Fast and sensitive pretargeted labeling of cancer cells through a tetrazine/trans-cyclooctene cycloaddition

Full text of DOI:10.1002/anie.200903233

Angewandte
Chemie
DOI: 10.1002/anie.200903233
Live-Cell Imaging
Fast and Sensitive Pretargeted Labeling of Cancer Cells through a
Tetrazine/trans-Cyclooctene Cycloaddition**
Neal K. Devaraj, Rabi Upadhyay, Jered B. Haun, Scott A. Hilderbrand,* and Ralph Weissleder*
There is considerable interest in the use of bioorthogonal
covalent chemistry, such as “click” reactions, to label small
molecules located on live or fixed cells.[1] Such labeling has
been used for the visualization of glycans, activity-based
protein profiling, the site-specific tagging of proteins, the
detection of DNA and RNA synthesis, investigation of the
fate of small molecules in plants, and the detection of
posttranslational modification in proteins.[2–4] Most reported
applications rely on either the copper-catalyzed azide–alkyne
cycloaddition, which is limited to in vitro application owing to
the cytotoxicity of copper, or the elegant strain-promoted
azide–alkyne cycloaddition, which is suitable for live-cell and
in vivo application but is hindered by relatively slow kinetics
and the often difficult synthesis of cyclooctyne derivatives.[4,5]
New bioorthogonal reactions that do not require a catalyst
and show rapid kinetics are therefore of interest for different
molecular-imaging applications at the cellular level. Herein
we demonstrate the use of an inverse-electron-demand Diels–
Alder cycloaddition between a serum-stable 1,2,4,5-tetrazine
and a highly strained trans-cyclooctene to covalently label live
cells. We applied this reaction to the pretargeted labeling of
epidermal growth factor receptor (EGFR) tagged with
cetuximab (Erbitux) on A549 cancer cells. We found that
the tetrazine cycloaddition to trans-cyclooctene-labeled cells
is fast and can be amplified by increasing the loading of the
dienophile on the antibody. This highly sensitive targeting
strategy can be used to label proteins by treatment with a
secondary agent at nanomolar concentrations for short
durations of time.
cyclooctyne cycloaddition reactions and requires micromolar
concentrations for sufficient labeling.[3,4] On the basis of
previously reported rate constants, we decided to investigate
the coupling of tetrazines with more-strained dienophiles.[8]
Higher rate constants would enable faster and more efficient
labeling. Thus, less of the labeling agent would be required,
and the background signal would be decreased.
Fox and co-workers recently reported the use of a highly
strained trans-cyclooctene for bioconjugation.[6,9] Although
the rates reported were impressive, the tetrazine that yielded
the fastest rate has limited stability to nucleophiles and
aqueous media, with significant degradation observed after
several hours. In contrast, we reported the use of the novel
asymmetric tetrazine 1, which is very stable in water as well as
in whole serum: a prerequisite for in vivo applications.[7] We
hypothesized that tetrazine 1 would react with trans-cyclo-
octene significantly faster than the previously reported
norbornene and this would greatly improve the sensitivity
of cell labeling by tetrazine cycloaddition.
With this goal in mind, we synthesized the trans-cyclo-
octene dienophile 2 in two steps from a commercially
available cyclooctene epoxide. The trans-cyclooctene reacted
readily with tetrazine 1 to form isomeric dihydropyrazine
conjugation products in greater than 95% yield (Figure 1a;
see also the Supporting Information). The trans-cyclooctenol
2 can be converted into the reactive succinimidyl carbonate,
and the carbonate can be conjugated to amine-containing
biomolecules, such as monoclonal antibodies, through the
formation of a carbamate linkage. To determine the second-
order rate constant for the reaction of the tetrazine with the
trans-cyclooctene, we modified surface arrays of trans-cyclo-
octene-functionalized antibodies with a fluorescent tetrazine
probe and monitored the fluorescence signal over time (see
Figure S3a in the Supporting Information). From these data,
we determined a second-order rate constant of 6000 Æ
200mÀ1 sÀ1 at 378C (see Figure S3b in the Supporting Infor-
mation). This rate constant is several orders of magnitude
higher than the previously reported value for the cyclo-
addition of tetrazine 1 with a norbornene derivative, as well as
the previously reported rate constants for bioorthogonal click
reactions used to label live cells covalently.[3,4,7]
To demonstrate the utility of the reaction of a tetrazine
with a trans-cyclooctene for live-cell imaging, we chose to
label EGFR expressed on A549 lung cancer cells with the
anti-EGFR monoclonal antibody cetuximab. The concept of
pretargeting is illustrated in Figure 1b. The multistep labeling
of monoclonal antibodies is of interest as a result of the long
blood half-life of antibodies. This property leads to poor
target-to-background ratios when the antibodies are labeled
directly with imaging agents or cytotoxins.[10] A small-
Recently, we and others explored strain-promoted
inverse-electron-demand Diels–Alder cycloaddition reac-
tions of 1,2,4,5-tetrazines for bioconjugation.[6,7] We showed
that the cycloaddition of a tetrazine with a norbornene can be
applied to the pretargeted imaging of live breast cancer cells.
However, the rate of cycloaddition of the tetrazine with
norbornene was 1.6mÀ1 sÀ1 in serum at 208C. This rate is
comparable to previously reported rates for optimized azide–
[*] Dr. N. K. Devaraj, R. Upadhyay, Dr. J. B. Haun, Dr. S. A. Hilderbrand,
Prof. R. Weissleder
Center for Systems Biology, Massachusetts General Hospital
Richard B. Simches Research Center
185 Cambridge Street, Suite 5.210, Boston, MA 02114 (USA)
Fax: (+1)617-643-6133
E-mail: scott_hilderbrand@hms.harvard.edu
rweissleder@mgh.harvard.edu
Homepage: http://csb.mgh.harvard.edu/
[**] We thank Dr. Ned Keliher for helpful advice. This research was
supported in part by NIH grants U01-HL080731 and T32-CA79443.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200903233.
Angew. Chem. Int. Ed. 2009, 48, 7013 –7016
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7013

Communications
octene/AF555 (100 nm) for
45 minutes in serum. The
cells were then washed,
incubated at 378C with tet-
razine–VT680 (500 nm) for
10 min in 100% fetal bovine
serum (FBS), washed again,
and imaged immediately by
confocal
microscopy
(Figure 2). Direct labeling
of the antibodies with
AF555 was monitored in
the red channel (Figure 2b,
red). The antibody was
clearly visible both on the
surface of the cells and
inside the cells as a result
of
EGFR
internaliza-
tion.[16,17] Covalently bound
tetrazine–VT680 could be
visualized clearly in the
near-infrared (NIR) channel
(Figure 2c, green). Merging
of the red and NIR channels
revealed excellent colocali-
zation of the AF555 and
VT680 signals with little
Figure 1. a) Reaction of benzylaminotetrazine 1 with trans-cyclooctenol 2 by an inverse-electron-demand
Diels–Alder cycloaddition. Dinitrogen is released, and dihydropyrazine coupling products, such as 3, are
formed. b) Live-cell pretargeting. Cancer cells (blue), which overexpress EGFR, are exposed to the cetuximab–
trans-cyclooctene conjugate (red). In the next step, the pretargeted cells are labeled with a tetrazine bearing a
fluorophore such as VT680 (green).
molecule-based pretargeting strategy that relies on irrever-
sible covalent chemistry may circumvent these problems and
find general application for the delivery of imaging agents and
therapeutics.[11]
For cell-labeling studies, we chose an anti-EGFR antibody
as our target, as EGFR is of central importance in cancer-cell
signaling[12] and therefore a key target for therapeutic
inhibition,[13] and because prior studies with fluorophore-
labeled antibodies would serve as a reference.[14] Commer-
cially available cetuximab was labeled with trans-cyclooctene
succinimidyl carbonate and used for pretargeting experi-
ments. We chose to work with a green fluorescent protein
(GFP) positive A549 lung cancer cell line that has been shown
to have upregulated levels of EGFR.[15] To label cells
pretargeted with trans-cyclooctene-bearing antibodies, the
tetrazine amine 1 was conjugated to a commercially available
far-red indocyanine fluorophore, Vivo-Tag 680 (VT680 pur-
chased from VisEn Medical). The decision to use tetrazine–
fluorophore probes was based on the commercial availability
of numerous amine-reactive fluorophores. Furthermore, we
had used this compound previously and shown that the
tetrazine moiety is serum-stable and reacts rapidly with
strained dienophiles.[7]
Figure 2. Confocal microscopy of cetuximab-pretargeted GFP-positive
A549 lung cancer cells after tetrazine–fluorophore labeling. A) GFP
channel. Scale bar: 30 mm. B) Red channel: Cetuximab–trans-cyclo-
octene antibodies were also labeled with AF555 directly and imaged in
the rhodamine channel. Some of the antibody has been internalized,
as indicated by the signal inside the cells. C) Near-IR channel showing
the location of bound tetrazine–VT680 probe (500 nm, 10 min, 100%
FBS, 378C). D) Merging of GFP, red, and near-IR channels. The red
and near-IR channels show excellent colocalization, especially at the
surface of the cells. Intracellular cetuximab that has not reacted with
tetrazine–VT680 is clearly visible and was probably internalized prior
to the addition of the extracellular tetrazine–VT680 probe.
Initial pretargeting experiments used cetuximab modified
by both trans-cyclooctene and a single Alexa Fluor 555 dye
(AF555 purchased from Invitrogen). The AF555 dye was
conjugated to the antibody in addition to the trans-cyclo-
octene to track the location of the antibody and determine the
specificity of sequential VT680–tetrazine labeling. A549
cancer cells were first incubated with cetuximab–trans-cyclo-
background fluorescence. This result indicates that the
reaction of the tetrazine is extremely selective. As expected,
the reaction occurred primarily on the surface of the cells,
where EGFR concentrations are highest. A smaller amount
of cell-internalized, vesicle-associated NIR fluorescence was
7014
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7013 –7016

Angewandte
Chemie
also observed. This fluorescence is probably a result of EGFR
internalization after treatment with tetrazine–VT680.[16,17]
Control experiments with either unlabeled cetuximab and
tetrazine–VT680 or trans-cyclooctene–cetuximab and unla-
beled VT680 resulted in no NIR fluorescence (see Figure S4
in the Supporting Information).
Next, we tested whether labeling could be observed
without a washing step to remove the probe. The desire to
avoid such a step is relevant to applications in which one is
unable to perform stringent and multiple washing steps, such
as intracellular labeling, experiments in which cell handling
has to be minimized (with rare cells or highly specialized
cells), and in vivo labeling. The concentration of the
tetrazine–VT680 label was lowered to 50 nm to enable
observation of the covalent modification in real time. The
images in Figure 3 were taken during continuous imaging of
trans-cyclooctene. The conjugates were modified with tetra-
zine–VT680, and the resulting fluorochrome absorbance was
used to estimate the number of reactive trans-cyclooctene
units per antibody. In this fashion, cetuximab bearing one,
three, five, and six tetrazine–VT680-reactive trans-cyclooc-
tene moieties were prepared. Owing to the large size of
indocyanine dyes, for the higher loadings, the number of
reactive trans-cyclooctene moieties is probably lower than the
actual number of trans-cyclooctene moieties on the antibody.
These trans-cyclooctene–cetuximab conjugates bound to
EGFR-expressing A549 cells with excellent stability (see
Figure S5 in the Supporting Information).
We used flow cytometry to gain a more quantitative
understanding of live-cell fluorescent labeling with the
tetrazine. A549 cells were incubated with cetuximab (50 nm)
modified with either zero, one, three, five, or six reactive
trans-cyclooctene units. The cycloaddition was carried out
with tetrazine–VT680 (500 nm) at 378C in 100% FBS. After
30 min, the cells were washed, and the fluorescence intensity
was analyzed by flow cytometry. Figure 4a shows the relative
VT680 signal intensity after 30 min for all five loadings of the
trans-cyclooctene. To illustrate the practical effect of this
Figure 3. Real-time imaging of the tetrazine labeling of pretargeted
A549 cells. Cells were exposed to cetuximab–trans-cyclooctene,
washed, and imaged in 100% FBS by using the near-IR channel (top
left). The FBS was removed and immediately replaced with FBS
containing tetrazine–VT680 (50 nm; top middle image). Images were
taken periodically over 40 min. The signal around the cell surfaces
continues to increase as a function of time. Scale bar (top left): 30 mm.
the cycloaddition of the tetrazine–VT680 to the pretargeted
trans-cyclooctene on live cancer cells in 100% FBS. Tetra-
zine–VT680 first became visible as it reacted and concen-
trated on the surface of cells; at later times, punctate spots
within the cell were visible as tetrazine-labeled cetuximab was
internalized.
In an attempt to improve the signal-to-background ratio,
we increased the loading density of the reactive trans-
cyclooctene on the targeted antibodies. A greater number
of reactive sites per antibody should lead to more fluorophore
per antibody after labeling and thus result in signal amplifi-
cation. To vary the trans-cyclooctene loading, we exposed
cetuximab to different molar excesses of the amine-reactive
Figure 4. a) Analysis of live-cell labeling by flow cytometry. Fluores-
cence intensity after treatment for 30 min with tetrazine–VT680
(500 nm) versus the loading of reactive trans-cyclooctene on the
antibody. b) Confocal microscopy of A549 cells pretargeted with
cetuximab loaded with A) one, B) three, C) five, or D) six trans-cyclo-
octene moieties and then labeled by treatment with tetrazine–VT680
(100 nm) for 10 min. The increasing fluorescence signal correlates with
the increasing level of labeling of the antibody with the trans-cyclo-
octene. (All cells were exposed to the antibody in the same concen-
tration.) Scale bar (top left): 30 mm.
Angew. Chem. Int. Ed. 2009, 48, 7013 –7016
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7015

Communications
2008, 105, 15779 – 15784; F. Kaschani, S. H. L. Verhelst, P. F.
van Swieten, M. Verdoes, C.-S. Wong, Z. Wang, M. Kaiser, H. S.
Overkleeft, M. Bogyo, R. A. L. Van der Hoorn, Plant J. 2009, 57,
373 – 385; B. R. Martin, B. F. Cravatt, Nat. Methods 2009, 6, 135 –
138.
amplification on the imaging of live cells, we pretargeted
A549 cells with cetuximab (100 nm) conjugated to one, three,
five, and six reactive trans-cyclooctene moieties, then exposed
the cells to tetrazine–VT680 (100 nm) for 10 min prior to
imaging with confocal microscopy (Figure 4b). Cells pre-
targeted with trans-cyclooctene-conjugated cetuximab con-
structs with higher loadings of the trans-cyclooctene were
visualized readily, and the signal diminished as the amount of
the dienophile on the antibody decreased. The ability to
amplify signals by loading increased amounts of a small
molecule on the antibody could provide a strategy for
increasing the signal-to-background ratio for in vivo pretar-
geting schemes.
[3] X. H. Ning, J. Guo, M. A. Wolfert, G. J. Boons, Angew. Chem.
2008, 120, 2285 – 2287; Angew. Chem. Int. Ed. 2008, 47, 2253 –
2255.
[4] J. M. Baskin, J. A. Prescher, S. T. Laughlin, N. J. Agard, P. V.
Chang, I. A. Miller, A. Lo, J. A. Codelli, C. R. Bertozzi, Proc.
Natl. Acad. Sci. USA 2007, 104, 16793 – 16797.
[5] N. J. Agard, J. A. Prescher, C. R. Bertozzi, J. Am. Chem. Soc.
2004, 126, 15046 – 15047.
[6] M. L. Blackman, M. Royzen, J. M. Fox, J. Am. Chem. Soc. 2008,
130, 13518 – 13519.
[7] N. K. Devaraj, R. Weissleder, S. A. Hilderbrand, Bioconjugate
Chem. 2008, 19, 2297 – 2299.
[8] J. Sauer, D. K. Heldmann, J. Hetzenegger, J. Krauthan, H.
Sichert, J. Schuster, Eur. J. Org. Chem. 1998, 2885 – 2896.
[9] M. Royzen, G. P. Yap, J. M. Fox, J. Am. Chem. Soc. 2008, 130,
3760 – 3761.
[10] A. M. Wu, P. D. Senter, Nat. Biotechnol. 2005, 23, 1137 – 1146.
[11] D. A. Goodwin, C. F. Meares, Biotechnol. Adv. 2001, 19, 435 –
450; D. A. Goodwin, C. F. Meares, N. Watanabe, M. McTigue, W.
Chaovapong, C. M. Ransone, O. Renn, D. P. Greiner, D. L.
Kukis, S. I. Kronenberger, Cancer Res. 1994, 54, 5937 – 5946;
R. M. Sharkey, H. Karacay, T. M. Cardillo, C. H. Chang, W. J.
McBride, E. A. Rossi, I. D. Horak, D. M. Goldenberg, Clin.
Cancer Res. 2005, 11, 7109s – 7121s.
In conclusion, we have developed a highly sensitive
technique for the covalent labeling of live cancer cells on
the basis of the cycloaddition of a tetrazine to a highly
strained trans-cyclooctene. With the appropriate choice of a
cell-permeable labeling agent, this method should be readily
extendable to intracellular labeling and could facilitate the
tracking of tagged small-molecule drugs, signaling proteins, or
other components of the cellular machinery within live cells.
Furthermore, it may be possible to incorporate tetrazines and
trans-cyclooctene-substituted nonnatural amino acids into
proteins of interest in a site-specific manner and then to
reveal them in live cells. Given its speed and sensitivity in
whole serum, this labeling reaction should be adaptable to in
vivo imaging applications.
[12] S. V. Sharma, D. W. Bell, J. Settleman, D. A. Haber, Nat. Rev.
Cancer 2007, 7, 169 – 181.
[13] R. M. Bachoo, E. A. Maher, K. L. Ligon, N. E. Sharpless, S. S.
Chan, M. J. You, Y. Tang, J. DeFrances, E. Stover, R. Weissleder,
D. H. Rowitch, D. N. Louis, R. A. DePinho, Cancer Cell 2002, 1,
269 – 277; T. J. Lynch, D. W. Bell, R. Sordella, S. Gurubhagava-
tula, R. A. Okimoto, B. W. Brannigan, P. L. Harris, S. M.
Haserlat, J. G. Supko, F. G. Haluska, D. N. Louis, D. C. Chris-
tiani, J. Settleman, D. A. Haber, N. Engl. J. Med. 2004, 350,
2129 – 2139.
[14] T. Barrett, Y. Koyama, Y. Hama, G. Ravizzini, I. S. Shin, B. S.
Jang, C. H. Paik, Y. Urano, P. L. Choyke, H. Kobayashi, Clin.
Cancer Res. 2007, 13, 6639 – 6648; A. J. Bogdanov, H. W. Kang,
M. Querol, P. H. Pretorius, A. Yudina, Bioconjugate Chem. 2007,
18, 1123 – 1130; Y. Koyama, T. Barrett, Y. Hama, G. Ravizzini,
P. L. Choyke, H. Kobayashi, Neoplasia 2007, 9, 1021 – 1029; E. L.
Rosenthal, B. D. Kulbersh, R. D. Duncan, W. Zhang, J. S.
Magnuson, W. R. Carroll, K. Zinn, Laryngoscope 2006, 116,
1636 – 1641; E. L. Rosenthal, B. D. Kulbersh, T. King, T. R.
Chaudhuri, K. R. Zinn, Mol. Cancer Ther. 2007, 6, 1230 – 1238.
[15] W. J. Rachwal, P. F. Bongiorno, M. B. Orringer, R. I. Whyte, S. P.
Ethier, D. G. Beer, Br. J. Cancer 1995, 72, 56 – 64; N. S. Wang, C.
Liu, J. Emond, M. S. Tsao, Ultrastruct. Pathol. 1992, 16, 439 – 449.
[16] B. Vincenzi, G. Schiavon, M. Silletta, D. Santini, G. Tonini, Crit.
Rev. Oncol. Hematol. 2008, 68, 93 – 106.
Received: June 15, 2009
Published online: August 20, 2009
Keywords: bioorthogonal reactions · cancer · cycloaddition ·
.
imaging agents · tetrazines
[1] J. M. Baskin, C. R. Bertozzi, QSAR Comb. Sci. 2007, 26, 1211 –
1219; J. A. Prescher, C. R. Bertozzi, Nat. Chem. Biol. 2005, 1,
13 – 21; A. S. Raghavan, H. C. Hang, Drug Discovery Today
2009, 14, 178 – 184; A. M. Sadaghiani, S. H. L. Verhelst, M.
Bogyo, Curr. Opin. Chem. Biol. 2007, 11, 20 – 28; H. C. Kolb,
M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113, 2056 –
2075; Angew. Chem. Int. Ed. 2001, 40, 2004 – 2021; V. V.
Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew.
Chem. 2002, 114, 2708 – 2711; Angew. Chem. Int. Ed. 2002, 41,
2596 – 2599.
[2] A. B. Neef, C. Schultz, Angew. Chem. 2009, 121, 1526 – 1529;
Angew. Chem. Int. Ed. 2009, 48, 1498 – 1500; A. Salic, T. J.
Mitchison, Proc. Natl. Acad. Sci. USA 2008, 105, 2415 – 2420;
A. E. Speers, G. C. Adam, B. F. Cravatt, J. Am. Chem. Soc. 2003,
125, 4686 – 4687; A. Deiters, T. A. Cropp, M. Mukherji, J. W.
Chin, J. C. Anderson, P. G. Schultz, J. Am. Chem. Soc. 2003, 125,
11782 – 11783; C. Y. Jao, A. Salic, Proc. Natl. Acad. Sci. USA
[17] D. Patel, A. Lahiji, S. Patel, M. Franklin, X. Jimenez, D. J.
Hicklin, X. Kang, Anticancer Res. 2007, 27, 3355 – 3366.
7016
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7013 –7016