Isotopologous Organotellurium Probes Reveal Dynamic Hypoxia In Vivo with Cellular Resolution

Article abstract of DOI:10.1002/anie.201607483

Changes in the oxygenation state of microenvironments within solid tumors are associated with the development of aggressive cancer phenotypes. Factors that influence cellular hypoxia have been characterized; however, methods for measuring the dynamics of oxygenation at a cellular level in vivo have been elusive. We report a series of tellurium-containing isotopologous probes for cellular hypoxia compatible with mass cytometry (MC)—technology that allows for highly parametric interrogation of single cells based on atomic mass spectrometry. Sequential labeling with the isotopologous probes (SLIP) in pancreatic tumor xenograft models revealed changes in cellular oxygenation over time which correlated with the distance from vasculature, the proliferation of cell populations, and proximity to necrosis. SLIP allows for capture of spatial and temporal dynamics in vivo using enzyme activated probes.

Full text of DOI:10.1002/anie.201607483

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201607483
German Edition: DOI: 10.1002/ange.201607483
Activity-Based Probes
Isotopologous Organotellurium Probes Reveal Dynamic Hypoxia In
Vivo with Cellular Resolution
Landon J. Edgar, Ravi N. Vellanki, Trevor D. McKee, David Hedley, Bradly G. Wouters, and
Mark Nitz*
Abstract: Changes in the oxy-
genation state of microenviron-
ments within solid tumors are
associated with the develop-
ment of aggressive cancer phe-
notypes. Factors that influence
cellular hypoxia have been
characterized; however, meth-
ods for measuring the dynamics
of oxygenation at a cellular
level in vivo have been elusive.
We report a series of tellurium-
containing
isotopologous
probes for cellular hypoxia
compatible with mass cytome-
try (MC)—technology that
allows for highly parametric
interrogation of single cells
Figure 1. Design and synthesis of a probe for monitoring hypoxia dynamics. a) Schematic of the SLIP
approach to measuring dynamic cellular hypoxia in vivo. Murine tumor models are injected with
based on atomic mass spec- temporally spaced doses of isotopologues of Telox 2 followed by sacrifice, staining of cells with metal-
trometry. Sequential labeling chelating polymer–antibody conjugates and analysis of tumor tissue via MC. The difference in the
quantity of tellurium isotopes reports on the oxygenation state of a cell at a given time point. In the plot
with the isotopologous probes
1
25
122
on the right, a low Te signal would be expected compared to Te due to the lower pO during
2
(
SLIP) in pancreatic tumor
1
22
exposure to Telox 2. b) Reaction conditions: i) N-bromosuccinimide (1.15 equiv), AgNO (1.0 equiv),
3
xenograft models revealed
changes in cellular oxygenation
over time which correlated with
acetone, RT, 3 hours, 85%. ii) 3-Butyn-1-ol (1.2 equiv), CuCl (2 mol%), 30% BuNH (aq.), 08C!RT,
2
0
.5 hours, 80%. iii) Tetrabutylammonium fluoride (TBAF) (4.0 equiv), THF, 08C, 0.5 hours, not isolated.
0
iv) Te , sodium hydroxymethylsulfinate (rongalite) (11.0 equiv), NaOH(aq.), 708C, 4 hours, 90%. v) PPh3
the distance from vasculature, (1.05 equiv), DIAD (1.05 equiv), azomycin (1.05 equiv), THF, 08C!RT, 3 hours, 90%. c) Mechanism of
the proliferation of cell popula- enzyme-mediated reduction of the 2-NI functionality to produce the electrophilic 2-nitrenium ion leading
[
7c–e]
to non-specific protein labeling.
tions, and proximity to necrosis.
SLIP allows for capture of
spatial and temporal dynamics
in vivo using enzyme activated probes.
survival advantage as compared with chronically hypoxic
[
3]
cells. Cyclic hypoxia may be particularly important in its
ability to activate pathways that promote metastasis, genomic
instability, and resistance to treatment, but an understanding
of temporal changes in hypoxia at the cellular level is largely
R
egions with limited access to oxygen are common in solid
[1]
tumors from many cancers. Poor patient prognosis and an
increased risk of metastasis have both been correlated with
[
2]
[4]
the degree of tumor hypoxia. Cells exposed to repeated
cycles of hypoxia and reoxygenation have a long-term
lacking. The ideal method for monitoring oxygenation
dynamics would allow for highly parametric spatiotemporal
analysis with cellular resolution. To achieve this, we turned to
mass cytometry (MC); a technology that employs bioorthog-
onal heavy isotopes to report on a large (> 40) number of
[*] Dr. L. J. Edgar, Prof. M. Nitz
[
5]
Department of Chemistry, University of Toronto
biological parameters simultaneously. MC is compatible
with activity-based profiling of single cells using organo-
tellurium mass tags, and we reasoned that this approach could
be translated to spatiotemporal investigation of hypoxia using
80 St. George Street, Toronto, Ontario, M5S 3H6 (Canada)
E-mail: mnitz@chem.utoronto.ca
Dr. R. N. Vellanki, Dr. T. D. McKee, Dr. D. Hedley, Prof. B. G. Wouters
Departments of Radiation Oncology, Medical Biophysics, and the
STTARR Innovaton Centre, Princess Margaret Cancer Centre, Uni-
versity Health Network
[
6,7]
recently developed imaging MC (IMC).
We envisioned an approach where isotopically coded
tellurium-containing probes, which form covalent conjugates
in hypoxic cells, could be serially injected into a tumor
xenograft-bearing mouse. Changes in cellular hypoxia would
101 College Street, Toronto, Ontario, M5G 1L7 (Canada)
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201607483.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 2. Evaluation of Telox 2 as a probe for cellular hypoxia in vitro and in vivo. a) PANC-1 cells in differently oxygenated atmospheres (3 h) were stained
1
03
193/191
with metal-containing nucleic acid intercalators ( Rh and/or
Ir), mixed, and analyzed via MC. Cartoons of expected MC readouts are presented at
1
30
130
right. b) Event length vs. Te signal. c) Intercalator signals from low, d) intermediate, and e) high Te gates. f) PANC-1 tumor xenograft-bearing SCID mice
ꢀ
1
were co-injected with Telox 2 and Pimo (60 mgkg each). After 3 h exposure, sacrifice, and digestion of tumor tissue, single cells were sorted via FACS and
1
30
then analyzed by MC for Te. Notes: All MC signals are reported as median values (arbitrary units). % of cells within drawn gates are indicated for (c)–(e).
be quantified by comparing tellurium isotope labeling by MC
Figure 1a). This strategy of serial labeling with isotopologous
specifically label nucleophile-bearing proteins (Fig-
ure 1c).
[7b–e]
(
probes (SLIP) has the unique advantage of removing
complications that arise from using structurally different
probes to report on the time-dependence of a biological
phenomenon. Specifically, these isotopologous probes would
have no significant differences in pharmacokinetics or
metabolism that could confound interpretation of results,
The performance of Telox 2 was investigated in vitro using
the human pancreatic cancer cell line PANC-1, which is
known to form xenografts with a high hypoxic fraction.
[10]
Cells were incubated with Telox 2 under either near-anoxia
(< 0.02% O ), hypoxia (1% O ), or normoxia (21% O )
2
2
2
(Figure 2a). The cells were then washed, fixed, and coded by
staining with one of three combinations of heavy isotope-
[
8]
unlike other approaches to this type of strategy.
1
03
193
Our previous MC-compatible hypoxia probe exhibited
favorable activity in vitro, but failed to reliably identify
hypoxic cells in vivo likely due to the stability of the
containing nucleic acid intercalators ( Rh and/or Ir). The
1
30
intercalator code allowed validation of the Te signals from
individual cells following mixing and analysis by MC
[7a]
130
telluroether. Chemical optimization of the organotellurium
mass tag revealed that a more stable tellurophene would
(CyTOF2, Fluidigm). A density map of
Te vs. event
1
03
length revealed three populations of cells (Figure 2b). Rh
vs. Ir output from each of these three populations retrieved
[9]
193
likely yield a better performing probe. Indeed, combining
a 2-nitroimidazole (2-NI) with a tellurophene produced
a probe (Telox 2) with excellent solution stability and low
toxicity both in vitro and in vivo (Figure 1b; Supporting
Information, Figures S1, S2, and S6). We reasoned that this
probe would label hypoxic cells similarly to other 2-NI-
bearing probes, where bioreductive metabolism converts the
the original nucleic acid staining combination expected for
each population demonstrating that Telox 2 was capable of
labeling cells in an oxygen concentration-dependent manner
(Figure 2c–e). Labeling with Telox 2 was found to be dose-
dependent, linear with time, and was observed over a wide
range of O concentrations suggesting that Telox 2 may be
2
2
-NI to a 2-nitrenium imidazole which proceeds to non-
used as a probe for measuring hypoxia on a gradient of O2
(Supporting Information, Figures S3 and S4). Additionally,
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
murine plasma using LC/MS/
[
13]
MS.
This experiment indi-
cated that a second isotopo-
logue could be administered
1
2–16 hours (ca. 4t ) after the
1/2
first, as the majority of the
initial isotopologue would have
cleared circulation by that time
(Supporting Information, Fig-
ure S7). Additionally, we mea-
sured the in vitro stability of
Telox 2 conjugates via MC to
determine how long the tellu-
rium mass tag would remain
detectable in cells after reoxy-
genation (Supporting Informa-
tion, Figure S8). Even after 48
hours tellurium levels were well
Figure 3. Isotopically enriched Telox 2 and imaging mass cytometry report on diffusion-limited hypoxia in
1
25
ꢀ1
vivo. a) A mouse bearing a PANC-1 tumor xenograft was injected with Te-enriched Telox 2 (60 mgkg ). above the limit of detection
After 3 h the mouse was sacrificed, and the tumor cryosectioned. Tissue sections were stained with MC-
indicating
that
conjugates
1
68
191/193
compatible reagents (i.e. anti-Ki-67 antibody ( Er-labeled) and
Ir nucleic acid intercalator) and
should remain detectable at
the end of a SLIP experiment.
In order to evaluate
analyzed via IMC. b) Tissue from experiment in (a) stained with H&E (top). A separate adjacent section
1
25
125
168
193
analyzed via IMC (large panel). Red= Te ( Telox 2), green= Er (Ki-67), blue= Ir (DNA intercala-
1
68
125
tor). Er (middle) and Te (bottom) channels are in grayscale on left. c) Processed image of data from
b). Segmented cells are shaded on a thermal scale that represents the intensity of the Te mass
channel (thermal legend on the right, arbitrary units). d) Te intensity as a function of distance from the
blood vessel. The vertical axis represents the average Te signal for each segmented cell within 20 mm-
1
25
(
dynamic cellular oxygenation
in the context of underlying
tissue morphology, we investi-
gated Telox 2 labeling using
1
25
1
25
thick circular slices of increasing distance from the blood vessel. Error bars represent the standard
1
25
deviation of Te for all cells within the corresponding 20 mm thick circular slice. Note: Scale
bars=100 mm. The blood vessel is outlined with a dashed oval.
[6]
IMC. We synthesized isotopi-
cally enriched variants of
Telox 2
containing
either
1
25
122
cellular Te accumulation was found to be potentiated by the
over-expression of cytochrome P450 oxidoreductase (POR)
in HCT116 cells, and reduced following POR knockout,
consistent with the mechanism of action of the 2-NI activity-
based group (Figure 1c; Supporting Information, Fig-
a
Te or Te nucleus (> 92% and > 91% enrichment,
respectively). Since Te does not share a mass with any other
stable isotope and Te only shares its mass with a low-
abundance (4.6%) isotope of tin, there should be no isotope
background in the tissue.
1
25
1
22
[
11]
ure S5).
Initially, a PANC-1 tumor xenograft-bearing mouse model
was injected with Telox 2 and introduced to an oxygen-
1
25
Next, we confirmed the ability of Telox 2 to label hypoxic
cells in vivo using the previously validated hypoxia probe
pimonidazole (Pimo, Hypoxyprobe) as a reference (Fig-
reduced atmosphere (7% O ) for 3 hours in order to promote
2
[
10]
an increase in tumor hypoxia (Figure 3a). The tumor was
then harvested, cryosectioned (5 mm), and stained with an
antibody against the proliferation marker Ki-67 (labeled with
[12]
ure 2 f).
PANC-1 tumor xenograft-bearing mice were
injected with both Pimo and Telox 2. After three hours the
mice were sacrificed and the tumors excised/digested into
single cells. These cells were then stained with a FITC-tagged
antibody against Pimo conjugates and sorted using fluores-
cence-based flow cytometry (FACS) into four fractions. We
reasoned that cells enriched in Pimo should also be enriched
in Telox 2 since both probes were designed to label hypoxic
cells using the same 2-NI-dependent mechanism. Injection of
each fraction obtained through FACS into a CyTOF2
revealed Pimo-dependent enrichment of
that the two probes labeled similar populations of cells
in vivo.
We investigated parameters that would influence the
ability of Telox 2 to measure dynamic changes in tumor
hypoxia using the SLIP approach (Figure 1a). In order to
obtain meaningful data from this type of experiment, it was
necessary to understand when a second isotopologue of
Telox 2 could be administered without interference from the
first. Thus, we measured the circulating t1/2 of the probe in
1
68
191/193
Er) and the
Ir nucleic acid intercalator (an IMC-
compatible surrogate for DAPI nuclear stain). IMC identified
an example of diffusion-limited hypoxia, where the most
1
25
intense Te signal was detected in viable cells most distant
from an apparent blood vessel, adjacent to necrosis, and anti-
localized with proliferating cells as identified by H&E stain
from an adjacent tissue section and the Ki-67 antigen,
respectively (Figure 3b). Using a cell segmentation algorithm,
the approximate size and shape of individual cells were traced
1
30
Te, suggesting
1
25
and a Te signal intensity assigned to each individual cell
(Figure 3c). This data processing revealed that the maximum
1
25
Te signal was observed at a distance of ca. 300 mm (beyond
the average diffusion distance of O in tissue) from the blood
2
vessel, and upon transition into necrotic tissue the signal
decreased, likely due to decreased enzymatic activity in this
[
14]
region.
We next investigated hypoxia dynamics using isotopo-
logues of Telox 2. Mice with PANC-1 tumor xenografts were
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 4. The SLIP strategy reports on dynamic hypoxia in pancreatic tumors. a) Schematic of a SLIP IMC experiment. Mice bearing PANC-1 tumor
1
25
122
xenografts were injected with Telox 2 + EF5 and left to breathe 21% O . After 24 h, Telox 2 was injected and animals were exposed to either 7%, 95%,
2
or 21% O . Sacrifice occurred 24 h later and tissues were processed as in Figure 3a. b) H&E-stained section, c) fluorescence microscopy image (anti-EF5
2
1
22
122
125
125
antibody=pink, DAPI=blue), and d) raw IMC image of tissue from a reduced O -breathing (7%) model. Red= Te ( Telox 2), green= Te ( Telox 2),
blue= Er (Ki-67). Individual mass channels on right. e) Processed image of data in (d). Segmented cells are shaded on a thermal scale (right of panel i)
representing the value of Te– Te. f) Processed image from an increased O -breathing (95%) model, and g) constant normoxia (21%) model.
h) Enhanced view of (g). Ki-67+ ( Er) cells are in the grayscale panel. Proposed region of hypoxic cell turnover and direction of cell displacement indicated
2
1
68
1
22
125
2
1
68
by the white box and arrow, respectively. i) Processed image from a 21% O -breathing OCIP-51 tumor model. j) Histograms of segmented cells from panels
2
ꢀ1
193
(
e), (f), (g), and Figure S13. Notes: All probes dosed at 60 mgkg . All scale bars=400 mm. Blue channel in processed images is Ir (nuclear stain). Gray
masks mark necrotic zones.
1
25
ꢀ1
injected with a mixture of Telox 2 and EF5 (60 mgkg
tumor. The first was stained with H&E for brightfield imaging
(Figure 4b). The second was prepared for fluorescence
microscopy via staining with Hoechst nuclear stain and
a Cy5.5-labeled anti-EF5 antibody—a well-validated probe
against hypoxia. The third adjacent tissue section was stained
with reagents compatible with detection via IMC including
each) and left to breathe normal atmosphere to establish
baseline levels of hypoxia (Figure 4a). After 24 hours the
mice were injected with Telox 2 and subsequently split into
four treatment groups. The first group was transferred to
a sealed chamber containing 7% O , the second to a sealed
chamber containing 95% O , and the third was left in normal
1
22
2
1
91/193
168
the
Ir-containing nucleic acid intercalator and the Er-
2
atmosphere (21% O ). The fourth group was injected with the
labeled anti-Ki-67 antibody. Fluorescence microscopy
revealed regions of the tumor section that had been labeled
with EF5. We used this information to guide our IMC
experiments since this technology is currently lower-through-
put than optical techniques making ablation of an entire
2
vasodilator hydralazine one hour post-injection with
1
22
Telox 2 (Supporting Information, Figure S9). After an
additional 24 hours the tumors were harvested and cryosec-
tioned. Three adjacent tissue sections were stained for each
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
tumor section impractical (Figure 4c). As expected, a similar Acknowledgements
1
25
localization of EF5 (Cy5.5) and Telox 2 ( Te) in adjacent
sections was observed; co-localization was not expected as
adjacent tissue sections were being examined and the
structurally distinct probes have different pharmacokinetics
The authors acknowledge Fluidigm for providing access to
instrumentation, reagents, and expertise in collecting MC and
IMC data. Specifically, we would like to thank D. Bandura, V.
Baranov, T. Closson, O. Loboda, O. Ornatsky, and J. Watson.
We also acknowledge Q. Chang for productive discussions
and G. McKeown for assistance processing IMC data. Addi-
tionally, we recognize W. Wilson for providing access to the
mutant HCT116 cell lines. The authors would like to
acknowledge the Spatio-Temporal Targeting and Amplifica-
tion of Radiation Response (STTARR) program and its
affiliated funding agencies. Finally, the authors acknowledge
funding from the Canadian Cancer Society, Natural Sciences
and Engineering Research Council of Canada, and Fluidigm
Canada.
(
Figure 4c,d). Following IMC analysis, single cells were
digitally segmented and the difference between signals for
1
22
125
the two tellurium isotopes ( Te– Te) was reported for each
cell that exhibited significant labeling with either isotope
1
22
(
Figure 4e). We propose that cells exhibiting a value of Te–
1
25
Te > 0 became increasingly hypoxic following administra-
tion of the second isotopologue whereas those with a value
0 became less hypoxic (reoxygenated). As expected, tumors
<
from mice subjected to reduced O -breathing following
2
injection of the second isotopologue exhibited a markedly
1
22
125
increased number of cells with Te– Te > 0 as compared to
mice subjected to increased O -breathing, where a high
2
1
22
125
proportion of cells exhibited negative
Te– Te values
Keywords: activity-based probes · cellular hypoxia ·
(
Figure 4e,f,j). Samples from the group treated with hydra-
dynamic biology · mass spectrometry · tellurium
lazine after dosing with the second isotopologue also exhib-
ited exacerbated hypoxia, consistent with reduction of blood
flow through tumor vasculature as a result of lowered blood
[15]
pressure (Figure 4j; Supporting Information, Figure S16).
[
[
[
1] W. R. Wilson, M. P. Hay, Nat. Rev. Cancer 2011, 11, 393 – 410.
2] N. Rohwer, T. Cramer, Drug Resist. Updates 2011, 14, 191 – 201.
3] P. Martinive, F. Defresne, C. Bouzin, J. Saliez, F. Lair, V.
Grꢀgoire, C. Michiels, C. Dessy, O. Feron, Cancer Res. 2006, 66,
11736 – 11744.
In all cases, hypoxic regions were generally anti-correlated
with cell proliferation as reported by the presence of Ki-67
(
Figure 4d; Supporting Information, Figures S12, S21, S30,
S39, and S48).
Having demonstrated that the SLIP approach could
[4] M. W. Dewhirst, Y. Cao, B. Moeller, Nat. Rev. Cancer 2008, 8,
25 – 437.
[
4
reveal controlled changes in oxygenation in vivo, we used
these tools to measure hypoxia dynamics in a tumor model
that was a closer analogue of a clinical situation. The fourth
group of tumors, having not been subjected to artificial O2
manipulation, were analyzed for regions of increased and
decreased hypoxia taking place over 48 hours. Indeed, regions
5] O. Ornatsky, D. Bandura, V. Baranov, M. Nitz, M. A. Winnik, S.
Tanner, J. Immunol. Methods 2010, 361, 1 – 20.
6] C. Giesen, H. A. O. Wang, D. Schapiro, N. Zivanovic, A. Jacobs,
B. Hattendorf, P. J. Schꢁffler, D. Grolimund, J. M. Buhmann, S.
Brandt, Z. Varga, P. J. Wild, D. Gꢁnther, B. Bodenmiller, Nat.
Methods 2014, 11, 417 – 422.
[
1
25
122
were detected that presented with either higher Te, Te, or
near-equal quantities of both isotopes suggesting reoxygena-
tion, reduced oxygenation, or no change during the second
isotopologue-exposure phase, respectively (Figure 4g). Of
particular interest was our observation of a region bordering
necrosis that appeared to experience a dramatic increase in
the number of hypoxic cells (Figure 4h). We speculate that
proliferating cells displace cells close to the necrotic region,
[7] a) L. J. Edgar, R. N. Vellanki, A. Halupa, D. Hedley, B. G.
Wouters, M. Nitz, Angew. Chem. Int. Ed. 2014, 53, 11473 – 11477;
Angew. Chem. 2014, 126, 11657 – 11661; b) K. Tanabe, Z. Zhang,
T. Ito, H. Hatta, S. Nishimoto, Org. Biomol. Chem. 2007, 5,
3745 – 3757; c) R. A. McClelland, R. Panicucci, A. M. Rauth, J.
Am. Chem. Soc. 1985, 107, 1762 – 1763; d) J. L. Bolton, R. A.
McClelland, J. Am. Chem. Soc. 1989, 111, 8172 – 8181; e) R. J.
Hodgkiss, Anti-Cancer Drug Des. 1998, 13, 687 – 702.
[8] J. Russell, S. Carlin, S. A. Burke, B. Wen, K. M. Yang, C. C. Ling,
Int. J. Radiat. Oncol. Biol. Phys. 2009, 73(4), 1177 – 1186.
[9] H. Park, L. J. Edgar, M. A. Lumba, L. M. Willis, M. Nitz, Org.
Biomol. Chem. 2015, 13, 7027 – 7033.
10] N. Pham, J. M. M. M. Magalhaes, T. Do, J. Schwock, N. Dhani, P.
Cao, R. P. Hill, D. W. Hedley, Int. J. Cancer 2009, 124, 280 – 286.
forcing them into an O -deficient microenvironment, thereby
2
1
22
inducing hypoxia as reported by the Telox 2 isotopologue.
Our data suggests that without artificial O -manipulation
2
[
nearly equal quantities of cells become newly hypoxic and
newly normoxic over the time frame investigated (Figure 4j,
green histogram). Dynamic changes in oxygenation were also
observed in a patient-derived pancreatic cancer xenograft
model demonstrating the general applicability of this
approach to the study of different tumor types (Figure 4i).
Taken together, these data indicate that isotopologous
probes compatible with IMC can provide a general approach
for investigating dynamic biological phenomena with spatio-
temporal resolution without confounding effects from differ-
ential pharmacology of distinguishable probes. Telox 2 shows
promise for translation to the clinic to capture dynamic
oxygenation as a biomarker for better understanding cancer
biology and treatment.
[11] J. Su, Y. Gu, F. B. Pruijn, J. B. Smaill, A. V. Patterson, C. P. Guise,
W. R. Wilson, J. Biol. Chem. 2013, 288, 37138 – 37153.
[
12] K. L. Bennewith, J. A. Raleigh, R. E. Durand, Cancer Res. 2002,
2, 6827 – 6830.
13] C. J. Koch, Methods Enzymol. 2002, 352, 3 – 31.
14] R. H. Thomlinson, L. H. Grey, Br. J. Cancer 1955, 9, 539 – 549.
15] M. R. Horsman, M. Nordsmark, M. Hoyer, J. Overgaard, Br. J.
Cancer 1995, 72, 1474 – 1478.
6
[
[
[
Received: August 2, 2016
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Activity-Based Probes
L. J. Edgar, R. N. Vellanki, T. D. McKee,
D. Hedley, B. G. Wouters,
M. Nitz*
&&&& — &&&&
Isotopologous Organotellurium Probes
Reveal Dynamic Hypoxia In Vivo with
Cellular Resolution
Monitoring oxygenation: Incorporation of
isotopically enriched tellurium into a bio-
compatible tellurophene enabled con-
struction of a series of analytically dis-
tinguishable, but pharmacologically
identical activity-based probes sensitive
towards cellular hypoxia. Serial injection
of these isotopologous probes into
human tumor-bearing mice revealed cel-
lular hypoxia dynamics with spatiotem-
poral resolution as reported by imaging
mass cytometry.
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!