Aggregation-induced emission based PET probe for liver function imaging

Article abstract of DOI:10.1039/c9nj04537f

To locate the site of a liver lesion by positron emission tomography (PET) imaging and then remove the lesion site under the guidance of fluorescence imaging, we designed an aggregate-induced emission (AIE)-based PET probe [^nat/68Ga] 5 based on

Full text of DOI:10.1039/c9nj04537f

NJC
View Article Online
View Journal | View Issue
PAPER
Aggregation-induced emission based PET probe
for liver function imaging†
Cite this: New J. Chem., 2019,
43, 16305
Song Liu,a Yong Huang,a Yajing Liu,b Renbo Wu,a Zequn Yang,a Yuli Sun,a
Hao Xiao,a Xuebo Chenga and Zehui Wu
*
a
To locate the site of a liver lesion by positron emission tomography (PET) imaging and then remove the
lesion site under the guidance of fluorescence imaging, we designed an aggregate-induced emission
(AIE)-based PET probe [nat/68Ga] 5 based on the effects of the reticuloendothelial system. By modifying
derivatives of tetraphenylethene, we obtained a bifunctional chelating agent with AIE characteristics,
which can complex with natGa3+ or 68Ga3+ at room temperature to form [natGa] 5 or [68Ga] 5. [natGa] 5
has typical AIE characteristics and can form nanomicelles when its concentration is greater than the
critical micelle concentration (CMC). Cellular fluorescence confocal imaging experiments and
radioactive uptake experiments showed that Kupffer cells had a high uptake of [nat/68Ga] 5 while HepG2
cells had lower uptake of [nat/68Ga] 5, indicating that liver function imaging was achieved by the
phagocytosis of [nat/68Ga] 5 by Kupffer cells into the liver. A biodistribution study showed that [68Ga] 5
shows high accumulation in the liver after intravenous injection into rats. The in vivo PET imaging profile
demonstrated an excellent liver to muscle ratio. Thus, [nat/68Ga] 5 could potentially be used as a liver
function PET imaging agent for clinical diagnosis and intraoperative navigation.
Received 3rd September 2019,
Accepted 29th September 2019
DOI: 10.1039/c9nj04537f
rsc.li/njc
tomography (CT),9 magnetic resonance imaging (MRI),10 single
photon emission computed tomography (SPECT)11 and positron
1. Introduction
The liver is the largest gland in the human body and is an organ emission tomography (PET).12,13 Each of these imaging modalities
mainly characterized by metabolism. The liver plays a very important has its own advantages and disadvantages, and it requires compre-
role in human metabolism, bile production, detoxification, hensive consideration to achieve a three-dimensional qualitative/
coagulation, immunity, heat production and regulation of water quantitative assessment of liver function.14 Accordingly, attempts
and electrolytes.1 There are many diseases associated with the have been made to overcome their respective shortcomings and
liver, including viral and non-viral hepatitis, cirrhosis, alcoholic are driving the ongoing efforts to develop dual modality imaging
liver disease, autoimmune liver disease, non-alcoholic fatty liver instruments and contrasts so that the advantages of these
disease and liver cancer. Most liver diseases are occult and the techniques can be synergistically combined to provide accurate
early clinical symptoms are not obvious; therefore, diseases are physiological and anatomical information. In the past few years,
often found in the late stages of cirrhosis and liver cancer.2–4 different strategies have been applied to achieve multimodal
Liver tumour resection and liver transplantation are the main liver function imaging, such as reports of MRI-FL liver imaging
methods for curing liver cancer.5,6 Accurate evaluation of liver agents.15
function before surgery is expected to be the most important
PET can be non-invasive, highly sensitive, and highly specific
factor for the safe resection of the liver and accurate assessment for the quantitative determination of functional liver mass for
of postoperative residual liver function.7 Therefore, the detection patient management in a variety of clinical settings, including
of liver function by high-sensitivity imaging technology is of great liver surgery and transplantation as well as the diagnostic and
significance for the diagnosis, treatment and prognosis of liver therapeutic monitoring of cancer.12,13 However, its spatial reso-
disease, especially liver cancer. Various imaging methods have lution is very poor (Z1 cm for a clinical scanner). Fluorescence
been developed, including ultrasound imaging,8 computed imaging (FL) has high sensitivity and high spatial resolution,16
which complements the shortcoming of PET. Therefore, a
PET-FL bimodal liver function imaging agent can evaluate the
function of the liver and locate the tumour or liver necrosis
according to PET imaging and then accurately remove the
tumour or liver necrosis site under the guidance of fluorescence
a Brain Institute of Brain Disorders, Capital Medical University, Beijing 100069,
China. E-mail: wzhhuey2012@ccmu.edu.cn
b School of Pharmaceutical Science, Capital Medical University, Beijing 100069,
China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9nj04537f imaging.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16305--16313 | 16305

View Article Online
Paper
NJC
The development of PET-FL probes for liver function imaging ATCC, Manassas, VA. Fetal bovine serum were purchased from
is more challenging than single modality agents.17,18 These Beijing YuanHeng ShenMa Biology Technology Research Institute.
probes usually consist of fluorophores, linkers, radionuclides Culture medium and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-
and their complexing agents, and targeting molecules, each of tetrazolium bromide (MTT) were purchased from Sigma.
which plays a very important role in imaging.19 Additionally,
2.2 Instruments
fluorophores experience some effects from emission quenching,
either partially or completely in many conventional systems.20 NMR spectra were recorded at 300 MHz at ambient temperature.
The harsh conditions of radiolabelling can result in changes to Chemical shifts are reported in parts per million downfield from
the molecular structure of the target molecule.21 Complex agents TMS (tetramethylsilane). Coupling constants in 1H NMR are
or fluorophores can affect the affinity of the probe for the target. expressed in Hertz. High-resolution mass spectrometry (HRMS)
Therefore, to solve the above problems, a novel strategy for data was obtained with an Agilent (Santa Clara, CA) G3250AA LC/
designing dual-mode imaging agents for PET-FL is particularly MSD TOF system. Thin-layer chromatography (TLC) analysis were
important for the development of new imaging probes.
performed using Merck (Darmstadt, Germany) silica gel 60 F254
Compared with the traditional PET-FL bimodal imaging probes plates. Crude compounds were generally purified by flash
consisting of five units, a new strategy attempts to integrate the column chromatography (FC) packed with Teledyne ISCO.
fluorophore, nuclides and complexing agents into one unit in this Photoluminescence (PL) spectra were recorded on a F-2500
paper. Fluoregens bearing aggregate-induced emission (AIE) char- spectrofluorometric Hitachi, Ltd. The morphology of 5 aggre-
acteristics have been introduced into dual-mode imaging probes gates was investigated using transmission electron microscopy
in new strategies to overcome conventional aggregation-caused (TEM; Japan, JEM-2100) at an accelerating voltage of 100 kV.
quenching (ACQ), which has been extensively studied in optoelec- Dynamic light scattering (DLS) and zeta potential measurements
tronic devices, chemical and biosensors, bioimaging, etc.15,22–33 were performed on a Malvern Zetasizer Nano-ZS90. For cell
AIE, conceptually termed by Tang in 2001,34 can actively utilize imaging experiments, Kupffer cells and HepG2 cells were imaged
luminogen aggregation to enhance its emissions. Tetrapheny- under a Zeiss LSM-880 confocal laser scanning microscope (Zeiss,
lethene (TPE) is one of the most famous AIE fluorogens,22 and Germany). 68Ge/68Ga generator supplied by Isotope Technologies
by structural modification, its amphiphilic derivatives have been Garching (ITG), Germany. For cytotoxicity experiments, cells were
shown to act as a good chelating agent to label 68Ga3+, a useful PET treated with MTT and the absorbance was measured on a
radionuclide with a 68 min half-life and mild labelling conditions.35 1420 Victor Multi-Label Counter (PerkinElmer Life Science).
Thus, the integration of fluorophores, nuclides and complexing Small animal PET imaging data were recorded on a microPET
agents into amphiphilic TPE derivatives reduces the difficulty of (Inveon, Siemens, Germany, Detector Type: LSO with 1.6 ꢀ 1.6 mm
design and links to different targeting molecules to develop dual- detector pixel spacing; Spatial Resolution: 1.4 mm).
mode PET-FL imaging probes for the diagnosis of different diseases.
The amphiphilic natGa3+ or 68Ga3+ complex may form nanomicelles
2.3 Synthesis
larger than 100 nm at a suitable concentration, which in turn can be
2,20-((Ethane-1,2-diylbis(azane diyl))bis(methylene))bis(4-(1,2,2-
specifically and mainly trapped in the macrophages of the triphenylvinyl)phenol) (2). A solution of 1 (0.69 g, 2 mmol) was
reticuloendothelial system (RES).15 Kupffer cells located in the added in 20 mL methanol at room temperature, ethane-1,
liver are specialized macrophages that act as an integral part of 2-diamine (60 mg, 1 mmol) were added. The solution was
the sinusoids. The Kupffer cells can remove the PET-FL probe stirred at 60 1C for 3 h. NaBH4 (0.14 g, 4 mmol) was added
from the bloodstream through the phagocytosis pathway, resulting above solution in portions at 0 1C for 1 h. The reaction was then
in the accumulation of the PET-FL probe in the liver tissue, which in acidified with 1 M HCl until the pH = 7 under cooling with ice
turn allows liver function imaging.
bath. The methanol was removed and water phase was
To obtain the PET-FL liver function imaging agent, we extracted with ethyl acetate (20 mL ꢀ 3). The organic phases
synthesized a bifunctional chelating agent with AIE characteristics were combined, washed with Sat. NaHCO3 (20 mL) and brine
that can complex with natGa3+ or 68Ga3+ at room temperature to (20 mL), and dried with MgSO4. The filtrate was evaporated and
form [natGa/68Ga] 5. Because of the formation of nanoaggregates, extracted by ethyl acetate, dried by Na2SO4, purified by FC
[68Ga] 5 displays high liver uptake in rats with an excellent liver to (DCM/methanol = 10/1) to give 2 (0.43 g, 56.1%) as white solid.
muscle ratio. The utilization of [natGa] 5 for cell imaging was 1H NMR (300 MHz, CDCl3) d 7.11–7.03 (m, 30H), 6.83 (d, J = 7.5 Hz,
examined, and the feasibility of using [68Ga] 5 as a PET imaging 2H), 6.73–6.48 (m, 4H), 3.73 (s, 4H), 2.60 (s, 4H).13C NMR (75 MHz,
agent was demonstrated.
CDCl3) d 156.64, 144.28, 144.17, 143.87, 140.33, 139.60, 134.71,
131.96, 131.75, 131.40, 131.33, 127.58, 126.35, 126.18, 126.08,
121.22, 115.46, 100.07, 77.43, 77.01, 76.59, 51.98, 47.37. HRMS calcd
for C56H48N2O2 780.3716; found, 781.3632 [M + H]+.
2. Experimental
2.1 Materials
Di-tert-butyl 2,20-(ethane-1,2-diylbis((2-hydroxy-5-(1,2,2-triphenyl-
All reagents were commercial products used without further vinyl)benzyl)azanediyl))diacetate (3). A solution of 2 (0.4 g,
purification unless otherwise indicated. Deionized water used 0.51 mmol) and K2CO3 (0.28 g, 2 mmol) were added in 45 mL
in our experiments was obtained from a Milli-Q water system. acetonitrile was stirred at room temperature for 0.5 h, t-butyl
Kupffer cell and HepG2 cell were obtained from the American, 2-bromoacetate (190 mg, 1 mmol) was added. The reaction was
16306 | New J. Chem., 2019, 43, 16305--16313 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
stirred at 60 1C for overnight. The reaction was filtered and
evaporated in vacuo, purified by FC (hexane/ethyl acetate = 4/1)
to give 3 (0.24 g, 46.7%) as white solid. 1H NMR (300 MHz,
DMSO) d 9.54 (s, 1H), 7.17–7.01 (m, 18H), 6.95–6.90 (m, 12H),
6.67 (d, J = 6.4 Hz, 4H), 6.52 (d, J = 8.8 Hz, 2H), 3.42 (s, 4H), 3.03
(s, 4H), 2.35 (s, 4H), 1.39 (s, 18 H).13C NMR (75 MHz, DMSO)
d 170.62, 155.69, 144.20, 143.96, 143.83, 141.06, 139.72, 134.12,
131.14, 128.18, 128.08, 126.83, 126.59, 122.70, 115.36, 80.97, 55.34,
40.87, 40.59, 40.32, 40.04, 39.76, 39.48, 39.20, 28.23. HRMS calcd for
C68H68N2O6 1008.5077; found, 1009.4784 [M + H]+.
2,20-(Ethane-1,2-diylbis((2-hydroxy-5-(1,2,2-triphenylvinyl)benzyl)-
azanediyl))diacetic acid (4). A solution of 3 (100 mg, 0.099 mmol)
and trifluoroacetic acid (TFA, 4 mL) was stirred at room temperature
for 3 h. The solvent was removed under vacuum, diethyl ether was
added, washed by diethyl ether, white solid 4 was collected (89 mg,
1
100%). H NMR (300 MHz, DMSO) d 7.06 (s, 18H), 6.99–6.88 (m,
12H), 6.67 (s, 4H), 6.54 (s, 2H), 3.43 (s, 4H), 3.08 (s, 4H), 2.40 (s, 4H).
13C NMR (75 MHz, DMSO) d 172.77, 155.76, 144.18, 143.98, 143.85,
140.94, 139.73, 134.12, 131.64, 131.16, 128.17, 128.10, 126.85,
126.69, 122.15, 115.46, 67.45, 55.34, 54.19, 40.82, 40.54, 40.26,
39.98, 39.70, 39.43, 39.15. HRMS calcd for C60H52N2O6 896.3825;
found, 897.3557 [M + H]+.
Fig. 1 (a) HPLC profiles of [natGa] 5 and [68Ga] 5 using the following
system (column: Gemini-NX 5u C18 110A (150 ꢀ 4.6 mm), mobile phase
ACN/0.1% TFA in H2O, 1 mL minꢁ1 (0–2 min, 10% ACN; 2–30 min, 10–60%
ACN; 20–21 min, 60–100% ACN; 20–26 min, 100% ACN), and column
temperature of 30 1C). [natGa] 5 and [68Ga] 5 had retention times of 23.3
and 23.6 min, respectively. Labelling yield of [68Ga] 5 in HEPES buffer
100 1C at (b) different pH values, (c) concentrations and (d) concentrations
of metal cations. Each data point represents the mean ꢂ SD from triplicate
trials.
Gallium N-(2-((carboxylatomethyl)(2-hydroxy-5-(1,2,2-triphenyl-
vinyl)benzyl)amino)ethyl)-N-(2-oxido-5-(1,2,2-triphenylvinyl)benzyl)-
glycinate (5). A solution of 4 (100 mg, 0.11 mmol) and GaCl3 (0.19 g,
1.1 mmol) was stirred in 0.5 mL DMSO and 0.5 mL H2O at room
temperature for overnight. The reaction was purified by semi-HPLC
1
and afforded white solid 5 (95 mg, 89.2%). H NMR (300 MHz,
GLUCOSE (HyClone) supplemented with 10% fetal bovine
serum (YHSM) and 1% penicillin/streptomycin (Gibco). The
cells were maintained in T-75 culture flasks under humidified
incubator conditions (37 1C, 5% CO2) and were routinely
passaged at confluence. Cells were plated (1.0 ꢀ 105 cells per
well) overnight in the media prior to ligand incubation. [natGa] 5
was dissolved in DMSO with the concentration of 10 mM as stock
solution. For cell imaging, Kupffer cells and HepG2 cells were
seeded overnight on a coverslip mounted onto a 35 mm Petri
dish and then stained with different concentrations [natGa] 5.
DMSO) d 7.26–6.86 (m, 30H), 6.55 (dd, J = 8.5, 2.3 Hz, 2H), 6.38 (d,
J = 2.2 Hz, 2H), 6.28 (d, J = 8.5 Hz, 2H), 3.17–3.13 (m, 2H), 3.01–2.95
(m, 3H), 2.77–2.72 (m, 7H).13C NMR (75 MHz, DMSO) d 171.58,
163.13, 144.69, 144.31, 143.06, 141.47, 138.12, 133.78, 132.12,
131.82, 131.40, 131.30, 131.14, 129.26, 128.23, 128.11, 128.01,
126.70, 126.41, 119.96, 99.99, 60.99, 56.14, 55.33, 41.98. HRMS
calcd for C60H49GaN2O6 962.2846; found, 963.2499 [M + H]+.
2.4 Radiosynthesis of [68Ga] 5
0.2 mL eluent in 0.05 M HCl of 68Ge/68Ga generator (ITG,
Germany) and 0.2 mL 2 N HEPES (pH = 7) were added and mixed
with 10 mL of 4 ꢀ 10ꢁ3 mol Lꢁ1 and reacted at room temperature for
10 min (Scheme S1, ESI†). Labeling efficiency and radiochemical
purity were determined using Radio-HPLC. Radiochemical purity
of [68Ga] 5 was 4 95%. Therefore, the tracer was diluted and used
in vitro and in vivo experiments without further purification.
Analytical reversed-phase high performance liquid chromatography
(RP-HPLC) was performed on a Gemini-NX 5u C18 110A column
(150 ꢀ 4.6 mm) using an Agilent gradient HPLC System. The [68Ga]
5 was eluted applying gradients of in 0.1% TFA in H2O and aceto-
nitrile (ACN) at a constant flow of 1 mL minꢁ1 (0–2 min, 10%ACN;
2–30 min, 10–60% ACN; 20–21 min, 60% ACN; 20–26 min, 100%
ACN). [68Ga] 5 had retention times of 23.6 min (Fig. 1a).
2.6 Cell uptake of [68Ga] 5
The cells were cultured for 2 days at 37 1C in a 5% CO2
incubator. Each cell line was set at four time points of 5, 30,
60 and 120 min, and four replicate wells were set at each time
point. The well plates were removed, the medium was washed
away, and each well was washed three times with 1 mL of pre-
heated PBS in 37 1C (containing Ca2+ and Mg2+). Add 1 mL of
PBS-configured 37 KBq (1 mCi) [68Ga] 5 to each well and place in
a 37 1C incubator. The plates were removed after four time
points, PBS was aspirated and washed three times with 1 mL of
cold PBS without Ca2+ and Mg2+. The cells were harvested; the
radioactivity was counted by adding a gamma counter.
2.5 Cell culture and imaging
2.7 In vitro stability in PBS and plasma
Kupffer cells were cultured in RPMI 1640 (SIGMA) supplemented [68Ga] 5 was added to 0.1 M phosphate buffered saline (PBS,
with 10% fetal bovine serum (YHSM) and 1% penicillin/strep- pH = 7.0) or rat plasma, and the solutions were incubated at
tomycin (Gibco). HepG2 cells were cultured in DMEM/HIGH 37 1C for 30, 60 and 120 min. Plasma was filtered through a
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16305--16313 | 16307

View Article Online
Paper
NJC
filter (0.22 mm) and radiochemical purity of the filtrate was isoflurane in oxygen at 2 L minꢁ1, in an induction chamber.
analyzed by RP-HPLC. The RCP was measured and analyzed by When fully anesthetized, the animals were placed on the
the radio-HPLC to determine the stability of [68Ga] 5 as a scanner bed, with a nose cone used to maintain anesthesia at
function of time.
1% isoflurane in oxygen at 2 L minꢁ1. Body temperature was
maintained at 37 1C by using a water-circulated pad under the
animals. Immediately after the start of the scan, a dose of radio
2.8 Cell proliferation assay
Kupffer cells were seeded into a 96-well plate at a density of tracer (B300 mCi, 11.1 MBq) was injected via the tail vein. The
3000 cells per well and incubated for 12 h at 37 1C and 5% CO2 in animals were visually monitored for breathing and any other
culture medium. The cells were washed with 0.1 mL ꢀ PBS. Then signs of distress throughout the entire imaging period. All
different concentrations of [natGa] 5 and fresh culture medium were scans were conducted over a period of 120 min (dynamic, 5 min
added. After 24 h incubation, all cells were then exposed to 20 mL of per frame). Regions of interest (ROIs) were drawn over liver guided
a MTT solution (5 mg mLꢁ1). After 4 h at 37 1C, culture media was by CT images using Amira 3.1 image visualization and analysis
removed and 0.15 mL dimethyl sulfoxide was added. The absor- software. Radioactivity within the ROIs of each frame was
bance of individual wells at 570 nm was measured on a 1420 Victor calculated. The radioactivity in the liver ROI for each frame
Multi-Label Counter. All experiments were repeated three times.
was decay-corrected. The time–activity curve of liver was plotted
from the liver ROI activities in each frame.
2.9 In vivo autoradiography and fluorescence imaging of liver
sections
2.12 Statistical analysis
For in vivo autoradiographic studies, Healthy BALB/c mice
(18–22 g, purchase from Beijing Charles river Laboratories).
All animal procedures were performed in accordance with the
Guidelines for Care and Use of Laboratory Animals of Captical
Medical University and experiments were approved by the
Animal Ethics Committee of Captical Medical University. Studies
of the in vitro autoradiography and fluorescence imaging of [68Ga]
5 was performed in mice. Approximately 1.29 MBq [68Ga] 5 was
administrated via the tail vein in conscious animals. At 60 min
postinjection, the mice were sacrificed under anesthesia. The
liver was rapidly removed, placed in a Tissue-Tek OCT compound
(Sakura, Japan), and frozen in a dry ice–acetone bath for 2 h.
After reaching equilibrium to ꢁ20 1C, 35 mm sections were
consecutively cut by a cryostat microtome (Leica, CM1950,
GER), thaw mounted on gelatin-coated microscope slides, and
air-dried at room temperature. The slides were then exposed to a
phosphor screen (PerkinElmer Cyclone plus, C431200, USA) in
an autoradiographic cassette for 20 h. The autoradiography
images were acquired through a Typhoon FLA 7000 (Perkin-
Elmer Cyclone plus, C431200, USA). The fluorescence images were
acquired through Pannoramic scan (3DHistech II, Hungary).
Data were presented as mean ꢂ standard deviation (SD). All
statistical tests were conducted using the IBM SPSS Statistics
Version 20.0 for Windows (SPSS, Inc., IBM Company). The
independent samples nonparametric test was used to compare the
difference of two quantitative groups. Pearson product-moment
correlation coefficient (r) was used for correlation analysis
between continuous variables. A P value of less than 0.05 was
considered statistically significant.
3. Results and discussion
3.1 Synthesis and characterization of [natGa] 5
To enable the TPE derivative to efficiently complex 68Ga3+,
[
natGa] 5 was designed in accordance with the structure of
HBED-CC (Scheme 1).36 Intermediate 1 was prepared by the
reported conditions. Condensation of 1 with ethylenediamine
produced the Schiff base without need for further purification.
2.10 Biodistribution studies in rats
Healthy male Sprague Dawley rats (180–220 g, purchase from
Beijing Charles river Laboratories) were injected with [68Ga] 5
(740 KBq, 0.1 mL) through a tail vein under anesthesia with
2% isoflurane. The injected mice were sacrificed at 5, 30, 60 and
120 min postinjection. Brain, blood, liver, muscle, kidney, lung,
heart, skin, bone and spleen were then excised, blotted, and
weighed, and then 68Ga radioactivity of each organ was counted
by a gamma scintillation counter. Results were expressed as the
percentage of injected dose per gram of tissue (% ID per g). The
percent dose per gram to brain ratios was calculated by a compar-
ison of the tissue activity counts to counts of 1% of the initial dose.
2.11 MicroPET imaging
Dynamic PET/CT imaging studies were conducted in healthy
male Sprague Dawley rats. Rats were anesthetized using 2%
Scheme 1 Synthetic route to dual-modal [natGa] 5.
16308 | New J. Chem., 2019, 43, 16305--16313 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
The corresponding secondary amine 2 was obtained from the
Schiff base after a reduction reaction with sodium borohydride in
56.1% yield. The secondary amine, 2, was then reacted with two
equivalents of t-butyl bromoacetate to give 3 in 46.7% yield. The tert-
butyl group of 3 was then removed with trifluoroacetic acid (TFA)
to afford 4, in 100% yield. Gratifyingly, the desired [natGa] 5 was
afforded by the reaction of 4 with Ga3+ at room temperature in
89.2% yield. All compounds were confirmed by NMR and HRMS.
3.2 Radiosynthesis and characterization of [68Ga] 5
The initial conditions for the [68Ga] 5 labelling were in reference
to the optimal conditions for HBED-CC labelling (total volume
of 300 mL at pH = 4.0 with a concentration of 4 of 20 mM),36 but
the temperature was 100 1C to ensure that the 4 was fully reacted
and the labelled product [68Ga] 5 was confirmed by radio-HPLC
(Scheme S1 (ESI†) and Fig. 1a). Gratifyingly, [68Ga] 5 can be
obtained by the above conditions, with a labelling yield of 95%
and a radiochemical purity of 495%. Further detailed studies
aim to determine the optimal conditions for the radiosynthesis
of 4 by varying the pH value, temperature, time and concentration
based on the optimized initial reaction conditions. When the pH
was increased from 1 to 4, the yield of [68Ga] 5 greatly increased,
and a slight improvement in yield was observed at pH 5–7 (Fig. 1c).
At pH = 7, [68Ga] 5 can still be obtained in 95% yield at room
temperature after 10 min. As the concentration increased from
8 mM to 90 mM, the labelling yield increased correspondingly
under the same reaction conditions (Fig. 1b). Further investigation
of the reaction time showed that when the reaction time exceeds
5 min, the labelling yield can be greater than 90% at room
temperature. According to the above experimental results, the
optimal labelling conditions were obtained; that is, the reaction
solution was in an acetate buffer at a pH of 5–7, the con-
centration of 4 was not less than 20 mM, and the reaction was
carried out for 5–15 min at room temperature. When the
reaction solution was changed from acetic acid buffer solution
to HEPES buffer solution, the above optimal conditions still
apply. The effects of metal ions with ionic radii similar to Ga3+, such
as Fe3+, Zn2+, and Cu2+, on 68Ga3+ labelling were also investigated.
The results show that Cu2+ had the greatest influence on the
labelling, as an equivalent amount of Cu2+, Fe3+ or Zn2+ was
added to the reaction solution and the labelling yield reduced
from 95% to 20%, 60% and 70%, respectively (Fig. 1d).
Fig. 2 (a) Photographs taken under UV irradiation and (b) emission
spectra of [natGa] 5 in aqueous solutions at different ratios of DMSO
and water. Concentration of [natGa] 5 : 15 mM. (c) Emission spectra and
(e) photographs taken under UV irradiation of [natGa] 5 in aqueous solutions
at different concentrations. (d) Plots of PL intensity versus concentration of
[
natGa] 5 in DMSO/H2O = 1/99. The CMC of [natGa] 5 is 0.7 mm. Slit width =
5 nm, excitation wavelength: 330 nm.
(Fig. 2d). Two lines are generated by analysing the relationship
between the fluorescence intensity and the concentration of
[
natGa] 5, the intersection of which gives the CMC of 1 mM
(Fig. 2d). Using the same method, the CMC value of 4 was
determined to be 3.5 mM (Fig. S2, ESI†). Comparing the CMC
values of 4 and [natGa] 5, the value of [natGa] 5 is lower probably
because of the increasing repulsive forces of the hydrophilic end
helping to form micelles at low concentrations. The size and
morphology of the micelle size formed by the aggregation of
[
natGa] 5 was further verified by a zeta potential particle size
analyser and transmission electron microscopy. Transmission
electron microscopy (TEM) images show that the nanomicelles
were non-uniform, easily adhesive spheres with an average
diameter of approximately 55 nm (Fig. 3b), which is consistent
with the hydrodynamic size measured by the zeta potential
particle size analyser (Fig. 3a).
3.3 AIE properties of [natGa] 5
After confirming the structure of [natGa] 5, we studied its photo-
physical properties. [natGa] 5 is weakly fluorescent when the
water fraction (fw) in the DMSO/water mixture is less than 60%
at 330 nm excitation. As the amount of water increases, the
fluorescence of the solution gradually increases, showing a
typical AIE effect (Fig. 2a). The labelling precursor 4 has AIE
characteristics similar to [natGa] 5 in a mixed solvent of DMSO
and water (Fig. S1, ESI†). It has been reported that metal
complexes of amphiphilic tetraphenylethene may form micelles.15
Therefore, we speculate that [natGa] 5 can also form micelles above a
CMC. We attempted to estimate the CMC of [natGa] 5 based on the
fluorescence intensity of different concentrations of [natGa] 5
Fig. 3 (a)Transmission electron micrographs and (b) particle size analysis
of [natGa] 5 in DMSO/H2O (10 mM).
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16305--16313 | 16309

View Article Online
Paper
NJC
3.4 In vitro fluorescent imaging of [natGa] 5
10 min, while under the same conditions, [natGa] 5 fails to induce
luminescence on the HepG2 cell membrane (Fig. 4). As the
incubation time increased, [natGa] 5 further induced lumines-
cence on the Kupffer cell membrane (Fig. 4), while [natGa] 5 still
failed to illuminate the HepG2 cell membrane. The above
results indicate that the possible mechanism of [nat/68Ga] 5 liver
function imaging is mainly achieved by [nat/68Ga] 5 engulfing the
Kupffer macrophages cell.
To explore the applications of [natGa] 5 in liver function imaging,
we first tested its potential toxicity to Kupffer cells using the
methyl thiazolyl tetrazolium (MTT) assay. After incubation with
[
natGa] 5 at various concentrations for 24 h, no significant cell
cytotoxicity from [natGa] 5 was observed even at high concentrations
up to 10 mM, which is five times higher than the working
concentration in cell imaging experiments, and the cell viability
was still approximately 85.2% (Fig. 5a). To further explore the
mechanism of liver function imaging and the scope of the
imaging applications, in vitro cell confocal imaging experiments
were performed. Kupffer cells and HepG2 cells were incubated
with [natGa] 5 for different periods of time. As shown in Fig. 4,
3.5 In vitro stability in PBS and plasma
The in vitro stability of [68Ga] 5 was investigated by incubation
both in PBS and in rat plasma at 37 1C. After 2 h of incubation,
the radiochemical purity was unchanged, maintaining a stable
level of approximately 490% (Fig. 5b). [68Ga] 5 was stable in
plasma and PBS in vitro without any measurable decomposition.
The release of 68Ga3+ from the complex was not observed under the
respective experimental conditions in the examined time frame.
[
natGa] 5 illuminates the Kupffer cell membrane for approximately
3.6 Cell uptake of [68Ga] 5
To further explore the mechanism of [68Ga] 5 liver function
imaging and verify the results of the confocal cell imaging,
radioactive cell uptake experiments were performed. As shown
in Fig. 5c, Kupffer cells showed higher uptake of [68Ga] 5,
increasing by 1.25%, 2.18%, 3.13% and 6.24% at 5, 30, 60
and 120 min, respectively. The uptake of [68Ga] 5 by HepG2 cells
was lower at the different time points. Based on the in vitro
fluorescence imaging experiments, [68Ga] 5 could be used to
discriminate between Kupffer cells and HepG2 cells.
3.7 In vivo autoradiography and fluorescence imaging of liver
sections
To realize precise image-guided resection for liver necrosis or
liver tumours, it is key to ensure that the full outline of liver
necrosis or liver tumours of various sizes can be visualized
during the surgical process. Therefore, we attempted to co-localize
normal liver tissue by fluorescence and autoradiography. As shown
in Fig. 5d, the liver margin can be readily distinguished by
fluorescence imaging. The liver profile detected by fluorescence
imaging exhibited consistency with that of the autoradiography
results (Fig. 5e). It can be seen from the experimental results that the
lesion points of the liver are first found by PET imaging, and the
liver necrotic tissue is removed under the guidance of fluores-
cence imaging.
3.8 Biodistribution studies in rats
To further investigate the pharmacokinetics of [68Ga] 5 in rats,
biodistribution experiments were performed. The results of the
biodistribution studies with rats 5, 30, 60 and 120 min post-
injection are shown in Table 1. [68Ga] 5 showed a very fast
accumulation throughout the entire liver with low background
activity in all other organs and a very fast blood clearance
(Table 1). The liver uptake of [68Ga] 5 was rapid with 13.21 ꢂ
3.27% ID per g at 5 min and remained relatively constant over
the course of the study to 14.49 ꢂ 2.3% ID per g at 30 min
and 16.37 ꢂ 0.69% ID per g at 60 min (Table 1). After 2 h,
the liver uptake continued to increase to 20.65 ꢂ 2.2% ID per g.
Fig. 4 Confocal images of HepG2 cells and Kupffer cells stained with
natGa] 5 conc. = 2 mM, incubation time 10, 30 and 120 min, excitation
wavelength: 405 nm; emission filter: 460–520 nm.
[
16310 | New J. Chem., 2019, 43, 16305--16313 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
clear liver uptake is visualized, and other organs did not see
significant intake (Fig. 6a–c and Fig. S8, ESI†). To confirm this,
region of interest analysis was performed using Amira 3.1
image visualization and analysis software on the reconstructed
images to generate the time–activity curves for [68Ga] 5. The
kinetics indeed confirm that the tracer exhibited higher liver
uptake than muscle (background) region uptake. Rapid liver
uptake is visualized within the first 5 min. Liver uptake remains
rather consistent throughout the 2 h scan time, with an
observed slow washout rate (Fig. 6d). As shown in Fig. S3 (ESI†),
the circulation lifetime of [68Ga] 5 in the blood is very short,
which may be because [68Ga] 5 forms nanoparticles larger than
100 nm in the blood, which is not conducive to blood
circulation.37,38 Nanoparticles (4100 nm) may be particularly
trapped in macrophages of the reticuloendothelial system
(RES),37,38 which may be an important reason for the high uptake
of [68Ga] 5 by the liver. [68Ga] 5 was phagocytosed by Kupffer cells
when blood flows into the liver, thereby achieving the purpose of
liver function imaging. Since hepatic metastases lack Kupffer
cells, they would not show uptake after [68Ga] 5 administration,
therefore, there is no signal or weak signal in the liver lesions
compared with normal liver function PET imaging. However,
focal liver lesions with Kupffer cells, such as focal nodular
hyperplasia, adenoma, and well-differentiated hepatocellular
carcinoma, will show significant uptake,15 therefore, the liver
lesions signal was enhanced compared with normal liver function
pet imaging. Based on the results we obtained and by comparing
the data from previously reported liver imaging agents and TPE
derivatives,13,15,39,40 [68Ga] 5 is expected to be a promising PET
imaging agent for liver functional imaging.
Fig. 5 (a) Relative Cell proliferation of Kupffer cells after treating with
[
natGa] 5 with the concentration of Y ranging from 2 to 10 mM for 24 h. No
significant difference was observed after the addition of [natGa] 5 compared
with the control group (P 4 0.05). (b) The stability in PBS and plasma of [68Ga] 5.
(c) In vitro cell uptake of [68Ga] 5. The two cell was compared using two-
tailed t tests (*, P o 0.01 and **, P o 0.001). Each data point represents
mean ꢂ SD from five trails. The fluorescence images (d) and in vitro auto-
radiography (e) of liver sections from mice injection of [68Ga] 5 into mice.
Table 1 In vivo biodistribution of [68Ga] 5 in rats
Organ
5 min
30 min
60 min
120 min
Brain
Liver
Heart
Spleen
Lung
Kidney
Muscle
Bone
0.41 ꢂ 0.1
0.21 ꢂ 0.2
0.09 ꢂ 0.01
0.15 ꢂ 0.05
13.21 ꢂ 3.27 14.49 ꢂ 2.3
16.37 ꢂ 0.69 20.65 ꢂ 2.2
0.62 ꢂ 0.06
5.83 ꢂ 0.51
1.75 ꢂ 0.21
0.37 ꢂ 0.04
1.84 ꢂ 0.76
4.39 ꢂ 0.97
2.05 ꢂ 1.36
5.03 ꢂ 1.01
0.28 ꢂ 0.17
5.72 ꢂ 0.29
1 ꢂ 0.25
0.2 ꢂ 0.02
5.3 ꢂ 0.56
1.49 ꢂ 0.41
0.17 ꢂ 0.01
0.31 ꢂ 0.12
1.04 ꢂ 0.37
0.66 ꢂ 0.2
3.39 ꢂ 0.92
0.25 ꢂ 0.05
6.75 ꢂ 1.13
1.03 ꢂ 0.28
0.2 ꢂ 0.02
0.29 ꢂ 0.09
1.51 ꢂ 0.82
0.83 ꢂ 0.24
3.01 ꢂ 0.61
0.26 ꢂ 0.22
0.33 ꢂ 0.19
1.13 ꢂ 0.48
0.52 ꢂ 0.08
3.95 ꢂ 5.08
Skin
Blood
The in vivo biodistribution results suggested that [68Ga] 5 is
mainly taken up by the liver, which is consistent with the
cellular fluorescence imaging and cellular uptake results.
1.29 MBq of [68Ga] 5 was administrated via tail vein injection
without anesthesia. The animals were euthanized at 5, 30, 60,
120 min (n = 5) after injection. The data are expressed as mean
% ID per g with standard deviation. There was significant
difference in tumor and muscle uptake values after 5 min
( p o 0.001), 30 min ( p o 0.001), 60 min ( p o 0.0001), and
120 min ( p o 0.001) of injection.
3.9 MicroPET imaging
Fig. 6 Representative PET images of rats after intravenous injection of
Dynamic small-animal PET studies on rats were performed with
[68Ga] 5. Animal PET images of summed 2 h coronal sections
were selected for visualization. As the images demonstrate,
[
68Ga] 5. The images of the (a) transverse, (b) coronal, and (c) sagittal views
are from a summed 2 h scan. (d) Time–activity curves of [68Ga] 5 uptake in
the liver and muscle.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16305--16313 | 16311

View Article Online
Paper
NJC
W. Kim, J. Kluwe, R. Kochhar, N. Dhaka, P. M. Costa,
M. A. Nabeshima Pharm, S. K. Ono, D. Reis, A. Rodil, C. R.
Domech, F. Saez-Royuela, C. Scheurich, W. Siow, N. Sivac-
Burina, E. S. Dos Santos Traquino, F. Some, S. Spreckic,
S. Tan, J. Vorobioff, A. Wandera, P. Wu, M. Yakoub, L. Yang,
Y. Yu, N. Zahiragic, C. Zhang, H. Cortez-Pinto and R. Bataller,
Clin. Gastroenterol. Hepatol., 2019, 29, 30073.
4. Conclusions
In this work, the integration of fluorophores and complexing
agents reduces the influencing factors of multimodal imaging
agent design, which strongly promotes the development of PET-FL
multimodal imaging agents. The bifunctional complexing agent
with AIE characteristics designed in this paper has a wider range
of labelling pH values and milder labelling conditions than
HBED-CC and is suitable for the labelling of sensitive peptides
or antibodies. In addition, 68Ga3+ provides a simple, rapid
radiolabelling method, which is suitable for use in a convenient
lyophilized kit formulation. In summary, this work designed an
AIE-based PET liver function imaging agent. [natGa] 5 can form
nanoscale micelles in DMSO/H2O = 1/99. These nanoaggregates
have high emissivity and exhibit a typical AIE effect. Fluorescence
imaging experiments and radioactivity uptake experiments showed
that Kupffer cells had high uptake of [nat/68Ga] 5, while HepG2 cells
had lower uptake of [nat/68Ga] 5. [nat/68Ga] 5 has simple labelling
conditions, high stability in vivo, and enters and accumulates in the
liver through phagocytosis by Kupffer cells after intravenous
injection into rats. Biodistribution and small animal PET/CT
studies showed that [68Ga] 5 shows a high uptake in the liver
after intravenous injection. Therefore, PET imaging of [68Ga] 5
can be used to evaluate liver function, locate liver lesions, and
then guide fluorescent removal of the lesions, which is important
for the clinical diagnosis, treatment and postoperative evaluation of
liver disease. Based on this work, PET probes with red AIE
characteristics that target different diseases and complex radio-
active metals are being synthesized in our laboratory to enhance
in vivo PET and fluorescence imaging as well as radioactive
targeted therapy.
3 P. Schmeltzer and K. E. Sherman, Dig. Dis. Sci., 2010, 55,
3328.
4 S. P. Georgiadou, K. Zachou, C. Liaskos, S. Gabeta, E. I.
Rigopoulou and G. N. Dalekos, Liver Int., 2009, 29, 434.
5 J. Zhou, Z. Wang, S. J. Qiu, X. W. Huang, J. Sun, W. Gu and
J. Fan, J. Cancer Res. Clin. Oncol., 2010, 136, 1453.
6 J. Belghiti and D. Fuks, Liver Cancer, 2012, 1, 71.
7 K. Hasegawa, N. Kokudo and M. Makuuchi, Saudi. Med. J.,
2007, 28, 1171.
8 P. Bharti, D. Mittal and R. Ananthasivan, Ultrason Imaging,
2017, 39, 33.
9 T. Tajima, K. Yoshimitsu, H. Irie, A. Nishie, M. Hirakawa,
K. Ishigami, Y. Ushijima, D. Okamoto and H. Honda, Acta
Radiol., 2009, 50, 743.
10 D. R. Martin and R. C. Semelka, Semin Ultrasound CT MR,
2005, 26, 116.
11 D. R. Vera, K. A. Krohn, R. C. Stadalnik and P. O. Scheibe,
J. Nucl. Med., 1984, 25, 779.
12 R. Haubner, A. M. Schmid, A. Maurer, C. Rangger, L. G.
Roig, B. J. Pichler and I. J. Virgolini, Mol. Imaging Biol., 2017,
19, 723.
13 R. Haubner, D. R. Vera, S. Farshchi-Heydari, A. Helbok,
C. Rangger, D. Putzer and I. J. Virgolini, Eur. J. Nucl. Med.
Mol. Imaging, 2013, 40, 1245–1255.
14 W. de Graaf, R. J. Bennink, R. Vetelainen and T. M. van
Gulik, J. Nucl. Med., 2010, 51, 742–752.
15 Y. Chen, M. Li, Y. Hong, J. W. Lam, Q. Zheng and B. Z. Tang,
ACS Appl. Mater. Interfaces, 2014, 6, 10783–10791.
Conflicts of interest
There are no conflicts to declare.
´
´
`
16 E. Ranyuk, R. Lebel, Y. Berube-Lauziere, K. Klarskov,
´
R. Lecomte, J. E. van Lier and B. Guerin, Bioconjugate Chem.,
2013, 24, 1624–1633.
Acknowledgements
This research was supported by the National Natural Science
Foundation of China (81701753) and Scientific Research Common
Program of Beijing Municipal Commission of Education
(KM201810025021). We thank Professor Kung Hank F., Professor
Zhu Lin and all members of their team. We also acknowledge Core
Facility Center of Capital Medical University and Capital Medical
University Beijing Anzhen Hospital PET Center.
17 A. Louie, Chem. Rev., 2010, 110, 3146–3195.
18 D. Patel, A. Kell, B. Simard, B. Xiang, H. Y. Lin and G. Tian,
Biomaterials, 2011, 32, 1167–1176.
19 L. Sun, J. Ding, W. Xing, Y. Gai, J. Sheng and D. Zeng,
Bioconjugate Chem., 2016, 27, 1200.
¨
20 T. K. Forster and K. Ein, Z. Phys. Chem., 1954, 275.
21 L. Perk, G. Visser, M. Budde, M. Vosjan, P. Jurek, G. Kiefer
and G. van Dongen, Facile radiolabeling of monoclonal anti-
bodies and other proteins with zirconium-89 or gallium-68 for
PET imaging using p-isothiocyanatobenzyl-desferrioxamine,
Nature Publishing Group, 2008.
22 J. Mei, N. L. Leung, R. T. Kwok, J. W. Lam and B. Z. Tang,
Chem. Rev., 2015, 115, 11718–11940.
23 X. Li, M. Jiang, J. W. Y. Lam, B. Z. Tang and J. Y. Qu, J. Am.
Chem. Soc., 2017, 139, 17022.
Notes and references
1 T. Muller, S. Mehrle, A. Schieck, U. Haberkorn, S. Urban and
W. Mier, Mol. Pharmaceutics, 2013, 10, 2230.
2 N. D. Shah, M. Ventura-Cots, J. G. Abraldes, M. Alboraie,
A. Alfadhli, J. Argemi, E. Badia-Aranda, E. A. Soler, A. S. T.
Barritt, F. Bessone, M. Biryukova, F. J. Carrilho, M. C.
Fernandez, Z. D. Guiridi, M. El Kassas, T. Eng-Kiong,
A. Farias, J. George, W. Gui, P. Harichander-Thurairajah,
J. C. Hsiang, A. Husic-Selimovic, V. Isakov, M. Karoney,
24 P. Zhang, Z. Zhao, C. Li, H. Su, Y. Wu, R. T. K. Kwok,
J. W. Y. Lam, P. Gong, L. Cai and B. Z. Tang, Anal. Chem.,
2018, 90, 1063.
16312 | New J. Chem., 2019, 43, 16305--16313 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
25 N. Alifu, A. Zebibula, J. Qi, H. Zhang, C. Sun, X. Yu, D. Xue, 33 Y. Ooyama, M. Sugino, T. EnoKi, K. Yamamoto, N. Tsunoji
J. W. Y. Lam, G. Li, J. Qian and B. Z. Tang, ACS Nano, 2018,
12, 11282.
26 C. W. Leung, Y. Hong, S. Chen, E. Zhao, J. W. Lam and
B. Z. Tang, J. Am. Chem. Soc., 2013, 135, 62.
and J. Ohshita, New J. Chem., 2017, 41, 4747.
34 J. Luo, Z. Xie, J. W. Lam, L. Cheng, H. Chen, C. Qiu, H. S.
Kwok, X. Zhan, Y. Liu, D. Zhu and B. Z. Tang, Chem.
Commun., 2001, 1740.
27 R. Zhang, S. H. P. Sung, G. Feng, C. J. Zhang, Kenry, 35 A. Roivainen, T. Tolvanen, S. Salomaki, G. Lendvai, I. Velikyan,
B. Z. Tang and B. Liu, Anal. Chem., 2018, 90, 1154.
P. Numminen, M. Valila, H. Sipila, M. Bergstrom, P. Harkonen,
28 L. Cheng, A. Kamkaew, H. Sun, D. Jiang, H. F. Valdovinos,
H. Lonnberg and B. Langstrom, J. Nucl. Med., 2004, 45, 347.
H. Gong, C. G. England, S. Goel, T. E. Barnhart and W. Cai, 36 Z. Zha, J. Song, S. R. Choi, Z. Wu, K. Ploessl, M. Smith and
ACS Nano, 2016, 10, 7721. H. Kung, Bioconjugate Chem., 2016, 27, 1314.
29 J. Mei, Y. Huang and H. Tian, ACS Appl. Mater. Interfaces, 37 W. Hou, F. Xia, C. S. Alves, X. Qian, Y. Yang and D. Cui, ACS
2018, 10, 12217. Appl. Mater. Interfaces, 2016, 8, 1447.
30 C. Han, S. Jiang, J. Qiu, H. Guo and F. Yang, New J. Chem., 38 Q. Chen, X. Wang, C. Wang, L. Feng, Y. Li and Z. Liu, ACS
2019, 43, 3317. Nano, 2015, 9, 5223.
31 Q. Jiang, Y. Xu, T. Yu, X. Qiu, R. Zhao, D. Zhao, N. Zheng, 39 J. Choi, J. M. Jeong, B. C. Yoo, M. K. Hong, Y. J. Kim, Y. S. Lee,
D. Hu, Z. Xie and Y. Ma, New J. Chem., 2019, 43, 6721. D. S. Lee and J. K. Chung, Nucl. Med. Biol., 2015, 42, 53.
32 P. A. A. M. Vaz, J. Rocha, A. M. S. Silva and S. Guieu, New 40 C. Tsopelas and W. Hsieh, J. Labelled Compd. Radiopharm.,
J. Chem., 2016, 40, 8198. 2017, 60, 528.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16305--16313 | 16313