Figure S6a−d). It was
shown that once translocated, eDHFR-GFP remained in the
nucleus or on the microtubule for at least 24 h, even after
washing the cells (Figure S5f and Figure S6e). However, it was
possible to readily return the protein to the original location
upon the addition of excess free TMP (Figure S7). Thus,
Figure 3. Protein translocation to various organelles. (a) Nuclear
translocation by hoeTMP. Confocal fluorescence images of HeLa cells
expressing eDHFR-GFP were obtained before and 90 min after the
addition of 5 μM hoeTMP. (b) Microtubule translocation by taxTMP.
Images were obtained before and 60 min after the addition of 5 μM
taxTMP. (c) Simultaneous control of two distinct proteins in the same
cell with orthogonal SLLs. Confocal fluorescence images of HeLa cells
coexpressing eDHFR-GFP and FKBP36V-mCherry were obtained
before (left), 90 min after the first addition of 2.5 μM hoeSLF*
(center), and 20 min after the second addition of 2.5 μM mgcTMP
(right). Top, GFP fluorescence; middle, mCherry fluorescence;
bottom, merged images. For translocation using hoeTMP and
taxTMP, see Supporting Information, Figure S9. Scale bar, 10 μm.
reversible small-molecule control of protein localization can be
readily achieved with the SLL system.
Orthogonal Control of Multiple Proteins in the Same
Cell. Inducible translocation of a protein distinct from eDHFR
can be achieved by replacing the ligand moiety. We designed
hoeSLF* (5) by linking SLF*, a specific synthetic ligand for the
F36V mutant of FKBP12 (FKBP36V),25 to the Hoechst motif
(Figure 1b). SLF* has a high (subnanomolar) affinity to
FKBP36V and 1000-fold selectivity to this protein over wild-type
FKBP12,25a allowing hoeSLF* to preferentially bind to
FKBP36V-fusion proteins in cells. The addition of hoeSLF*
induced the nucleus translocation of an FKBP36V-mCherry
fusion protein (FKBP36V-mCherry) (Figure S8). Because
TMP−eDHFR and SLF*−FKBP36V pairs are orthogonal to
each other, we were able to control the two distinct proteins
simultaneously and independently in the same cell. When
D
dx.doi.org/10.1021/ja4046907 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
hoeSLF* and one of the self-localizing TMPs were added
sequentially, FKBP36V-mCherry and eDHFR-GFP moved to
the nucleus and its expected site, respectively (Figure 3c and
Figure S9). Similarly, TMP-based SLLs were orthogonal to the
rapamycin CID-based translocation system (Figure S10).
SLL-Based Spatial Control of an Endogenous Protein.
Finally, we investigated whether the SLL approach is applicable
to control the spatial location of an endogenous protein in cells.
As a pilot study, here we targeted native (wild-type) FKBP12
(nFKBP), a cytoplasmic protein, and sought to relocate it to
the nucleus. Accordingly, we synthesized hoeSLF (6) consisting
of the SLF (no asterisk) ligand for nFKBP25a and the Hoechst
motif (Figure S11a). SLF specifically binds to nFKBP with high
(KD of 20 nM) affinity.25a Immunoblotting of cell fractions
indicated that nFKBP was predominantly localized in the
cytoplasm in intact HeLa cells, whereas the majority of the
protein moved to the nucleus upon addition of hoeSLF (Figure
S11b−d). The hoeSLF-induced translocation did not occur in
the presence of a competitive ligand (rapamycin), verifying that
nFKBP was directed to the nucleus through its direct binding
to hoeSLF. We therefore demonstrated that the intracellular
distribution of a specific endogenous protein can be
manipulated by the SLL method.
signaling activators is more challenging.29 Because many cell
signaling processes are triggered by protein translocation, we
envision that the SLL strategy can be a key component in
expanding the repertoire of small molecule-based signaling
activators.30,31
Further identification of small-molecule localization motifs
targeting a variety of intracellular locations is definitely required
to exploit the full potential of the SLL methodology. For
example, organelle-staining dyes,32 natural products,33 or short
(lipidated) peptide segments found in natural membrane-
associated proteins34 would be an important source for this
purpose. Such an effort is currently underway.
In closing, we believe that the concept of SLL (self-localizing
ligand) described herein opens a new direction in the design of
small-molecule tools or drugs for the study and control of living
biological systems.
ASSOCIATED CONTENT
* Supporting Information
■
S
Figures S1−S12 and detailed experimental procedures for
chemical synthesis, plasmid construction, and cell biological
experiments. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
tsukiji@vos.nagaokaut.ac.jp; ihamachi@sbchem.kyoto-u.ac.jp
Notes
CONCLUSIONS AND OUTLOOK
■
■
In this work, we have established a unique design methodology
for synthetic ligands that control spatial protein location in
living mammalian cells. A simple, diffusible ligand can be
converted to a synthetic protein translocator by endowing it
with self-localization ability. In the SLL system, the small-
molecule localization motif determines the subcellular destina-
tion and tethers the binding protein to the target site. The SLL
molecules could manipulate intracellular processes in diverse
ways, such as inducible (reversible) protein translocation,
conditional activation of synthetic and natural signaling
pathways, and simultaneous control of multiple proteins at
different times and locations in the same cell. No significant
cytotoxicity was observed for all SLLs reported herein under
the conditions used (Figure S12). The SLL-induced protein
relocation technique will become a powerful new chemical
approach for probing the role of protein localization and
spatiotemporally regulated signaling in cell biology.
The modular design is a chief advantage of our SLL system.
It allows us to generate various SLLs for different proteins by
simply exchanging the ligand moiety. In addition to eDHFR
and FKBP36V used in this work, other orthogonal protein
labeling techniques, for example, SNAP-tag26 and HaloTag,27
are now established. Therefore, by developing SLLs for these
protein tags using corresponding ligands/substrates, we will be
able to construct an integrated synthetic cell wherein the
location of diverse engineered (tag-fusion) proteins can be
flexibly controlled. Such a system enables more sophisticated
modulation of cellular functions based on protein location
control, providing an attractive new platform for synthetic
biology.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Prof. T. Nagamune (The University of Tokyo) for
eDHFR plasmid; Prof. Y. Gotoh (The University of Tokyo) for
Akt1 plasmid; Prof. T. Meyer (Stanford University) for
Tiam1DH‑PH, p85iSH and LDR plasmids; Prof. C. A. Voigt
(Massachusetts Institute of Technology) for PakGBD-mCherry
plasmid; Prof. C. Montell (Johns Hopkins University) for
AktPH-GFP plasmid; ARIAD Pharmaceuticals Inc. for FKBP36V
plasmid. We also thank Dr. Keiko Kuwata (Nagoya University)
for HRMS measurement and Prof. K. Takimoto (Nagaoka
University of Technology) for helpful discussions and critical
reading of the manuscript. Y.K. acknowledges the JSPS
Research Fellowships for Young Scientists. This work was
supported by the Nakajima Foundation and the Naito
Foundation and in part by a Grant-in-Aid for Challenging
Exploratory Research (No. 23651214) from the Japan Society
for the Promotion of Science (all to S.T.).
REFERENCES
■
(1) (a) Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.;
Walter, P. Molecular Biology of The Cell, 5th ed.; Garland Science: New
York, 2008. (b) Nabi, I. R. Cellular Domains, 1st ed.; Wiley-Blackwell:
NJ, 2011.
(2) (a) Teruel, M. N.; Meyer, T. Cell 2000, 103, 181. (b) Schles-
singer, J. Cell 2000, 103, 211.
(3) As reviews of CID techniques: (a) Fegan, A.; White, B.; Carlson,
J. C. T.; Wagner, C. R. Chem. Rev. 2010, 110, 3315. (b) Rutkowska, A.;
Schultz, C. Angew. Chem., Int. Ed. 2012, 51, 8166. (c) Putyrski, M.;
Schultz, C. FEBS Lett. 2012, 586, 2097. (d) DeRose, R.; Miyamoto, T.;
Another notable feature of the SLL approach is its
applicability to endogenous proteins. The use of appropriate
ligands may allow us to create novel SLL tools or drugs that
regulate cell behavior by altering the intracellular location of
specific endogenous proteins. Although there has been
significant progress in the development of bioactive small
molecules, the majority of these molecules are inhibitors of
enzyme activity or protein−protein interactions.28 Finding
Inoue, T. Pflugers Arch. 2013, 465, 409.
̈
(4) Liang, F.-S.; Ho, W. Q.; Crabtree, G. R. Sci. Signal. 2011, 4, rs2.
(5) Miyamoto, T.; DeRose, R.; Suarez, A.; Ueno, T.; Chen, M.; Sun,
T.; Wolfgang, M. J.; Mukherjee, C.; Meyers, D. J.; Inoue, T. Nat.
Chem. Biol. 2012, 8, 465.
E
dx.doi.org/10.1021/ja4046907 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(6) Recently, Ottman and co-workers developed a reversible CID
system based on fusicoccin and its engineered binding proteins. See:
Skwarczynska, M.; Molzan, M.; Ottman, C. Proc. Natl. Acad. Sci. U.S.A.
2012, 110, E377.
(7) More recently, Inoue and co-workers demonstrated reversible
(on and off) control of signaling processes by combining rapamycin-
and gibberellin-based CID techniques. See: Lin, Y.-C.; Nihongaki, Y.;
Liu, T.-Y.; Razavi, S.; Sato, M.; Inoue, T. Angew. Chem., Int. Ed. 2013,
52, 6450.
(8) (a) Schreiber, S. L. Chem. Eng. News 2003, 81 (9), 51. (b) Wang,
Q. J.; Bhattacharyya, D.; Garfield, S.; Nacro, K.; Marquez, V. E.;
Blumberg, P. M. J. Biol. Chem. 1999, 274, 37233.
(9) To date, several attempts have been made to localize bioactive
small molecules or peptides to specific intracellular compartments.
However, the major purpose of these studies is to improve the efficacy
of the molecules by delivering them to the locations where their target
been dismissed during standard drug development/screening.
However, they may acquire potent biological activities when
transformed to SLLs.
(31) For example, we have previously demonstrated that an
endogenous cell signaling pathway can be rapidly and efficiently
activated by subcellular targeting of a (microinjected) synthetic
peptide ligand. See: Hashiro, S.; Tsukiji, S.; Nagamune, T. J. Am.
Chem. Soc. 2009, 131, 13568.
(32) (a) The Molecular Probes Handbook: A Guide to Fluorescent
Probes and Labeling Technologies, 11th ed.; Johnson, I., Spencer, M. T.
Z., Eds.; Life Technologies Corporation: Carlsbad, CA, 2010.
(b) Rosania, G. R.; Lee, J. W.; Ding, L.; Yoon, H.-S.; Chang, Y.-T. J.
Am. Chem. Soc. 2003, 125, 1130. (c) Kawazoe, Y.; Shimogawa, H.;
Sato, A.; Uesugi, M. Angew. Chem., Int. Ed. 2011, 50, 5478.
(33) Alexander, M. D.; Burkart, M. D.; Leonard, M. S.; Portonovo,
P.; Liang, B.; Ding, X.; Joullie, M. M.; Gulledge, B. M.; Aggen, J. B.;
Chamberlin, R. A.; Sandler, J.; Fenical, W.; Cui, J.; Gharpure, S. J.;
Polosukhin, A.; Zhang, H.-R.; Evans, P. A.; Richardson, A. D.; Harper,
M. K.; Ireland, C. M.; Vong, B. G.; Brady, T. P.; Theodorakis, E. A.; La
Clair, J. J. ChemBioChem 2006, 7, 409.
proteins reside. See: (a) Rajendran, L.; Knolker, H.-J.; Simons, K. Nat.
̈
Rev. Drug Discovery 2010, 9, 29. (b) Yousif, L. F.; Stewart, K. M.;
Kelley, S. O. ChemBioChem 2009, 10, 1939.
(10) Gunning and co-workers previously reported a similar strategy
using a cholesterol-tethered synthetic ligand for STAT3 (signal
transducer and activator of transcription 3). See: Avadisian, M.;
Fletcher, S.; Liu, B.; Zhao, W.; Yue, P.; Badali, D.; Xu, W.; Schimmer,
A. D.; Turkson, J.; Gradinaru, C. C.; Gunning, P. T. Angew. Chem., Int.
Ed. 2011, 50, 6248.
(34) (a) Fivaz, M.; Meyer, T. Neuron 2003, 40, 319. (b) Kaufmann,
T.; Schlipf, S.; Sanz, J.; Neubert, K.; Stein, R.; Borner, C. J. Cell. Biol.
2003, 160, 53.
(11) Corson, T. W.; Aberle, N.; Crews, C. M. ACS Chem. Biol. 2008,
3, 677.
(12) Miller, L. W.; Cai, Y.; Sheetz, M. P.; Cornish, V. W. Nat.
Methods 2005, 2, 255.
(13) Creaser, S. P.; Peterson, B. R. J. Am. Chem. Soc. 2002, 124, 2444.
(14) Manning, B. D.; Cantley, L. C. Cell 2007, 129, 1261.
(15) Sarbassov, D. D.; Guertin, D. A.; Ali, S. M.; Sabatini, D. M.
Science 2005, 307, 1098.
(16) Andjelkovic, M.; Alessi, D. R.; Meier, R.; Fernandez, A.; Lamb,
́
N. J. C.; Frech, M.; Cron, P.; Cohen, P.; Lucocq, J. M.; Hemmings, B.
A. J. Biol. Chem. 1997, 272, 31515.
(17) Burridge, K.; Wennerberg, K. Cell 2004, 116, 167.
(18) Cantley, L. C. Science 2002, 296, 1655.
(19) (a) Worthylake, D. K.; Rossman, K. L.; Sondek, J. Nature 2000,
408, 682. (b) Inoue, T.; Heo, W. D.; Grimley, J. S.; Wandless, T. J.;
Meyer, T. Nat. Methods 2005, 2, 415.
(20) Levskaya, A.; Weiner, O. D.; Lim, W. A.; Voigt, C. A. Nature
2009, 461, 997.
(21) (a) Klippel, A.; Escobedo, J. A.; Hu, Q.; Williams, L. T. Mol.
Cell. Biol. 1993, 13, 5560. (b) Suh, B.-C.; Inoue, T.; Meyer, T.; Hille,
B. Science 2006, 314, 1454.
(22) Kwon, Y.; Hofmann, T.; Montell, C. Mol. Cell 2007, 25, 491.
(23) Pjura, P. E.; Grzeskowiak, K.; Dickerson, R. E. J. Mol. Biol. 1987,
197, 257.
(24) Guy, R. K.; Scott, Z. A.; Sloboda, R. D.; Nicolaou, K. C. Chem.
Biol. 1996, 3, 1021.
(25) (a) Clackson, T.; Yang, W.; Rozamus, L. W.; Hatada, M.;
Amara, J. F.; Rollins, C. T.; Stevenson, L. F.; Magari, S. R.; Wood, S.
A.; Courage, N. L.; Lu, X.; Cerasoli, F., Jr.; Gilman, M.; Holt, D. A.
Proc. Natl Acad. Sci. U.S.A. 1998, 95, 10437. (b) Marks, K. M.; Braun,
P. D.; Nolan, G. P. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9982.
(26) Keppler, A.; Gendreizig, S.; Gronemeyer, T.; Pick, H.; Vogel,
H.; Johnsson, K. Nat. Biotechnol. 2003, 21, 86.
(27) Los, G. V.; Encell, L. P.; McDougall, M. G.; Hartzell, D. D.;
Karassina, N.; Zimprich, C.; Wood, M. G.; Learish, R.; Ohana, R. F.;
Urh, M.; Simpson, D.; Mendez, J.; Zimmerman, K.; Otto, P.; Vidugiris,
G.; Zhu, J.; Darzins, A.; Klaubert, D. H.; Bulleit, R. F.; Wood, K. V.
ACS Chem. Biol. 2008, 3, 373.
(28) (a) Walsh, D. P.; Chang, Y.-T. Chem. Rev. 2006, 106, 2476.
(b) Arkin, M. R.; Wells, J. A. Nat. Rev. Drug Discov. 2004, 3, 301.
(29) Zorn, J. A.; Wells, J. A. Nat .Chem. Biol. 2010, 6, 179.
(30) For this purpose, small-molecule ligands should bind to their
target protein with high affinity and specificity, but without any
intrinsic inhibitory activity. Such inactive simple binders have usually
F
dx.doi.org/10.1021/ja4046907 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX