Nanoaggregate Probe for Breast Cancer Metastasis through Multispectral Optoacoustic Tomography and Aggregation-Induced NIR-I/II Fluorescence Imaging

Article abstract of DOI:10.1002/anie.201913149

An activatable nanoprobe for imaging breast cancer metastases through near infrared-I (NIR-I)/NIR-II fluorescence imaging and multispectral optoacoustic tomography (MSOT) imaging was designed. With a dihydroxanthene moiety serving as the electron donor, quinolinium as the electron acceptor and nitrobenzyloxydiphenylamino as the recognition element, the probe can specifically respond to nitroreductase and transform into an activated D-π-A structure with a NIR emission band extending beyond 900 nm. The activated nanoprobe exhibits NIR emission enhanced by aggregation-induced emission (AIE) and produces strong optoacoustic signal. The nanoprobe was used to detect and image metastases from the orthotopic breast tumors to lymph nodes and then to lung in two breast cancer mouse models. Moreover, the nanoprobe can monitor the treatment efficacy during chemotherapeutic course through fluorescence and MSOT imaging.

Full text of DOI:10.1002/anie.201913149

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Nanoaggregate for Detecting and Mapping Breast Cancer
Metastasis via Multispectral Optoacoustic Tomography and
Aggregation-induced Near-infrared Fluorescence-I and -II
Imaging
Authors: Juan Ouyang, Lihe Sun, Zhuo Zeng, Cheng Zeng, Fang
Zeng, and Shuizhu Wu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201913149
Angew. Chem. 10.1002/ange.201913149
Link to VoR: http://dx.doi.org/10.1002/anie.201913149
http://dx.doi.org/10.1002/ange.201913149

10.1002/anie.201913149
Angewandte Chemie International Edition
Nanoaggregate for Detecting and Mapping Breast Cancer Metastasis via Multispectral
Optoacoustic Tomography and Aggregation-induced Near-infrared Fluorescence-I and
-II Imaging
Juan Ouyang +, Lihe Sun+, Zhuo Zeng, Cheng Zeng, Fang Zeng* and Shuizhu Wu*
J. Ouyang +, L. H. Sun+, Z. Zeng, C. Zeng, Prof. F. Zeng, Prof. S. Z. Wu
State Key Laboratory of Luminescent Materials and Devices, Guangdong Provincial Key
Laboratory of Luminescence from Molecular Aggregates, College of Materials Science and
Engineering, South China University of Technology, Wushan Road 381, Guangzhou, 510640,
China
E-mail: mcfzeng@scut.edu.cn
shzhwu@scut.edu.cn
[+] These authors contributed equally to this work.
Supporting information and the ORCID identification number(s) for the author(s) of this article
can be found under http://doi.org/
1
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Abstract: An activatable nanoprobe for imaging breast cancer metastases via near infrared-I
(NIR-I) / NIR-II fluorescence imaging and multispectral optoacoustic tomography (MSOT)
imaging was designed. With dihydroxanthene moiety serving as the electron donor,
quinolinium as the electron acceptor and nitrobenzyloxydiphenylamino as the recognition
element, the probe can specifically respond to nitroreductase and transform into an activated
D--A structure with a NIR emission band extending beyond 900 nm. The activated nanoprobe
exhibits NIR emission enhanced by aggregation-induced emission (AIE) effect and produces
strong optoacoustic signal. The nanoprobe can detect and image sequential metastases from
the orthotopic breast tumors to lymph nodes and then to lung in two breast cancer mouse
models. Moreover, the nanoprobe can monitor and track the treatment efficacy during
chemotherapeutic course via NIR fluorescence and MSOT imaging. 3D MSOT images enable
us to locate and map the tumor metastases to lymph nodes and lung in a spatiotemporal
manner.
Introduction
Breast cancer has become the most common type of cancer and the leading cause of
cancer-related death for women, which affect up to one in eight women worldwide.[1,2]
For breast cancer, in early stages the invasive cancer is contained in the breast. If left
untreated, it would grow and progress to more advanced stages, at which points breast
cancer may spread to lymph nodes and then even invades distant organs such as
lungs.[3-7] If treated before metastasis occurs, breast cancer is considered a manageable
disease and more likely curable.[8,9] However, once metastasis is established, the
treatment response to the same treatment strategy becomes dismal, and over 90% of the
breast cancer mortalities are caused by metastasis.[10-14] Hence, the importance of
accurate and early detection for the metastasis of breast cancer has been widely
recognized, given that it can help plan optimized treatment and increase survival
rate.[15,16]
Currently, some medical imaging techniques are used for detection and diagnosis of
breast cancer.[17-19] However, accurate detection for breast cancer metastasis and
mapping of its metastasis remain challenging.[20,21] The examination for metastasis is
still a time-consuming and error-prone process, since the metastasized tumors are
usually of small size with various shapes at early stage of metastasis, and usually for the
tumors there are low contrast difference and unclear boundary with surrounding normal
tissues.[20,21] Furthermore, the metastatic routes and sites are unknown at early stages
and need to be pinpointed, so as to provide crucial information for precision treatment.
Optoacoustic imaging is a rapidly developing and non-invasive imaging technique
with great potential for preclinical and clinical applications.[22-26] Optoacoustic imaging
exhibits some attractive advantages of both optical imaging and ultrasound imaging,
such as high-contrast and spectroscopy-based specificity, reduced tissue scattering and
the larger penetration depth.[27] As a functional optoacoustic imaging method, the
2
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
multispectral optoacoustic tomography (MSOT) operates via irradiating a sample with
multiple wavelengths and detecting ultrasound signals generated by different
photoabsorbers in the sample.[28] Through spectral unmixing, MSOT can separately
visualize different photoabsorbers including endogenous ones such as hemoglobin, as
well as the exogenous optoacoustic probes in the target tissue.[29,30] Moreover, MSOT
can readily afford three-dimensional (3D) images by rendering a stack of cross-sectional
images as the maximum intensity projection (MIP) images.[31,32]
Another way to reduce tissue scattering and autofluorescence is the application of
near-infrared II (NIR-II) dyes in the biological imaging, [33-35] as NIR-II fluorescence
imaging can overcome the penetration/contrast bottleneck which is unsolvable by
fluorescence imaging within visible region.[33a,33b] However, compared to inorganic
NIR-II nano-fluorophores, the organic NIR-II dyes are relatively less despite of their
advantages such as flexibility in structural design and ease of spectral property tuning.
Interestingly, recent studies have found some NIR-I dyes show long emission tails that
stretch into the NIR-II region and can be utilized for NIR-II imaging. This may open a
new route for NIR-II imaging.[35]
Activatable (or responsive) probes are more advantageous than those inert ones
which only produce unchangeable fluorescent or optoacoustic signal.[36] The activatable
probes can respond specifically to the biomarkers of diseases and correspondingly
generate changed signals, thereby allowing disease-specific detection and diagnosis.[36-39]
To achieve high-quality optoacoustic and NIR (especially NIR-II) imaging, a probe
should has some congenial optical properties, including near-infrared absorption with
high molar extinction coefficient as well as strong fluorescence emission band at or
extended to long wavelength region. However, the activatable optoacoustic probes and
NIR-II fluorescence probes are still limited and thus far there is scarcely any activatable
NIR-II or activatable optoacoustic probe for detecting and mapping the breast cancer
metastasis. The electron donor--electron acceptor (D--A) structure is an effective
approach to manipulate the spectral properties of a molecular chromophore in long
wavelength.[40] Introducing the substituents with suitable electron-donating or
electron-accepting capabilities into the chromophores would lead to long-wavelength
absorption and emission. Moreover, analyte-triggered variation of electron-donating (or
electron-accepting) capability of the donor (or acceptor) substituent would be conducive
to achieving responsive changes of spectral property of chromophore.
Endeavoring toward developing an activatable and dual-mode probe for detecting
and tracking breast cancer metastasis optoacoustically and fluorescently, we designed a
probe which is specifically responsive toward nitroreductase, a biomarker highly
expressed in hypoxia microenvironment at solid tumors including breast tumor.[41-43] In
this probe (Figure 1), a hydrophobic dihydroxanthene moiety serves as the strong
electron donor (D) and a hydrophilic quinolinium as the electron acceptor (A). A
nitrobenzyloxydiphenylamino group is linked to the donor moiety and serves as the
recognition element for nitroreductase as well as the electron-accepting substituent that
quenches the fluorescence. The probe specifically responds to nitroreductase and thus
the electron-withdrawing group turns into the electron-donating one; subsequently the
3
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
activated probe exhibits aggregation-induced emission (AIE) feature and produces
strong NIR fluorescence and optoacoustic signals. With AIE-active nature, the activated
probe would emit in aggregated state in aqueous media more strongly than in
molecularly dissolved state in good solvents.[44-47] Therefore, the probe can function as a
dual-mode imaging probe and further allow mutual attestation between the two imaging
methods.
The synthesized chromophore (Q-NO2) readily forms the nanoprobe (NP-Q-NO2) in
aqueous media owing to its amphiphilic molecular structure. Moreover, due to the
existence of the nitrobenzyloxy and triarylamine moieties, the probe formed relatively
loosely packed nanoaggregates in water media, which would be conducive to the
nitroreductase-mediated enzymatic reaction. As shown in Figure 1, in the absence of
nitroreductase, the nanoprobe is neither absorptive nor fluorescent in NIR region.
Nitroreductase transforms the strong electron-withdrawing nitro group to amino group,
which induces self-elimination reaction and yields an electron-donating hydroxyl group,
thus generating the activated probe with strong optoacoustic and fluorescence signals.
Our results indicate that, the nanoprobe can respond to nitroreductase in two mouse
models of breast tumor metastasis, and image the sequential metastases from the
orthotopic breast tumors to lymph nodes and then to lung (Figure 1(II)). Moreover, the
nanoprobe can also be employed to monitor and track the tumor metastasis suppression
during a chemotherapeutic course. In addition, the 3D MSOT images enable us to locate
the focus of orthotopic breast cancer tumors as well as to map the metastases of tumor to
lymph nodes and lung in a spatiotemporal manner.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Figure 1. (I) Schematic display of the nanoprobe’s response to nitroreductase in tumor hypoxia.
(II) Illustration showing two scenarios of breast tumor metastasis. a: Metastasis from the
orthotopic tumors at the 3rd pair of mammary fat pads (MFPs) to subiliac lymph nodes (LNs),
axillary lympha nodes and lungs. b: Metastasis from the 5th pair of mammary fat pads to
subiliac lymph nodes and lungs.
Results and discussion
The molecular probe Q-NO2 was synthesized according to the route shown in Scheme S1.
The structure of Q-NO2, the activated probe Q-OH and intermediates were characterized by 1H
5
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
NMR, 13C NMR and mass spectrometry (Figures S1-S11). The nanoprobe NP-Q-NO2 was
prepared by dispersing DMSO solution of Q-NO2 into pH 7.4 PBS. The Q-NO2 molecules
readily formed nanoaggregates (NP-Q-NO2) in aqueous media due to their amphiphilic
structure. The transmission electronic microscopy (TEM) images of the nanoaggregate
(nanoprobe) and the size distribution measured by dynamic light scattering method are shown
in Figures S12-S13.
The AIE feature of the activated probe Q-OH and the relevant elements triarylamine and
quinolinium was investigated by separately dissolving them in mixtures of poor solvent (PBS)
and good solvent (DMSO), followed by measurement of fluorescence spectra, which are shown
in Figures S14-S15. In good solvent, fluorescence for the three compounds is weak; while it
gradually increases with the increasing ratio of poor solvent, indicating that these compounds
are all AIE active, and the activated nanoprobe is suitable for imaging. These results also
suggest that, both the triarylamine and the quinolinium moieties contribute to the AIE feature
of Q-OH, we suppose the triarylamine moiety plays the key role, since it has more rotatable
bonds in its structure.
The absorption and fluorescence spectra for the nanoprobe NP-Q-NO2 upon incubation
with nitroreductase of varied levels were measured at 37 °C in pH 7.4 PBS containing 1%
DMSO (Figure 2). Upon incubation with nitroreductase, the absorbance of the nanoprobe at
around 664 nm increases with the increasing nitroreductase level (Figure 2A and Figure S16).
Particularly, the absorbance at around 808 nm in the absorption tail also enhances with the
nitroreductase level (Figure 2B and Figure S17), which suggests that the nanoprobe may be
applicable for NIR-II imaging, as the 808 nm laser is commonly used as the excitation light for
NIR-II imaging.[48-50] In the absence of nitroreductase, the nanoprobe is weakly emissive; and
incubation with nitroreductase gradually enhances the fluorescence intensities at 780 nm
(excited at 680 nm) and 922 nm (excited at 808 nm at 80 mW cm-2) (Figures S18-S19). The
dose-dependent fluorescence spectra for the nanoprobe were also recorded and shown in Figure
2C-2F, which demonstrates that with increasing nitroreductase activities, the fluorescence at
780 nm and 922 nm all enhances greatly. These results demonstrate that the nanoprobe
NP-Q-NO2 could be explored for both NIR-I and NIR-II imaging. The imaging quality would
exhibit great improvement when using emissions beyond 900 nm (NIR-II),[51,52] since in NIR-II
region there are diminished tissue autofluorescence and reduced photon scattering.[51-53] The
limit of detections of the probe toward nitroreductase is determined as 0.052 g mL-1 (3
/slope with Figure 2D).
The optoacoustic (OA) intensities in the wavelength range from 680 nm to 900 nm for
NP-Q-NO2 upon incubation with nitroreductase are shown in Figure 2G, which suggests that
the activated nanoprobe is also suitable for optoacoustic imaging. Figure 2H reveals that, the
relative optoacoustic intensity for the nanoprobe upon incubation with nitroreductase enhances
with the increasing nitroreductase activity. Moreover, the nanoprobe displays good selectivity
toward nitroreductase over a variety of potential interferents including amino acids, ions and
some common bio-substances (Figures S20-S22).
To verify the transformation from Q-NO2 into Q-OH upon reaction with nitroreductase, the
absorption and emission spectra for NP-Q-NO2 and Q-OH nanoaggregate upon reaction with
nitroreductase were compared (Figures S23-S24). Clearly, the absorption and emission spectra
6
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
of the reaction product are quite similar to those for the Q-OH nanoaggregate, proving the
structural transformation of Q-NO2 upon incubation with nitroreductase. The emission of Q-OH
was found stable in the pH range of 5.8 – 8.0 (Figure S25). Furthermore, high resolution mass
spectrometry (HRMS) was performed to verify the generation of Q-OH. In the HRMS
spectrum (Figure S26), the peak at m/z 684.2866 arises from Q-NO2, while the peak at m/z
549.2553 arises from Q-OH. These results confirm that Q-OH is indeed generated after the
reaction between Q-NO2 and nitroreductase. Our observations also show that upon treatment
with nitroreductase, the nanosized NP-Q-NO2 transforms into another nanosized product
(NP-Q-OH) (Figures S12-S13), implying that the nanoprobe is formed by the loosely packed
molecular probe in aqueous media and this favors the nitroreductase-mediated reaction of the
probe.
Figure 2. Spectral properties of the nanoprobe NP-Q-NO2 in pH 7.4 PBS (containing 1%
DMSO). (A) Absorption spectra (in 500 - 900 nm) of NP-Q-NO2 (10 M) upon incubation with
varied levels of nitroreductase. (B) Absorption spectra (in 780 - 950 nm) of NP-Q-NO2 (10 M)
upon incubation with varied levels of nitroreductase. (C) NIR-I Fluorescence spectra (excited at
680 nm) for NP-Q-NO2 (10 M) upon incubation with varied levels of nitroreductase. (D)
Fluorescence intensity at 780 nm for NP-Q-NO2 (10 M) as a function of nitroreductase level
(n = 3). The equation of the least square regression line is y = 20.06 + 10.67 x. The inset shows
representative NIR-I fluorescent images of the nanoprobe after incubation with varied levels of
nitroreductase. (E) NIR-II Fluorescence spectra (excited at 808 nm laser, 80 mW cm-2) for
NP-Q-NO2 (10 M) upon incubation with varied levels of nitroreductase. (F) Fluorescence
intensity at 922 nm for the nanoprobe (10 M) as a function of the nitroreductase level (n = 3).
The equation of the least square regression line is y = 321.07 + 52.50 x. The inset shows
representative NIR-II fluorescent images of the nanoprobe after incubation with varied levels of
nitroreductase. (G) Normalized optoacoustic intensities for the nanoprobe (10 M) upon
incubation with 20 g mL-1 nitroreductase (for 10 min) as a function of excitation wavelength.
(H) Relative optoacoustic intensity [(I-I0)/I0] for the nanoprobe (10 M) upon incubation with
varied levels of nitroreductase (n = 3). The inset displays representative optoacoustic images of
the nanoprobe in phantom after incubation with varied levels of nitroreductase. Excitation
wavelength: 680 nm.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Before in vivo imaging experiments, the cytotoxicity of the nanoprobe was tested by MTT
assay. As displayed in Figure S27, NP-Q-NO2 shows low cytotoxicity against 4T1 cells.
Afterwards, the nanoprobe was employed to image endogenous nitroreductase in 4T1 cells (the
murine mammary carcinoma cell line, which is known to overexpress nitroreductase under
hypoxia).[54-57] From Figure S28, one can see that the cells incubated with the nanoprobe in
hypoxic atmosphere for 0.5 h exhibited evident fluorescence, either under the excitation of 675
nm LED light or 808 nm laser light (at 50 mW cm-2). While for the cells incubated in normoxia,
almost no intracellular fluorescence could be observed. Optoacoustic imaging of endogenous
nitroreductase in 4T1 cells incubated with the nanoprobe was also performed, as shown in
Figure S28. Obviously, after the nanoprobe was incubated with cells under hypoxia, significant
OA signals could be observed in the cells; while the opotoacoustic signal was insignificant in
the cells under normoxia. These experimental data indicate that the nanoprobe NP-Q-NO2 can
respond to the intracellular nitroreductase and generate fluorescent and optoacoustic signals.
Moreover, the biosafety of the nanoprobe was evaluated by comparing the histology sections
(H&E staining) of organs collected from the control and the mice intravenously injected with
the nanoprobe, and the blood retention of NP-Q-NO2 was measured as well (Figure S29). As
shown in Figure S29, there was no obvious histopathological difference between the sections of
the nanoprobe-treated mice and the control, indicating that the nanoprobe NP-Q-NO2 is of good
biosafety
.
Next, we used the nanoprobe to detect the orthotopic and the metastatic tumors via NIR-I
and NIR-II imaging. We established two orthotopic breast tumor models by injecting the
luciferase-expressing 4T1 cells into the 3rd or 5th pair of mammary fat pads of the mice
respectively, so as to diagnose and image both the regional and distant metastasis from the
primary site at mammary fat pads. The two mouse models are referred to as “the 3rd pair model”
and “the 5th pair model” respectively. Luciferase bioluminescence in tumor cells allows us to
locate orthotopic and metastatic tumors via injecting luciferin substrate (15 mg kg-1) and then
detecting the luminescence as the result of enzymatic reaction between luciferase and its
substrate (D-luciferin),[58-60] and the resultant bioluminescence images can be used to verify our
fluorescence imaging results.
NIR-I and NIR-II imaging for the mice injected with 4T1 cells at the 5th pair of mammary
fat pads (the 5th pair model) were performed first. The distances between the 5th pair of
mammary fat pads and the possible organs of metastasis (subiliac lymph nodes and lungs) are
relatively long (Figure S30), this mouse model was established to evaluate our nanoprobe’s
capability of imaging distant metastasis. As shown in Figure 3, at day 8 upon injection of
luciferase-expressing 4T1 cells into the 5th pair of mammary fat pads, bioluminescence signals
could be observed at the fat pads, indicating the formation of orthotopic tumor at the fat pads.
Afterward, weak bioluminescence signals could be identified at the subiliac lymph nodes at day
14 and day 20, but no signal could be observed in the lung (Figure 3A). The wavelength of the
bioluminescence produced by D-luciferin (upon reaction with luciferase) is below 600ꢀnm,
which is not suitable for in vivo imaging [61,62] as the absorption and scattering of light by tissue
(scaling with λ-α, α is up to 4 for tissues) reduces the imaging resolution and depth.[36,37,61-64]
8
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Figure 3A also shows that, the NIR-I and NIR-II fluorescence images are basically in line
with the bioluminescence ones, verifying the nanoprobe is capable of diagnosing and imaging
the orthotopic tumors and distant metastatic tumors. Compared to bioluminescence imaging,
the NIR-I and NIR-II fluorescence are more prominent. At day 20, obvious fluorescence
signals could be observed at the orthotopic tumors as well as at the lymph nodes and the lung
(Figure 3A and 3C). Moreover, the NIR-II imaging could afford more distinct signals in lymph
nodes and lung (Figure 3A and 3D), suggesting that the NIR-II fluorescence facilitates higher
resolution for in vivo imaging, which is due to its reduced photo scattering and minimal tissue
autofluorescence levels. The images of the mice were also obtained under tungsten halogen
light (wavelength range 400 – 850 nm), and it is clear that both fluorescence and bright field
images can be observed simultaneously using the halogen lamp as the light source and they
further verify the capability of the nanoprobe for imaging orthotopic breast tumors and the
metastases.
To evaluate the nanoprobe’s capability to track and monitor the therapeutic effects,
cyclophosphamide (CPA), which is a commonly-used antitumor drug for a variety of
cancers including breast cancer,[65-67] was used to treat the breast cancer and suppress the
metastases from the orthotopic breast tumors to other organs. At day 4 after the injection
of luciferase-expressing 4T1 cancer cells, cyclophosphamide was i.v. injected into the
mice and the treatment was repeated once every other day, and at day 20 the mice were
subject to bioluminescence imaging and fluorescence imaging. From Figure 3B, it is
obvious that, for the mice without treatment, overgrown tumors could be seen at the 5th
pair of mammary fat pads at day 20 upon injection of cancer cells into the mammary fat
pads. In contrast, for the mice receiving chemotherapy, no metastases could be observed
and the orthotopic tumors were still constrained in the mammary fat pads, and at day 20
the orthotopic tumors are much smaller compared to those of the control group which
did not receive treatment (Figure 3B, 3E and 3F). In addition, the ex vivo imaging of the
major organs and the tumors (orthotopic tumors at the mammary fat pads, metastatic
tumors at the lymph nodes and the lung) were performed after the mice were sacrificed,
and the resultant images are shown in Figure S31, which tally well with the in vivo
fluorescence imaging results. These experimental data support the idea that NP-Q-NO2
can serve as an ideal probe for tracking therapeutic effects via NIR-I and NIR-II
fluorescence imaging.
9
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Figure 3. Imaging of tumor metastasis from breast tumors at 5th pair of mammary fat pads. (A)
Representative bioluminescence images, NIR-I and NIR-II fluorescence images and
photographs under halogen light of the mice at different days after 4T1 cancer cell injection
into the 5th pair of mammary fat pads. For NIR-I fluorescence imaging: excitation filter: 675
nm, emission filter: 790 nm. For NIR-II fluorescence imaging: excitation: 808 nm laser at 50
mW cm-2, emission filter: > 900 nm; for halogen light: 400 – 850 nm. In the case of
fluorescence imaging, the imaging was performed 30 min after nanoprobe injection. (B)
10
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Representative bioluminescence images and fluorescence images at day 4 and day 20 after 4T1
cell injection for the mice receiving chemotherapy treatment (starting at day 4 and repeated
every other day) and for the mice without treatment (the control). (C) Mean NIR-I fluorescence
intensities for the 5th pair of mammary fat pads, the subiliac lymph nodes and the lung for the
healthy mice (day 0) and the tumor-bearing mice at different days after 4T1 cell injection. (n =
5). (D) Mean NIR-II fluorescence intensities of the 5th pair of mammary fat pads, the subiliac
lymph nodes and the lung for the healthy mice (day 0) and the tumor-bearing mice at different
days after 4T1 cell injection (n = 5). (E) Mean NIR-I fluorescence intensities of the 5th pair of
mammary fat pads, the subiliac lymph nodes and the lung at day 4 and day 20 after 4T1 cell
injection for the mice receiving chemotherapy treatment and for the control. (n = 5). (F) Mean
NIR-II fluorescence intensities of the 5th pair of mammary fat pads, the subiliac lymph nodes
and the lung at day 4 and day 20 after 4T1 cell injection for the mice receiving chemotherapy
treatment and for the control. (n = 5) (G) Immunohistochemical (IHC) analysis for CD206
levels and hematoxylin-eosin (H&E) staining analysis of the tissue sections from the 5th pair of
mammary fat pads, subiliac lymph nodes and the lung from the healthy mice and the
tumor-bearing mice.
The histological analysis using hematoxylin-eosin (H&E) staining and
immunohistochemical analysis (IHC) for evaluating the CD206 levels for the sections from
orthotopic tumor, the metastasized tumor at the lymph nodes and lung were performed. CD206,
which is macrophages mannose receptor, plays vital roles in tumor proliferative activity,
invasion and metastasis, and high level of CD206 is closely associated with tumor invasion.[68]
As shown in Figure 3G, for healthy mice the CD206 level is normally expressed for the
sections of mammary fat pads, the lymph nodes and the lung. While for tumor-bearing mice,
CD206 is obviously overexpressed in those sections, indicating the tumor cells have
metastasized to lymph nodes and lung. H&E staining results show that abnormal morphology
formed in the tissues of the tumor-bearing mice. These results prove the above imaging results.
The fluorescence imaging for the mice with their orthotopic tumor at the 3rd pair of
mammary fat pads (the 3rd pair model) was also performed to evaluate the nanoprobe’s
capability of imaging the regional (short-distance) tumor metastasis. For a female mouse,
the 3rd pair of mammary fat pads reside at the thoracic region which is close to axillary
lymph nodes (Figure S30), similar to human mammary gland. As shown in Figure S32,
with the aid of the nanoprobe, we can detect the orthotopic tumors at the 3rd pair of
mammary fat pads, and the metastatic tumors at subiliac lymph nodes and even the
metastasis at lung. However, we cannot clearly discriminate the orthotopic tumors from
the metastatic tumors at axillary lymph nodes and lung, since there exists overlap of the
fluorescence signals from the orthotopic tumors and the metastasized tumors in the
lymph nodes and the lung, probably because of the short distance among them and the
projection-based nature of fluorescence imaging and light scattering by tissues. The ex
vivo NIR-I and NIR-II images for the dissected organs shown in Figure S33 reveal that,
both the axillary and subiliac lymph nodes have been invaded by breast cancer cells at
day 12.
11
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
To further verify the in vivo fluorescence imaging results, the mice were humanely
sacrificed, and part of their skins were removed and the pleura was opened for imaging
(Figure 4A, 4C and 4D). As shown in Figure 4A, the orthotopic tumors, the
metastasized tumors in lymph nodes and lung could be directly observed. With the skin
removed, the bioluminescence and fluorescence imaging at day 17 for the 3rd pair
model and at day 20 for the 5th pair model indicate that, in addition to signals at the
orthotopic tumor and lymph nodes, the signals at the lungs can be observed. The ex vivo
images and the intensity data for the dissected lungs (Figure 4B, Figures S34-S35) also
support this observation. From the bright-field photographs, obvious nodules and lesions
could be found in the lungs. These results also verify the fluorescence imaging results.
12
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Figure 4. (A) Representative images (bright-field photograph, bioluminescence, NIR-I and
NIR-II fluorescence images, and photographs under halogen light) for healthy mice (day 0) and
the tumor-bearing mice at day 17 (for the 3rd pair model) or at day 20 (for the 5th pair model)
with the part of the mice’ skin removed and pleura opened for better visualization. In blue
circles: orthotopic tumors; white circles: metastases at the lung; green circles: metastasized
tumors at lymph nodes. For NIR-I: excitation filter: 675 nm, emission filter: 790 nm. For
NIR-II: excitation: 808 nm laser at 50 mW cm-2. For halogen light: 400 - 850 nm. (B)
Representative bright-field photographs, bioluminescence images and NIR-I/NIR-II
fluorescence images for the lungs from the healthy mice (day 0) and the tumor-bearing mice of
the 3rd pair model or the 5th pair model. (C) Mean NIR-I fluorescence intensities for the 3rd
and 5th pair of mammary fat pads, the subiliac lymph nodes, the axillary lymph nodes and the
lung for the healthy mice (at day 0) and the mice at day 17 (for the 3rd pair model) or day 20
(for the 5th pair model) after cancer cell injection. (n = 5). (D) Mean NIR-II fluorescence
intensities for the 3rd and 5th pair of mammary fat pads, the subiliac lymph nodes, the axillary
lymph nodes and the lung for the healthy mice (at day 0) and the mice at day 17 (for the 3rd
pair model) or day 20 (for the 5th pair model) after cancer cell injection. (n = 5).
We then evaluated the nanoprobe’s capability of MSOT imaging for breast cancer metastases
on the two mice models, and the imaging results are presented in Figure 5 and Figures S36-S39.
On the MSOT system, we can obtain a cross-sectional MSOT image at a specific site of a
mouse. The cross-sectional MSOT images at the 3rd pair of mammary fat pads, the subiliac
lymph node and the lung are shown in Figure 5A and Figure S36; while the cross-sectional
images for the 5th pair model are displayed in Figure 5B and Figure S37. It is clear that, at day
0, there was no MSOT signal from the activated nanoprobe in all the cross sections (Figure 5A,
5B and Figures S36 and S37). While at day 6 (for the 3rd pair model) or at day 8 (for the 5th
pair model) after cancer cell injection, MSOT signals at the mammary fat pads began to emerge
(Figure S36 and Figure S37). Significant MSOT signals at the fat pads and weak signals at the
lymph nodes were visible at day 12 (for the 3rd pair model) or at day 16 (for the 5th pair model)
(Figure S36 and S37), indicating that tumor cells from the orthotopic tumors in the mammary
fat pads have invaded the lymph nodes. At day 17 or at day 20, MSOT signals could be
observed at the lungs, demonstrating that the lungs have been invaded by tumor cells. In
addition, both the healthy mice and the tumor-bearing mice were sacrificed at 0.5 h after the
nanoprobe injection, and the orthotopic tumors, the metastasized tumors and the main organs
were dissected for ex vivo MSOT imaging (Figures S38-S39). For the healthy mice, weak
MSOT signals can only be observed in some organs like liver. While for the tumor-bearing
mice, strong MSOT signals are found in various tumors.
MSOT technique presents important advantages in offering efficient visualization via
recording multiple two-dimensional tomographic images and rendering them as 3D maximal
intensity projection (MIP) images, which allow us to precisely locate and map biological events.
In this study, we obtained 3D MIP MSOT images for the two mice models. Figure 5C and 5E
show the representative whole-body (not including head and tail) 3D MSOT images at 17 day
(for the 3rd pair model) and at 20 day (for the 5th pair model) after 4T1 cell injection,
respectively. Figure 5D and 5F display the top-view images for the two models at different time
13
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
points after 4T1 cell injection. The detailed 3D images for the two mouse models at different
time points after 4T1 cell injection are shown in Figures S40 and S41. These MSOT images
can not only manifest the tumor metastasis from the orthotopic sites to lymph nodes and lungs,
but also allow us to locate the orthotopic and metastatic tumors in a temporospatial manner. In
addition, the MSOT signals of the activated nanoprobe in metabolic organs (such as liver) at
abdominal region are also visible. Importantly, compared to fluorescence imaging, the MSOT
provides images with higher spatial resolution. As shown in Figure 5D (day 17), the MSOT
signals of the axillary lymph nodes can be discriminated from the signals of the lungs and the
3rd pair of mammary pads which are in close proximity. The less scattering of acoustic waves
by tissues as well as the cross-sectional imaging achieved by tomographic approach all
contribute to the high spatial resolution of the MSOT images.
14
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Figure 5. MSOT imaging for tumor metastasis for the 3rd pair model (A, C and D) and the 5th
pair model (B, E and F). (A) Representative cross-sectional MSOT images corresponding to
different locations for the control (day 0) and the mice 17 days after cancer cell injection into
the 3rd pair of mammary fat pads. The upper panel of an image represents the MSOT signal
from the activated nanoprobe, while the lower panel (Merge) represents the overlay of the
activated nanoprobe’s signal with the grayscale background signal. The mouse lay on its
stomach with a certain degree of tilt, with a labeling 1 representing the spinal cord and
15
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
reflecting the extent of the tilt. (B) Representative cross-sectional MSOT images for the control
(day 0) and the mice at 20 days after cancer cell injection into the 5th pair of mammary fat pads.
(C) and (E) representative 3D MSOT images for the trunk (excluding head and tail) of a mouse
17 days (or 20 days) after cancer cell injection into the 3rd (or 5th) pair of mammary fat pads,
respectively. (D) and (F) Top-view MSOT images for the 3rd pair model (or the 5th pair model)
different days upon cancer cell injection. Labeling: 1. Spinal cord; 2. Orthotopic tumor; 3.
Lymph node; 4. Lung. The imaging was performed at 30 min upon the nanoprobe injection.
Scale bar: 5 mm.
MSOT imaging using the nanoprobe was then employed to track the therapy of breast tumor
and evaluate the metastases inhibition efficacy, and the imaging results are presented in Figure
6 and Figures S42-S45. At day 4 upon injection of 4T1 cells into the mammary fat pads,
cyclophosphamide was administrated via intravenous injection, and the therapy was repeated
every other day thereafter. The MSOT imaging was then performed at day 17 (for the 3rd pair
model) or at day 20 (for the 5th pair model). As displayed in Figure 6, Figure S42 and Figure
S43, for the mice group receiving chemotherapy treatment, almost no MSOT signals could be
observed at lymph nodes or the lungs, indicating the tumor metastases were largely suppressed
by chemotherapy, and the MSOT signals at orthotopic tumors after the therapeutic course were
much weaker compared to these for the mice without receiving therapeutic course. In
comparison, for the control, at the day 17 for the 3rd pair model, obvious metastases from the
orthotopic breast tumors to the axillary lymph nodes, the subiliac lymph nodes and the lung
could be clearly observed (Figures S42-S43); and at day 20 for the 5th pair model, the
metastases from the orthotopic breast tumors to the subiliac lymph nodes and the lung could be
found (Figure 6A-6B). These results support that, the nanoprobe NP-Q-NO2 is capable of
tracking and monitoring the therapeutic efficacy via MSOT imaging.
16
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Figure 6. (A) Representative cross-sectional MSOT images corresponding to the locations at
the 5th pair of mammary fat pads, the subiliac lymph nodes and the lung of the tumor-bearing
mice without or with chemotherapy treatment starting at day 4 (4 days after cancer cell
injection into the 5th pair of mammary fat pads) and repeated every other day. For the images
collected at a certain day: the upper panel represents the MSOT signal from the activated
nanoprobe; and the lower panel represents the overlay of the activated nanoprobe’s signal with
the grayscale background signal. The mouse lay on its stomach with a certain degree of tilt, and
the labeling 1 representing spinal cord reflects the extent of the tilt. (B) 3D MSOT images
covering the trunk of the tumor-bearing mice without or with chemotherapy treatment starting
at day 4 (4 days after cancer cell injection into the 5th pair of mammary fat pads) and repeated
every other day. For the images collected at a certain day: the left panel represents the signal
from the activated nanoprobe; and the right panel represents the overlay of the activated
nanoprobe’s signal with the grayscale background signal. White dotted circle: the 5th pair of
17
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
mammary fat pads, yellow dotted square: subiliac lymph nodes, magenta dotted circle: the lung.
The imaging was performed 30 min upon the nanoprobe injection. Scale bar: 5 mm.
Conclusion
In this study, an activatable nanoprobe with aggregation-induced emission has been
developed for detecting and imaging breast cancer metastases in two mice models via NIR-I
and NIR-II fluorescence imaging as well as multispectral optoacoustic tomography (MSOT)
imaging. By responding to nitroreductase overexpressed in tumor hypoxia, the nanoprobe
generates significant NIR-I/NIR-II signals and optoacoustic signals, thus presenting us the
whole picture of breast tumor metastasis. The fluorescence and MSOT imaging can detect both
the regional and distant breast cancer metastases originated from the orthotopic tumors. The
nanoprobe is also found capable of tracking the treatment efficacy of chemotherapy. The
nanoprobe may serve as a robust in-vivo imaging tool for monitoring breast cancer metastases
and correspondingly aid in in-depth researches and understanding on the metastasis process of
breast cancer. This strategy may provide insights for designing other activatable
optoacoustic/fluorescent probes for locating tumors and monitoring their metastases.
Acknowledgements
This work was supported by NSFC (21788102, 21875069, 51673066 and 21574044), the
Natural Science Foundation of Guangdong Province (2016A030312002) and the Fund of
Guangdong Provincial Key Laboratory of Luminescence from Molecular Aggregates
(2019B030301003).
Conflict of interest
The authors declare no conflict of interest.
Keywords: Nanoaggregate, Breast cancer metastasis, Multispectral optoacoustic tomography,
Near infrared II, Imaging
——————————
[1] A. Lafourcade, M. His, L. Baglietto, M. C. Boutron-Ruault, L. Dossus, V. Rondeau, BMC
Cancer, 2018, 18, 171.
[2] F. Caiazza, N. S. McCarthy, L. Young, A. D. Hill, B. J. Harvey, W. Thomas, Brit. J.
Cancer, 2011, 104, 338-344.
18
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
[3]
Https://www.cancer.org/cancer/breast-cancer/understanding-a-breast-cancer-diagnosis/stage
s-of-breast-cancer.html.
[4] C. B. Winston, O. Hadar, J. B. Teitcher, J. F. Caravelli, N. T. Sklarin, D. M. Panicek, L.
Liberman, Am. J. Roentgenol., 2000, 175, 795-800.
[5] G. N. Sharma, R. Dave, J. Sanadya, P. Sharma, K. K. Sharma, J. Adv. Pharm. Tech. Res.,
2010, 1, 109–126.
[6] R. L. Siegel, K. D. Miller, A. Jemal, CA: Cancer J. Clin., 2019, 69, 7-34.
[7] P. M. Peiris, R. Toy, E. Doolittle, J. Pansky, A. Abramowski, M. Tam, A. Mayer, ACS
Nano, 2012, 6, 8783-8795.
[8] J. Yan, Q. Huang, Cancer Metast. Rev., 2012, 31, 419-428.
[9] X. Guan, Acta Pharm. Sin. B, 2015, 5, 402-418.
[10] G. P. Gupta, J. Massagué, Cell, 2006, 127, 679-695.
[11] T. N. Seyfried, L. C. Huysentruyt, Crit. Rev. Oncogenesis, 2013, 18, 43-73.
[12] C. L. Chaffer, R. A. Weinberg, Science, 2011, 331, 1559-1564.
[13] F. Bertucci, C. K. Ng, A. Patsouris, N. Droin, S. Piscuoglio, N. Carbuccia, S. Garnier,
Nature, 2019, 569, 560.
[14] M. M. Obradović, B. Hamelin, N. Manevski, J. P. Couto, A. Sethi, M. M. Coissieux, M.
Bentires-Alj, Nature, 2019, 567, 540.
[15] D. Di Gioia, P. Stieber, G. P. Schmidt, D. Nagel, V. Heinemann, A. Baur-Melnyk, Brit. J.
Cancer, 2015, 112, 809-818.
[16] I. Ullah, G. M. Karthik, A. Alkodsi, U. Kjällquist, G. Stålhammar, J. Lövrot, J. Bergh, J.
Clin. Invest., 2018, 128, 1355-1370.
[17] A. Karellas, S. Vedantham, Med. Phys., 2008, 35, 4878-4897.
[18] R. N. Strickland, H. I. Hahn, Med. Imaging, 1996, 15, 218-229.
[19] E. García, Y. Diez, O. Diaz, X. Lladó, A. Gubern-Mérida, R. Martí, A. Oliver, Med. Image
Anal., 2019, 54, 76-87.
[20] E. J. Schneble, L. J. Graham, M. P. Shupe, F. L. Flynt, K. P. Banks, A. D. Kirkpatrick, I.
Avital, Brit. J. Cancer, 2014, 5, 281-290.
[21] M. E. Menezes, S. K. Das, I. Minn, L. Emdad, X. Y. Wang, D. Sarkar, P. B. Fisher, Adv.
Cancer Res., 2016, 132, 1-44.
[22] Y. Zhang, M. Jeon. L. J. Rich, H. Hong, J. Geng, Y. Zhang, S. Shi, T. E. Barnhart, P.
Alexandridis, J. D. Huizinga, M. Seshadri. W. Cai, C. Kim, J. F. Lovell, Nature Nanotech.
2014, 9, 631-638.
[23] a) L. Feng, M. Gao, D. Tao, Q. Chen, H. Wang, Z. Dong, M. Chen, Z. Liu, Adv. Funct.
Mater. 2016, 26, 2207-2217; b) Y. Wu, J. Chen, L. Sun, F. Zeng, S. Wu, Adv. Funct. Mater.,
2019, 29, 1807960.
[24] J. Zhou, Y. Jiang, S. Hou, P. K. Upputuri, D. Wu, J. Li, H. Duan, ACS Nano, 2018, 12,
2643-2651.
[25] R. Ma, A. Taruttis, V. Ntziachristos, D. Razansky, Opt. Express, 2009, 17, 21414-21426.
[26] L. V. Wang, S. Hu, Science, 2012, 335, 1458-1462.
[27] a) V. Neuschmelting, K. Kim, J. Malekzadeh-Najafabadi, S. Jebiwott, J. Prakash, A.
Scherz, V. Ntziachristos, Theranostics, 2018, 8, 723-734; b) Z. Wang, X. Zhen, P. K.
19
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
Upputuri, Y. Jiang, J. Lau, M. Pramanik, B. Xing, ACS Nano, 2019, 13, 5816−5825; c) Y.
Wu, L. Sun, F. Zeng, S. Wu, Photoacoustics, 2019, 13, 6-17.
[28] a) D. Razansky, A. Buehler, V. Ntziachristos, Volumetric real-time multispectral
optoacoustic tomography of biomarkers, Nature Protocols. 2011, 6, 1121-1129; b) J. Huang,
Y. Wu, F. Zeng, S. Wu, Theranostics, 2019, 9, 7313-7324.
[29] V. Ntziachristos, D. Razansky, Chem. Rev. 2010, 110, 2783-2794.
[30] S. Zackrisson, S. M. W. Y. van de Ven, S. S. Gambhir, Cancer Res. 2014, 74, 979-1003.
[31] S. Tzoumas, N. C. Deliolanis, S. Morscher, V. Ntziachristos, IEEE Trans. Med. Imaging,
2014, 33, 48-60.
[32] S. Tzoumas, A. Kravtsiv, Y. Gao, A. Buehler, V. Ntziachristos, IEEE Trans. Med.
Imaging, 2016, 35, 2534-2545.
[33] a) S. Zhu, R. Tian, A. L. Antaris, X. Y. Chen, H. J. Dai, Adv. Mater., 2019, 31, 1900321; b)
S. Zhu, B. C. Yung, S. Chandra, G. Niu, A. L. Antaris, X. Chen, Theranostics, 2018, 8,
4141-4151; c) Q. Miao, Y. Lyu, D. Ding, K. Y. Pu, Adv. Mater., 2016, 28, 3662-3668; d) Y.
Wu, S. Huang, J. Wang, L. Sun, F. Zeng, S. Wu, Nat. Commun., 2018, 9, 3983; e) C. Li, L.
Cao, Y. Zhang, P. Yi, M. Wang, B. Tan, Z. Deng, D. Wu, Q. B. Wang, Small, 2015, 11,
4517-4525; f) X. Hao, Y. Zhang, H. Wang, G. Chen, M. Wang, Q. B. Wang, Adv. Mater.,
2018, 30, 1804437.
[34] a) X. Zhen, J. Zhang, J. Huang, C. Xie, Q. Miao, K. Y. Pu, Angew. Chem. Int. Ed., 2018,
57, 7804-7808; Angew. Chem., 2018, 130, 7930-7934; b) L. Sun, Y. Wu, J. Chen, J. Zhong,
F. Zeng, S. Wu, Theranostics, 2019, 9, 77-89.
[35] a) Q. Chen, C. Liang, X. Q. Sun, J. W. Chen, Z. J. Yang, H. Zhao, L. Z. Feng, Z. Liu,
PNAS 2017, 114, 5343-5348; b) C. Zhan, Y. Huang, G. Lin, S. Huang, F. Zeng, S. Wu,
Small, 2019, 1900309;
[36] a) V. Gujrati, A. Mishra, V. Ntziachristos, Chem. Commun., 2017, 53, 4653-4672; b) Y.
Lyu, K. Y. Pu, Adv. Sci., 2017, 4, 1600481.
[37] a) X. Jiang, B. Du, S. Tang, E. Hsieh, J. Zheng, Angew. Chem. Int. Ed., 2019, 58,
5994-6000; Angew. Chem., 2019, 131, 6055-6061; b) C. Yu, X. Li, F. Zeng, F. Zheng, S.
Wu, Chem. Commun., 2013, 49, 403-405; c) Y. Qi, Y. Huang, B. Li, F. Zeng, S. Wu, Anal.
Chem., 2018, 90, 1014-1020; d) G. Hong, A. L. Antaris, H. Dai, Nat. Bioimed. Eng., 2017, 1,
0010.
[38] a) K. Harmatys, J. Chen, D. M. Charron, C. M. MacLaughlin, G. Zheng, Angew. Chem. Int.
Ed., 2018, 57, 8125-8129; Angew. Chem., 2018, 130, 8257-8261; b) B. Li, X. Xie, Z. Chen,
C. Zhan, F. Zeng, S. Wu, Adv. Funct. Mater., 2018, 28, 1800692; c) K. Gu, W. Qiu, Z. Q.
Guo, C. Yan, S. Zhu, D. Yao, P. Shi, H. Tian, W. H. Zhu, Chem. Sci., 2019, 10, 398-405; d)
C. Yan, Z. Q. Guo, Y. Liu, P. Shi, H. Tian, W. H. Zhu, Chem. Sci., 2018, 9, 6176-6182.
[39] a) Z. Yang, W. Fan, W. Tang, Z. Shen, Y. Dai, J. Song, Z. Wang, Y. Liu, L. Lin, L. Shan,
Y. Liu, O. Jacobson, P. Rong,W. Wang, X. Y. Chen, Angew. Chem. Int. Ed., 2018, 57,
14101-14105; Angew. Chem., 2018, 130, 14297-14301; b) D. Lee, J. Kim, Y. Lee, S. Lee,
W. Miao, H. Kim, J. Min, S. Jon, Angew. Chem. Int. Ed., 2017, 56, 13684-13688; Angew.
Chem., 2017, 129, 13872-13876; c) Y. Wu, J. Wang, F. Zeng, S. Huang, J. Huang, H. Xie, C.
Yu, S. Wu, ACS Appl. Mater. Interfaces, 2016, 8, 1511-1519.
20
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
[40] a) R. Ni, M. Vaas, W. Ren, J. Klohs, Neurophotonics, 2018, 5, 015005; b) R. T. Hayes, M.
R. Wasielewski, D. Gosztola, J. Am. Chem. Soc., 2000, 122, 5563-5567.
[41] Y. Li, Y. Sun, J. Li, Q. Su, W. Yuan, Y. Dai, F. Li, J. Am. Chem. Soc., 2015, 137,
6407-6416.
[42] X. Zheng, H. Tang, C. Xie, J. Zhang, W. Wu, X. Jiang, Angew. Chem. Int. Ed., 2015, 54,
8094-8099; Angew. Chem., 2015, 127, 8212-8217.
[43] J. N. Liu, W. Bu, J. Shi, Chem. Rev., 2017, 117, 6160-6224.
[44] S. Sreejith, K. P. Divya, A. Ajayaghosh, Angew. Chem. Int. Ed., 2008, 47, 7883-7887;
Angew. Chem. 2008, 120, 8001-8005.
[45] a) D. Ding, D. Mao, K. Li, X. Wang, W. Qin, R. Liu, D. S. Chiam, N. Tomczak, Z. Yang,
B. Z. Tang, D. Kong, B. Liu, ACS Nano, 2014, 8, 12620-12631; b) W. Fu, C. Yan, Z. Q.
Guo, J. Zhang, H. Zhang, H. Tian, W. H. Zhu, J. Am. Chem. Soc., 2019, 141, 3171-3177; c)
J. Zhang, Q. Wang, Z. Q. Guo, S. Zhang, C. Yan, H. Tian, W. H. Zhu, Adv. Funct. Mater.,
2019, 29, 1808153.
[46] a) S. Wang, W. Wu, P. Manghnani, S. Xu, Y. Wang, C. C. Goh, L. G. Ng, B. Liu, ACS
Nano, 2019, 13, 3095-3105; b) C. Zhang, G. Feng, S. Xu, Z. Zhu, X. Lu, J. Wu, B. Liu,
Angew. Chem. Int. Ed., 2016, 55, 6192-6196; Angew. Chem., 2016, 128, 6300-6304; c) Y.
Chen, W. Zhang, Z. Zhao, Y. Cai, J. Gong, T. Kwok, J. Lam, H. Sung, I. Williams, B. Z.
Tang, Angew. Chem. Int. Ed., 2018, 57, 5011-5015; Angew. Chem., 2018, 130, 5105-5109.
[47] a) F. Hu, D. Mao, Kenry, X. Cai, W. Wu, D. Kong, B. Liu, Angew. Chem. Int. Ed., 2018,
57, 10182-10186; Angew. Chem., 130, 10339-10343; b) Q. Q. Li, Y. H. Tang, W. Hu, Z. Li,
Small, 2018, 14, 1801560.
[48] G. Hong, J. C. Lee, J. T. Robinson, U. Raaz, L. Xie, N. F. Huang, H. Dai, Nat. Med., 2012,
18, 1841-1846.
[49] Y. Cai, Z. Wei, C. Song, C. Tang, W. Han, X. Dong, Chem. Soc. Rev., 2019, 48, 22-37.
[50] Z. Lei, C. Sun, P. Pei, S. Wang, D. Li, X. Zhang, F. Zhang, Angew. Chem. Int. Ed., 2019,
58, 8166-8171; Angew. Chem., 2019, 131, 8250-8255 .
[51] M. Wang, N. Chen, J. Biophotonics, 2019, 12, e201800459.
[52] Z. Feng, X. M. Yu, M. X. Jiang, L. Zhu, Y. Zhang, W. Yang, W. Xi, G. H. Li, J. Qian,
Theranostics, 2019, 9, 5706-5719.
[53] B. N. Giepmans, S. R. Adams, M. H. Ellisman, R. Y. Tsien, Science, 2016, 312, 217-224.
[54] Z. Liu, F. Song, W. Shi, G. Gurzadyan, H. Yin, B. Song, X. Peng, ACS Appl. Mater.
Interfaces, 2019, 11, 15426-15435.
[55] P. Feng, H. Zhang, Q. Deng, W. Liu, L. Yang, G. Li, M. Li, Anal. Chem., 2016, 88,
5610-5614.
[56] K. H. Gebremedhin, Y. Li, Q. Yao, M. Xiao, F. Gao, J. Fan, J. Du, S. Long, X. Peng, J.
Mater. Chem. B, 2019, 7, 408-414.
[57] X. Tian, Z. Li, Y. Sun, P. Wang, H. Ma, Anal. Chem., 2018, 90, 13759-13766.
[58] D. M. Mofford, G. R. Reddy, S. C. Miller, J. Am. Chem. Soc., 2014, 136, 13277-13282.
[59] T. Nakatsu, S. Ichiyama, J. Hiratake, A. Saldanha, N. Kobashi, K. Sakata, H. Kato, Nature,
2006, 440, 372.
[60] J. C. Tiffen, C. G. Bailey, C. Ng, J. E. Rasko, J. Holst, Mol. Cancer, 2010, 9, 299.
21
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
[61] T. Kuchimaru, S. Iwano, M. Kiyama, S. Mitsumata, T. Kadonosono, H. Niwa, S.
Kizaka-Kondoh, Nat. Commun., 2016, 7, 11856.
[62] R. Weissleder, V. Ntziachristos, Nat. Med., 2003, 9, 123-128.
[63] K. Welsher, S. P. Sherlock, H. Dai, Proc. Natl. Acad. Sci. USA, 2011, 108, 8943-8948.
[64] R. S. Negrin, C. H. Contag, Nat. Rev. Immunol., 2006, 6, 484-490.
[65] E. J. Bae, D. A. Solimando Jr, J. A. Waddell, Hosp. Pharm., 2009, 44, 26-33.
[66] M. B. Jensen, A. V. Lænkholm, T. O. Nielsen, J. O. Eriksen, P. Wehn, T. Hood, B. Breast
Cancer Res., 2018, 20, 79.
[67] B. Peláez, J. A. Campillo, J. A. López-Asenjo, J. L. Subiza, J. Immunol., 2001, 166,
6608-6615.
[68] a) J. G. Quatromoni, E. Eruslanov, Am. J. Transl. Res., 2012, 4, 376-389; b) P. Allavena,
M. Chieppa, G. Bianchi, G. Solinas, M. Fabbri, G. Laskarin, A. Mantovani, Clin. Dev.
Immunol., 2010, 2010, 547179; c) A. Mantovani, B. Bottazzi, F. Colotta, S. Sozzani, L.
Ruco, Immunol. Today, 1992, 13, 265-270.
22
This article is protected by copyright. All rights reserved.

10.1002/anie.201913149
Angewandte Chemie International Edition
TOC graphical abstract:
Nanoaggregate for Detecting and Mapping Breast Cancer Metastasis via Multispectral
Optoacoustic Tomography and Aggregation-induced Near-infrared Fluorescence-I and
-II Imaging
Juan Ouyang +, Lihe Sun+, Zhuo Zeng, Cheng Zeng, Fang Zeng* and Shuizhu Wu*
An activatable nanoprobe with AIE feature has been developed as the dual-mode probe for
tracking breast cancer metastasis optoacoustically and fluorescently. Via 3D MSOT images the
focus of orthotopic breast tumors can be located and the sequential metastases of primary
tumors to lymph nodes and lung can be mapped spatiotemporally.
23
This article is protected by copyright. All rights reserved.