A macrophage uptaking near-infrared chemical probe CDnir7 for in vivo imaging of inflammation

Article abstract of DOI:10.1039/c4cc02038c

Visualization of macrophages in live animals has been of great interest for a better understanding of inflammation. We developed a near infrared (NIR) probe CDnir7 that can selectively detect macrophages and visualize inflammation in vivo using the IVIS spectrum, Fluorescence Molecular Tomography (FMT) and Multi-Spectral Optoacoustic Tomography (MSOT). This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4cc02038c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A macrophage uptaking near-infrared chemical
probe CDnir7 for in vivo imaging of inflammation†
Nam-Young Kang,‡a Sung-Jin Park,‡a Xiao Wei Emmiline Ang,a Animesh Samanta,a
Wouter H. P. Driessen,b Vasilis Ntziachristos,c Kristine O. Vasquez,d
Jeffrey D. Peterson,d Seong-Wook Yun*a and Young-Tae Chang*ae
Cite this: Chem. Commun., 2014,
50, 6589
Received 19th March 2014,
Accepted 2nd May 2014
DOI: 10.1039/c4cc02038c
www.rsc.org/chemcomm
Visualization of macrophages in live animals has been of great interest often resulting in high variation in data quality and contamination
for a better understanding of inflammation. We developed a near problems. Moreover, leakage of labeling agents from the injected
infrared (NIR) probe CDnir7 that can selectively detect macrophages cells can generate false background signals by labeling surrounding
and visualize inflammation in vivo using the IVIS spectrum, Fluores- cells non-specifically. On the other hand, cell type specific probes can
cence Molecular Tomography (FMT) and Multi-Spectral Optoacoustic be directly administered into the body without the need for extra cell
Tomography (MSOT).
preparation steps. Currently, only a few injectable imaging probes
based on fluorescently labelled magnetic nanoparticles for phago-
Inflammation is an immune response in the body normally triggered cytic uptake4 or extracellular protease substrate peptides5 have been
by infection, tissue damage or allergic substances. For the develop- developed to visualize macrophages in the body by optical imaging.
ment of inflammation in tissues, migration of immune cells, includ-
The conventional approach for fluorescent probe development,
ing monocytes from the blood, to the tissue is essential. Infiltrating however, has been limited to optical reporters conjugated to target-
monocytes then differentiate into macrophages and function primar- specific molecules such as antibodies, oligonucleotides and peptides.
ily as scavengers of foreign pathogenic organisms, apoptotic cells and As an alternative approach, we have developed a diversity oriented
dead cell debris and secrete pro- or anti-inflammatory cytokines.1 fluorescence library approach (DOFLA) using structurally diverse and
Macrophages are often recruited by solid tumors to become tumor- intrinsically fluorescent small organic dyes. By screening the DOFL in
associated macrophages which promote tumor progression and many different types of living cells, we have discovered fluorescent
metastasis by secreting growth factors for tumor cell proliferation dyes that selectively stain only certain types of cells such as embryonic
and angiogenesis.2 Therefore, monitoring the behavior of macro- stem cells, neural stem cells, basophils, pancreatic alpha cells and
phages in inflammation and tumor is crucial for the diagnosis and beta cells.6 While successful both in in vivo and in vitro imaging using
prognosis of a wide range of diseases.
DOFLA, a drawback of these previously reported dyes is the fact that
Methodologically, macrophage imaging is conducted by admin- their maximum excitation and emission wavelengths have generally
istration of macrophages that are pre-labelled in vitro, or by directly been between 400 and 600 nm and thus they have had limited
injectable probes that selectively label macrophages in inflamed tissue penetration capability, making them less suitable for in vivo
areas.3 Although employing in vitro labeled macrophages is the live animal imaging. To overcome this physico-optical limitation, we
most common method in current medical practice, the procedure recently developed photostable NIR compound libraries7 and applied
for isolating, labeling and reinjecting macrophages is complicated, these for macrophage selective probe development. Herein we report
the first NIR macrophage uptaking probe CDnir7 and its application
a Singapore Bioimaging Consortium, Agency for Science, Technology and Research,
11 Biopolis Way, # 02-02 Helios, Singapore 138667, Singapore.
E-mail: yun_seong_wook@sbic.a-star.edu.sg
of in vivo animal imaging in various modality and disease models.
For the primary screening of NIR compounds that selectively
stains macrophages versus leukocytes, we plated RAW 264.7 (mouse
macrophage cell line) and as the negative control, primary mouse
splenocytes (a mixture of various types of leukocytes) side by side in
96-well plates. The cells were incubated with two hundred NIR
compounds7 at 1 mM for 1 hour and the epi-fluorescence micro-
scopic image of each well was acquired. The fluorescence intensity of
the cells was analyzed using image analysis software as well as by
manual observation. Although a small number of macrophages were
expected to be in the splenocyte culture (approximately 10%), their
b iThera Medical, GmbH, Munich, Germany
c Institute for Biological and Medical Imaging (IBMI), Helmholtz Centre Munich
¨
Chair of Biological Imaging (CBI), Technische Universitat Munich, Germany
d Life Sciences & Technology, PerkinElmer, Inc, Boston, Massachusetts, USA
e Department of Chemistry & Med Chem Program, Life Sciences Institute,
National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore.
E-mail: chmcyt@nus.edu.sg; Web: http://ytchang.science.nus.edu.sg
† Electronic supplementary information (ESI) available: Experimental proce-
dures, spectroscopic data, cell image data. See DOI: 10.1039/c4cc02038c
‡ These authors contributed equally.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 6589--6591 | 6589

View Article Online
Communication
ChemComm
Fig. 1 Discovery of the NIR fluorescence probe CDnir7 and in vivo imaging of inflammation. (a) Chemical structure of CDnir7. (b) Fluorescence images
of mouse splenocytes and the macrophage cell line Raw 264.7. (c) Flow cytometry dot plot images of splenocytes and Raw 264.7. (d) CDnir7 detection in
inflamed regions of live mouse paws using the IVIS spectrum. LPS (1 mg mLÀ1, 200 mL) was injected into the right paws 2 days prior to the imaging. CDnir7
(100 mM, 250 mL, 1% PEG and 0.1% Tween 20 in PBS) was injected via the tail vein and the image was obtained 30 minutes later. (e) Immunostaining
images of CDnir7 and CD11b using the left rear foot (control region, e) and the right rear foot (f). Scale bar: 50 mm.
contribution to the total fluorescence signal was negligible in the culture media. In a living mammalian organism, however, the probes
screening. From the cell based screening, we identified 5 hit com- should selectively stain the target cells among high numbers of
pounds that stained RAW 264.7 cells more strongly than primary different types of cells, and unbound probes need to be either
splenocytes in epi-fluorescence microscopy imaging (Fig. 1b and metabolized or excreted from the body. These differences between
Fig. S1, ESI†). The five hit compounds were then evaluated for their cell culture and whole organism likely explain why the other four
efficiency in detecting inflammation in vivo using a well established primary hit compounds did not show specific signals in the inflamed
mouse model. We induced acute inflammation in mice by injecting regions of the mice.
lipopolysaccharide (LPS) into the paws (2 days) and then admini-
For further quantitative analysis of CDnir7 staining for
strated the hit compounds by tail vein injection. The fluorescence inflammation imaging in mice, we monitored the signal of
signals were monitored by the IVIS spectrum imaging system using CDnir7 by a tomographic fluorescence imaging method, FMT.
appropriate NIR filters, and we identified one compound CyNA-E10, For an acute animal model, we induced inflammation in mice
as the best imaging probe in vivo (Fig. 1d) based on its strong by injecting carrageenan (CG) into a footpad, an approach that
fluorescence signal only at the LPS injected site. Interestingly, the
other four hit compounds did not show a clear localization in the
inflammation site, when injected into this mouse model. The selected
in vivo probe compound, CyNA-E10 was dubbed CDnir7 (Compound
of Designation near infrared 7; Fig. 1a) and its optical properties
were fully characterized (lex/lem = 806/821 nm in DMSO: extinction
coefficient = 198 500 MÀ1 cmÀ1; quantum yield = 0.14 in Scheme S1 &
Fig. S6, ESI†). In addition to fluorescence imaging, CDnir7 demon-
strated a clear separation of macrophages from splenocytes using flow
cytometry analysis (Fig. 1c). Using the in vivo mice model, we further
confirmed that CDnir7 specifically stained macrophage cells in the
inflamed paw by immunostaining histological sections for the macro-
phage marker CD11b. Strong CDnir7 signals were observed only in
the inflamed paw and colocalized with CD11b positive cells (Fig. 1f),
whereas the control area remains largely negative for CD11b and
CDnir7 (Fig. 1e). CDnir7 did not show any apparent toxicity at the
working concentration (1 mM) over 24 hours in macrophage cells
(Fig. S2, ESI†) and in the liver, kidney and spleen tissue from CDnir7
injected mice even at 10 times the working concentration (Fig. S3–S5,
ESI†). There are many unpredictable factors that hinder the successful
application of imaging probes identified by cell-based screening
to whole animal in vivo imaging. In cell culture, only a few cell types
Fig. 2 Fluorescence Molecular Tomographic (FMT) imaging of paw inflam-
mation by CDnir7. Time-lapse FMT imaging of CDnir7 in edema induced by
the injection of 1% carrageenan (CG-marked with red box) into the right hind
are compared as negative control, under defined conditions
and unbound probes can be simply removed by exchanging the were imaged by FMT at 1, 2, 3 and 18 h after CDnir7 injection.
footpad. The left paw was injected with PBS as the internal control. Mice (n = 2)
6590 | Chem. Commun., 2014, 50, 6589--6591
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Fig. 3 In vivo imaging of CDnir7 kinetics and accumulation in breast tumor by Multi-Spectral Optoacoustic Tomography (MSOT). (a) Time lapse MSOT
imaging of CDnir7 signal in jugular veins (ROI for half-life determination marked with white circles). (b) MSOT image-based quantification of CDnir7 in
jugular veins over time. (c) Single wavelength optoacoustic (OA) image at 900 nm before CDnir7 injection (far left) and spectrally unmixed signals:
hemoglobin (oxygenated, HbO2; deoxygenated, Hb) and CDnir7 in the orthotopic breast tumor.
yields neutrophil and macrophage influx into the injected region. in mice bearing orthotopic 4T1 breast tumors, which are known to
After administration of CDnir7 into the blood stream through tail recruit large numbers of tumor-associated macrophages. In the tumor
vein injection, we measured the thickness of the paws and acquired region, which was identified by signals from oxygenated and deoxy-
fluorescence images periodically up to 18 hours. CG injection caused genated hemoglobin, the apparent accumulation of CDnir7 was
a rapid noticeable swelling in the paws, while PBS, injected as a observed (Fig. 3c). When the tumor region was monitored kinetically,
control, caused only slight swelling. A significantly stronger fluores- CDnir7 accumulation reached a maximum within 10 min and was
cence signal was detectable in the CG-injected paw compared to the detectable beyond 3 hours (Fig. S9, ESI†). In comparison to tumor
PBS control at 1 hour after CDnir7 administration. The fluorescence imaging kinetics, the kinetic changes in the CDnir7 signal in a nearby
signal in regions of CG-induced edema remained stable between 1 blood vessel were significantly different, decreasing rapidly in the first
and 3 h and then declined by 66% after 18 h. Despite the decline in 5 minutes and more slowly thereafter.
fluorescence intensity, the signal difference between inflamed and
In conclusion, we developed the first macrophage-targeted NIR
control paws was clearly detectable (Fig. 2), supporting the utility of probe and demonstrated in vivo imaging in mice using various
this probe in detecting and quantifying inflammation. Between paw modalities, IVIS, FMT and MSOT. Using in vivo imaging in various
thickness and CDnir7 signal intensity, there was an excellent linear modalities, CDnir7 provides a powerful fluorescence imaging probe
correlation (Fig. S7, ESI†). For broader application of the NIR probe, for the detection and quantification of inflammation occurring
we also assessed the capability of CDnir7 for in vivo photoacoustic in situ in regions of disease.
imaging by MSOT. First we tested whether the absorbance spectrum
This study was supported by an intramural funding from
and molar extinction coefficient of CDnir7 were affected by serum A*STAR (Agency for Science, Technology and Research, Singapore)
albumin, which has been known to bind to a wide range of organic Biomedical Research Council and a Singapore Ministry of Education
compounds in the blood potentially affecting their function and Academic Research Fund Tier 2 (MOE2010-T2-2-030).
property. In the presence of 10% bovine serum albumin (BSA) in
Notes and references
1 K. J. Moore, F. J. Sheedy and E. A. Fisher, Nat. Rev. Immunol., 2013, 13,
PBS, CDnir7 displayed maximum absorbance at 800 nm which was
slightly (15 nm) red-shifted compared to the value in PBS (Fig. S8a
and b, ESI†). In scattering agar phantoms (1.2% agar, 1% intralipid),
CDnir7 showed a strong photoacoustic signal and excellent linearity
at concentrations ranging from 0.15 mM to 5 mM regardless of the
presence of BSA (Fig. S8c and d, ESI†).
709–721.
2 P. Allavena and A. Mantovani, Clin. Exp. Immunol., 2012, 167, 195–205.
3 F. J. Van Hemert, C. Voermans, B. L. Van Eck-Smit and R. J. Bennink,
Q. J. Nucl. Med. Mol. Imaging, 2009, 53, 78–88.
4 R. Weissleder, K. Kelly, E. Y. Sun, T. Shtatland and L. Josephson,
Nat. Biotechnol., 2005, 23, 1418–1423.
5 R. Weissleder, C. H. Tung, U. Mahmood and A. Bogdanov Jr.,
Nat. Biotechnol., 1999, 17, 375–378.
To study the pharmacokinetics of CDnir7 by deep tissue in vivo
¨
imaging, we intravenously infused CDnir7 into naıve mice and
continuously monitored the neck region of the mice. The plasma 6 N. Y. Kang, H. H. Ha, S. W. Yun, Y. H. Yu and Y. T. Chang, Chem. Soc.
Rev., 2011, 40, 3613–3626.
7 A. Samanta, M. Vendrell, R. Das and Y. T. Chang, Chem. Commun.,
2010, 46, 7406–7408; A. Samanta, K. K. Maiti, K. S. Soh, X. Liao,
half-life of CDnir7, determined by spectral unmixing using the linear
regression of the signal from a jugular vein, was 21 min (Fig. 3a and
b). Plasma half-life of CDnir7 measured using MSOT is much shorter
than those of nanoparticle-based macrophage imaging probes, e.g.,
B3.9 h for 89 Zr-labeled dextran nanoparticles,8 and B10 h for cross-
M. Vendrell, U. S. Dinish, S. W. Yun, R. Bhuvaneswari, H. Kim,
S. Rautela, J. Chung, M. Olivo and Y. T. Chang, Angew. Chem., Int. Ed.,
2011, 50, 6089–6092; S. C. Lee, D. Zhai, P. Mukherjee and Y. T. Chang,
Materials, 2013, 6, 1779–1788.
linked iron oxide.9 This short plasma half-life of CDnir7 makes the 8 E. J. Keliher, J. Yoo, M. Nahrendorf, J. S. Lewis, B. Marinelli,
A. Newton, M. J. Pittet and R. Weissleder, Bioconjugate Chem., 2011,
22, 2383–2389.
9 P. Wunderbaldinger, L. Josephson and R. Weissleder, Bioconjugate
retained probe in macrophages rapidly distinguishable after injection
of the probe, thus offering an advantage over nanoparticles used to
detect phagocytic cells. We next studied the accumulation of CDnir7
Chem., 2002, 13, 264–268.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 6589--6591 | 6591