A "caged" Luciferin for Imaging Cell-Cell Contacts

Article abstract of DOI:10.1021/jacs.5b02774

Cell-cell interactions underlie fundamental biological processes but remain difficult to visualize over long times and large distances in tissues and live organisms. Bioluminescence imaging with luciferase-luciferin pairs is sufficiently sensitive to imag

Full text of DOI:10.1021/jacs.5b02774

Communication
pubs.acs.org/JACS
A “Caged” Luciferin for Imaging Cell−Cell Contacts
†
‡
†
,†,‡,§
William B. Porterfield, Krysten A. Jones, David C. McCutcheon, and Jennifer A. Prescher*
†
‡
§
Departments of Chemistry, Molecular Biology & Biochemistry, and Pharmaceutical Sciences, University of California, Irvine,
California 92697, United States
*
S Supporting Information
We recently exploited a modified luciferinknown as a
“caged” probeto visualize cell−cell proximity in rodent
ABSTRACT: Cell−cell interactions underlie fundamental
biological processes but remain difficult to visualize over
long times and large distances in tissues and live
organisms. Bioluminescence imaging with luciferase−
luciferin pairs is sufficiently sensitive to image cells in
vivo but lacks the spatial resolution to identify cellular
locations and interactions. To repurpose this technology
for visualizing cellular networks, we developed a “caged”
luciferin that produces light only when cells are in close
contact. This molecule comprises a nitroaromatic core that
can be selectively reduced (“uncaged”) by one cell type,
liberating a luciferin that can be selectively consumed by
neighboring, luciferase-expressing cells. When the two cell
types are in contact, robust light emission is observed. This
imaging strategy will enable the noninvasive visualization
of cell−cell interactions relevant to organismal biology.
5
models. “Caged” luciferins typically comprise appendages (i.e.,
“cages”) at the 6′-position of the scaffold, rendering the
molecules incapable of interacting with luciferase to produce
16
light. However, upon removal of the cage, a functional
luciferin (with an electron-donating group at the 6′-position) is
generated and available for the light-emitting reaction. “Caged”
probes have been previously employed to measure enzyme
1
7,18
19,20
activities
and track bioactive small molecules
in vitro
and in vivo. We utilized Lugal, a galactose-caged luciferin that is
selectively cleaved by the uncaging enzyme, β-galactosidase (β-
5
gal). If β-gal is expressed in one cell (i.e., an “activator” cell)
and Fluc is expressed in another (i.e., a “reporter” cell),
administration of Lugal can report on the proximity of the two
21
cells (Figure 1). In this scenario, luciferin released by activator
ell−cell contacts govern numerous biological processes1,
C
including cell growth, motility, and immune function.
Our understanding of these interactions is critically dependent
on our ability to “see” them, and several fluorescence imaging
1
,2
techniques have been developed for this purpose. While
powerful, these strategies require intense excitation sources and
are thus largely limited to visualizing interactions on a
2
−4
microscopic scale.
Bioluminescence imaging (BLI), a
complementary optical technique, is more suitable for capturing
macroscopic cell movements in whole tissues and organisms, as
it does not require excitation light. However, in its current
form, BLI is incapable of reporting on direct cell−cell
3
,5,6
Figure 1. General strategy for visualizing cell−cell interactions. (A)
Light is produced only when cells are in close contact. (B) “Caged”
luciferins enter activator cells, where an uncaging enzyme (e.g., β-gal
or NTR) liberates an active luciferin. This molecule diffuses into
nearby reporter cells (expressing Fluc), and light is produced. (C)
Structures of “caged” luciferins discussed in this work.
interactions owing to its low spatial resolution.
This leaves
a gap in our ability to investigate dynamic cellular interactions
across large length and time scales without knowing when and
where to look.
To address this void, we are engineering bioluminescent
probes that can report on cell−cell contacts. Bioluminescence
exploits enzymes (luciferases) that catalyze light production via
the oxidation of small molecule substrates (luciferins). The
most widely used enzyme−substrate pair comprises the
luciferase from the firefly (Fluc) and its small molecule
cells enters neighboring reporter cells and is used by Fluc to
5
produce light (Figure 1B). While Lugal is able to report on
relative distances between cell populations, the molecule is
3
,7
5
substrate, D-luciferin. When introduced into nonluminescent
cells, these components produce photons that can be captured
by sensitive cameras. The Fluc/D-luciferin pair has been widely
used for tracking cells and gene expression patterns in mouse
prone to nonspecific uncaging in biological media. Premature
uncaging results in luciferin release and bioluminescent light
emission in regions devoid of activator cells. Thus, sensitive
imaging of cellular interactions with Lugal remains challenging.
3
,7−9
models of human biology.
Synthetic analogs of D-luciferin
are also gaining traction in imaging studies, owing to their
unique light emission profiles and cell permeabilities.
Received: August 3, 2014
10−15
©
XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b02774
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
We aimed to develop an alternative caged luciferin with
improved robustness and specificity. The ideal molecule would
be readily accessible from common synthetic procedures,
produce little to no light with Fluc, and, importantly, be
resistant to nonspecific uncaging. Based on these consid-
erations, we were drawn to the nitroaromatic luciferin Luntr
Importantly, Luntr was found to be stable in both aqueous
solution and media (Figure S1). Chemically robust cages, as
noted earlier, are critical for our approach to imaging cell
contacts.
With the caged compound in hand, we evaluated its
responsiveness to NTR. Luntr was incubated with purified
2
2,32
(
1) that could potentially be uncaged by bacterial nitro-
NTR and NADH,
and substrate uncaging was monitored
reductase (NTR) (Figure 1B−C). NTR and related orthologs
have been widely used to activate nitroaromatic prodrugs
and imaging agents
via fluorescence spectroscopy. As shown in Figures 2A and S2,
uncaging was both selective and rapid, with product being
detected after only 15 min of incubation. We hypothesized that
the reduced compound was hydroxylamine 6, as this molecule
is capable of robust light emission with Fluc (Figure S3).
However, we were unable to directly isolate and characterize 6
22−27
2
8,29
in vivo. Mammalian cells also lack
endogenous NTR activity, suggesting that heterologous
expression of the enzyme in these cells would enable selective
5
uncaging in preclinical models. We further reasoned that the
nitro group would serve as an effective luciferin cage. Active
luciferins harbor electron-donating substituents at the 6′ ring
1
from the enzymatic reaction. Thus, we turned to H NMR to
monitor the enzymatic reduction in situ. Luntr was incubated
with NTR and NADH in deuterated PBS, and aliquots of the
reaction mixture were compared to synthetic standards of
hydroxylamine 6 and amine S1 (Scheme S1 and Figures S4−
3
0
position; thus, the electron-withdrawing nitro group in Luntr
would likely preclude robust light emission. Nitro group
reduction by NTR, though, should restore electron density and,
thus, provide a viable luciferin.
S5). NTR is known to reduce aryl nitro groups to
Luntr was readily synthesized using chemistries previously
22−25,33,34
1
1,12
hydroxylamines,
although nitroso and amine products
developed in our group (Scheme 1).
In brief, commercially
26,28,35
have been detected in a few cases.
Diagnostic proton
resonances for 6 were observed during the first few hours of
incubation (Figure 2B), prior to compound degradation
Scheme 1. Synthesis of the Caged Luciferin Luntr (1)
(
(
Figure S6). Formation of the fully reduced 6-amino luciferin
S1) was not observed (Figure S5) during the course of the
reaction. In the absence of NTR, no changes in the NMR
spectra for Luntr or NADH were noted. Importantly, when the
reaction mixtures (containing 6) were added to purified Fluc,
bioluminescent light was produced (Figure S7). These data
indicate that hydroxylamine 6 is the uncaged form of Luntr.
Uncaged Luntr was also sufficiently cell permeant to induce
+
bioluminescent light production in luciferase-expressing (Fluc )
reporter cells. In these experiments, Luntr (1) was incubated in
the presence or absence of recombinant NTR and NADH for
0−90 min. Aliquots of these reaction mixtures were added to
available 4-nitroaniline (2) was condensed with dithiazolium
chloride 3 (Appel’s salt). The product was fragmented in situ
with sodium thiosulfate to generate cyanothioformamide 4 in
good yield. This intermediate was subsequently cyclized to
provide cyanobenzothiazole 5. A final condensation step with
D-cysteine afforded the desired caged probe (Luntr, 1).
+
Fluc reporter cells, and bioluminescent light production was
measured over time. As shown in Figure 2C, an ∼120-fold
increase in bioluminescent signal was observed when Luntr was
incubated with NTR and NADH for 30 min. Reduced light
outputs were observed at longer incubation times, likely due to
31
Figure 2. NTR-mediated reduction of Luntr produces a light-emitting luciferin. (A) Fluorescence emission spectra of Luntr (100 μM) incubated
with NTR (2 μg/mL). The reaction was monitored over 170 min using an excitation wavelength of 350 nm. (B) Luntr (1, 5 mM) was combined
1
with NTR (2 μg/mL) and NADH (7.5 mM) in deuterated PBS, and the reduction was monitored by H NMR spectroscopy. Sample spectra (2 h
post-NTR addition) are shown, and diagnostic resonances are labeled. (C) Reduced Luntr is cell permeant. Luntr (2.75 mM) was incubated with+
NADH in the presence (+) or absence () of NTR for 0−90 min. Aliquots were collected over 90 min, transferred to wells containing Fluc
HEK293 (reporter) or control cells, and imaged. Sample images are shown, and the fold induction in bioluminescent signal (from reporter vs control
cells) is plotted. Error bars represent the standard deviation of the mean for n = 3 experiments.
B
DOI: 10.1021/jacs.5b02774
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
product inhibition or degradation of hydroxylamine 6 (Figure
S6).
NTR enzyme assays suggested that uncaging activity was
present in the activator cells (Figure S9). Thus, the low signals
were likely due to limited release of 6 out of the NTR activator
+
We hypothesized that Luntr would be beneficial for imaging
cell contacts, as the uncaged luciferin would remain localized
near the NTR source. To examine this possibility, we first
cells. To test this hypothesis, we generated cells that stably
expressed Fluc directly fused to NTR (Fluc-NTR). In these
cells, reduced Luntr can be immediately processed by Fluc,
without having to diffuse out of one cell and into another. Fluc-
+
monitored Luntr uncaging in NTR-expressing (NTR )
2
2
activator cells. Activator cells were seeded together with
+
+
Fluc reporter cells in a 96-well plate. When Luntr (250 μM)
NTR cells were plated and incubated with Luntr (250 μM) for
was added, only a modest increase in bioluminescent signal was
observed in the mixed cultures compared to reporter cells
0−60 min prior to imaging. Control experiments included
+
+
+
mixtures of NTR and Fluc cells, along with Fluc cells plated


incubated with control (NTR ) cells (Figure 3A). Similarly
with NTR cells. Comparable levels of NTR activity were
observed in the Fluc-NTR and NTR+ cells (Figure S9).
+
weak signals were observed when Luntr was first incubated with
However, the Fluc-NTR+ cells provided a nearly 40-fold
+
NTR cells, followed by media transfer to reporter cells (Figure
+
increase in bioluminescent signal compared to Fluc cells mixed
with NTR cells (Figure S10). These data suggest that once
S8). Previous experiments indicated that reduced Luntr 6 was
+
+
sufficiently cell permeant to enter Fluc cells (Figure 2C), and
Luntr is uncaged, the molecule does not readily escape activator
cells for use by neighboring reporter cells.
Limited release of uncaged Luntr is advantageous for imaging
direct cell−cell interactions, as active luciferin should be
completely consumed by the most proximal reporter cells.
+
Fluc cells located farther away should remain dark, owing to
insufficient quantities of luciferin reaching distant areas. Indeed,
+
+
when Fluc and NTR cells were plated in direct contact, light
emission was readily observed following Luntr addition
(
Figures 3 and S11−S12). When separated by just 1 mm,
+ +
though, Fluc and NTR cells produced little light in the
presence of Luntr (Figures 3B−C and S13). Similar trends
were observed using another cell type (Figure S14). Robust
+
+
light emission was also observed when Fluc and NTR cells
were allowed to grow into contact (Figures 3D−E and S15−
S16). In all cases, no cell death was observed upon trypan blue
staining. The stringent dependence of light emission on cellular
distance is ideal for use in tissues and other complex
environments, as only cells in close contact should produce
light.
The development of imaging tools to visualize cell contacts
addresses a void in our ability to “see” microscopic events at the
macroscopic level. In this work, we designed and synthesized a
novel caged luciferin, Luntr, which is stable in biological media.
Additionally, we verified that Luntr can be selectively uncaged
by bacterial nitroreductase in vitro for bioluminescent photon
production. NTR was expressed in various cell lines and used to
activate Luntr in mixed cell cultures. The limited release and
lifetime of uncaged Luntr proved advantageous for visualizing
cells in close contact. We anticipate that the technology will be
most useful for imaging cell−cell interactions where the largest
numbers of interacting cells are expressing NTR. We are
currently evaluating Luntr for visualizing host−pathogen
interactions in vivo, but the caged luciferin technology will be
broadly applicable to imaging cellular interactions relevant to
immune function, cancer progression, and numerous other
biological processes.
Figure 3. Luntr can be selectively uncaged and locally consumed. (A)
+

Activator (NTR ) or control (NTR ) (50,000) cells were plated with
+
5
0,000 reporter (Fluc ) cells. Luntr (250 μM) was added, and
bioluminescence images were recorded over 60 min. Sample images
are provided, and the fold induction in bioluminescent signal from
+

NTR vs NTR cells is plotted. (B and C) Close proximity is
necessary for signal induction. Fabricated stencils (left) were used to
separate (blue bars) or mix (black bars) activator and reporter cells.
Luntr (250 μM) was added, and images were recorded from 0 to 90
ASSOCIATED CONTENT
Supporting Information
■
*
S

min. NTR cells were also mixed (gray bars) or separated (white
Experimental details, full spectroscopic data for all new
compounds, and additional images. The Supporting Informa-
tion is available free of charge on the ACS Publications website
at DOI: 10.1021/jacs.5b02774.
bars) with reporter cells and imaged. For (A) and (C), error bars
represent the standard deviation of the mean for n = 3 experiments. (D

+
and E) NTR or NTR cells were plated 1 mm apart from reporter
cells (dotted lines indicate initial plating areas). The cells were allowed
to grow into contact over 7 d. Images were acquired daily (25 min
post-Luntr addition). Those shown are from day 6. Error bars in (E)
represent the standard deviation of the mean for n = 3 samples. Data
are representative of three replicate experiments.
AUTHOR INFORMATION
Corresponding Author
*jpresche@uci.edu
■
C
DOI: 10.1021/jacs.5b02774
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Notes
(24) Chung-Faye, G.; Palmer, D.; Anderson, D.; Clark, J.; Downes,
M.; Baddeley, J.; Hussain, S.; Murray, P. I.; Searle, P.; Seymour, L.;
Harris, P. A.; Ferry, D.; Kerr, D. J. Clin. Cancer Res. 2001, 7, 2662.
The authors declare no competing financial interest.
(
25) Johansson, E.; Parkinson, G. N.; Denny, W. A.; Neidle, S. J. Med.
Chem. 2003, 46, 4009.
26) Helsby, N. A.; Ferry, D. M.; Patterson, A. V.; Pullen, S. M.;
Wilson, W. R. Br. J. Cancer 2004, 90, 1084.
27) Barak, Y.; Thorne, S. H.; Ackerley, D. F.; Lynch, S. V.; Contag,
C. H.; Matin, A. Mol. Cancer Ther. 2006, 5, 97.
28) McCormack, E.; Silden, E.; West, R. M.; Pavlin, T.; Micklem, D.
R.; Lorens, J. B.; Haug, B. E.; Cooper, M. E.; Gjertsen, B. T. Cancer
Res. 2013, 73, 1276.
(29) Bhaumik, S.; Sekar, T. V.; Depuy, J.; Klimash, J.; Paulmurugan,
R. Gene Ther. 2012, 19, 295.
ACKNOWLEDGMENTS
■
(
This work was supported by the UC Irvine School of Physical
Sciences, the National Science Foundation (CBET-1351302
CAREER award to J.A.P.), and Research Corporation (Cottrell
Scholar award to J.A.P.) W.B.P. acknowledges support from the
BEST IGERT program (funded by NSF award DGE-1144901).
K.A.J. was supported by an institutional Chemical and
Structural Biology Training Grant predoctoral fellowship
(
(
(
T32-GM10856). We thank David Li and Elliot Hui for
(
30) White, E. H.; Worther, H.; McElroy, W. D. J. Am. Chem. Soc.
966, 88, 2015.
31) Inamoto, K.; Hasegawa, C.; Hiroya, K.; Doi, T. Org. Lett. 2008,
0, 5147.
32) Olekhnovich, I. N.; Goodwin, A.; Hoffman, P. S. FEBS J. 2009,
76, 3354.
33) Lee, M. K.; Williams, J.; Twieg, R. J.; Rao, J.; Moerner, W. E.
Chem. Sci. 2013, 4, 220.
providing the fabricated stencils, Logan Roberts and Todd
Holmes for help with bioluminescence microscopy, and the
Weiss, Martin, and Chamberlin laboratories for providing
reagents and equipment. Finally, we acknowledge members of
the Prescher lab for manuscript edits and helpful discussions.
1
(
1
(
2
(
REFERENCES
■
(
(
34) Christofferson, A.; Wilkie, J. Biochem. Soc. Trans. 2009, 37, 413.
́
35) Roldan, M. D.; Perez-Reinado, E.; Castillo, F.; Moreno-Vivian,
(
1) Germain, R. N.; Robey, E. A.; Cahalan, M. D. Science 2012, 336,
́
́
1676.
C. FEMS Microbiol. Rev. 2008, 32, 474.
(
2) Ntziachristos, V. Annu. Rev. Biomed. Eng. 2006, 8, 1.
(
3) Prescher, J. A.; Contag, C. H. Curr. Opin. Chem. Biol. 2010, 14,
80.
(
4) Paley, M. A.; Prescher, J. A. MedChemComm 2014, 5, 255.
(
5) Sellmyer, M. A.; Bronsart, L.; Imoto, H.; Contag, C. H.;
Wandless, T. J.; Prescher, J. A. Proc. Natl. Acad. Sci. U. S. A. 2013, 110,
567.
6) Cronin, M.; Akin, A. R.; Collins, S. A.; Meganck, J.; Kim, J. B.;
8
(
Baban, C. K.; Joyce, S. A.; van Dam, G. M.; Zhang, N.; van Sinderen,
D.; O’Sullivan, G. C.; Kasahara, N.; Gahan, C. G.; Francis, K. P.;
Tangney, M. PloS ONE 2012, 7, e30940.
(
(
7) Badr, C. E.; Tannous, B. A. Trends Biotechnol. 2011, 29, 624.
8) Rabinovich, B. A.; Ye, Y.; Etto, T.; Chen, J. Q.; Levitsky, H. I.;
Overwijk, W. W.; Cooper, L. J. N.; Gelovani, J.; Hwu, P. Proc. Natl.
Acad. Sci. U. S. A. 2008, 105, 14342.
(
9) Sacco, A.; Doyonnas, R.; Kraft, P.; Vitorovic, S.; Blau, H. M.
Nature 2008, 456, 502.
10) Woodroofe, C. C.; Meisenheimer, P. L.; Klaubert, D. H.; Kovic,
(
Y.; Rosenberg, J. C.; Behney, C. E.; Southworth, T. L.; Branchini, B. R.
Biochemistry 2012, 51, 9807.
(
11) McCutcheon, D. C.; Paley, M. A.; Steinhardt, R. C.; Prescher, J.
A. J. Am. Chem. Soc. 2012, 134, 7604.
12) McCutcheon, D. C.; Porterfield, W. B.; Prescher, J. A. Org.
Biomol. Chem. 2015, 13, 2117.
13) Reddy, G. R.; Thompson, W. C.; Miller, S. C. J. Am. Chem. Soc.
010, 132, 13586.
14) Evans, M. S.; Chaurette, J. P.; Adams, S. T.; Reddy, G. R.; Paley,
M. A.; Aronin, N.; Prescher, J. A.; Miller, S. C. Nat. Methods 2014, 11,
93.
15) Takakura, H.; Kojima, R.; Ozawa, T.; Nagano, T.; Urano, Y.
ChemBioChem 2012, 13, 1424.
(
(
2
(
3
(
(16) Li, J.; Chen, L.; Du, L.; Li, M. Chem. Soc. Rev. 2013, 42, 662.
(17) Yao, H.; So, M.; Rao, J. Angew. Chem., Int. Ed. 2007, 46, 7031.
(18) Rush, J. S.; Beatty, K. E.; Bertozzi, C. R. ChemBioChem 2010,
1
(
3
(
1, 2096.
19) Niwa, K.; Nakajima, Y.; Ohmiya, Y. Anal. Biochem. 2010, 396,
16.
20) Van de Bittner, G. C.; Dubikovskya, E. A.; Bertozzi, C. R.;
Chang, C. J. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 21316.
21) Lindberg, E.; Mizukami, S.; Ibata, K.; Miyawaki, A.; Kikuchi, K.
Chem.Eur. J. 2013, 19, 14970.
22) Friedlos, F.; Court, S.; Ford, M.; Denny, W. A.; Springer, C.
Gene Ther. 1998, 5, 105.
23) Knox, R. J.; Friedlos, F.; Boland, M. P. Cancer Metastasis Rev.
993, 12, 195.
(
(
(
1
D
DOI: 10.1021/jacs.5b02774
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX