Fluorescence light-up AIE probe for monitoring cellular alkaline phosphatase activity and detecting osteogenic differentiation

Article abstract of DOI:10.1039/c6tb00828c

Alkaline phosphatase (ALP) is an important monophosphate hydrolase during cell mineralization and osteogenic differentiation. Though traditional methods are provided for evaluating the ALP expression in the fixed and lysed cells, at present it is still ch

Full text of DOI:10.1039/c6tb00828c

Journal of
Materials Chemistry B
View Article Online
View Journal
PAPER
Fluorescence light-up AIE probe for monitoring
cellular alkaline phosphatase activity and
detecting osteogenic differentiation†
Cite this:DOI: 10.1039/c6tb00828c
Feng-Yi Cao, Yue Long, Shi-Bo Wang, Bin Li, Jin-Xuan Fan, Xuan Zeng* and
Xian-Zheng Zhang*
Alkaline phosphatase (ALP) is an important monophosphate hydrolase during cell mineralization and
osteogenic differentiation. Though traditional methods are provided for evaluating the ALP expression in
the fixed and lysed cells, at present it is still challenging to monitor the ALP activity in living cells. In this
work, three phosphorylated tetraphenylethylene (TPE) probes (TPE-PA, TPE-2PA and TPE-4PA) with
different numbers of –PO3H2 groups were synthesized for monitoring the ALP activity. It was found that
in aqueous solution, both the TPE-PA and TPE-2PA probes were highly sensitive to ALP. In the presence
of ALP, they could be quickly hydrolysed, resulting in an aggregation-induced emission (AIE) to light up
ALP. While in living cells, only TPE-2PA showed good cell penetrability and high fluorescence signal-to-
noise ratio during osteogenic differentiation. This probe provides us a new strategy to screen the ALP
activity in living stem cells for detecting osteogenic differentiation.
Received 4th April 2016,
Accepted 3rd June 2016
DOI: 10.1039/c6tb00828c
www.rsc.org/MaterialsB
mineralization.18,19 Therefore, monitoring cellular ALP activity
is of great significance for osteogenic differentiation.
Introduction
Bone marrow mesenchymal stem cells (BMSCs) are multi-potential
Traditional methods for evaluating the cellular ALP activity
stem cells that can be easily harvested from bone marrow and include the reverse transcription polymerase chain reaction
expanded in vitro.1–3 Due to their differentiation potential to specific (RT-PCR) for ALP gene expression, Western blot analysis for
cells and tissues,4 BMSCs have emerged as promising candidates ALP protein content, and enzyme assay based on its catalytic
for regenerative medicine and tissue engineering.5–8 As millions properties.20–22 However, for these methods a process of cell lysis
of patients are suffering from bone diseases worldwide, stem or fixing is required. Moreover, it is impossible to evaluate the
cell-based bone repair is booming as a novel method for bone cellular ALP activity in living cells. Very recently, Chung et al.
diseases.9,10 For stem cell-based therapy, stem cells act as developed a nanoprobe for monitoring the cellular ALP activity
osteoprogenitors for bone regeneration at the defect site.11 And based on the hydrolysis of phosphate groups on hydroxyapatite
specific differentiation factors (physical or chemical stimuli) are and recovering the fluorescence of ICG fluorophores.23 However,
needed to accelerate the osteogenic differentiation and bone the nanoprobe was non-sensitive due to the slow degradation
regeneration process.12–16
process. We also reported a nanoprobe based on the cleavage of
During osteogenic differentiation, a series of enzymes, genes monophosphate from fluoresceinamine isomer I, while the
and proteins are up-regulated, which were identified as osteo- signal-to-background ratio was still unsatisfactory.24 Therefore,
genic markers to evaluate cell osteogenic differentiation.17 it is highly desirable to develop a more sensitive bioprobe with
Among them, alkaline phosphatase (ALP) has been identified higher signal-to-background ratio for monitoring the cellular
as an important marker for osteogenic differentiation. During ALP activity in living stem cells.
osteogenic differentiation, ALP could increase the local concen-
In recent years, bioprobes based on aggregation-induced
tration of inorganic phosphate and promote the process for cell emission (AIE) have emerged as a new class of fluorescence turn-
on probe for biosensing and imaging. And a number of AIE probes
Key Laboratory of Biomedical Polymers of Ministry of Education & Department of
Chemistry, Wuhan University, Wuhan 430072, China. E-mail: zeng_xuan@163.com,
xz-zhang@whu.edu.cn; Fax: +86-27-68754509; Tel: +86-27-68754509
† Electronic supplementary information (ESI) available: Characterization of the
probes; dynamic light scattering characterization; linear relationships between
the probes and the ALP enzyme; and selectivity of the probes. See DOI: 10.1039/
c6tb00828c
have been reported for the detection of cellular proteins and small
molecules.25–30 Compared with traditional fluorophores suffering
from aggregation-induced fluorescence quenching, the fluorogens
with AIE characteristics, such as tetraphenylethylene (TPE)
derivatives, are non-emissive when dissolved in solution and
could emit when aggregated. For most AIE probes, a high
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
6.577–6.400 (m, 2H). TPE-2OH was synthesized from 4-hydroxy-
benzophenone (3.96 g, 20 mmol), zinc dust (1.28 g, 20 mmol)
and TiCl4 (2.19 mL, 20 mmol), and purified by silica gel. Yield:
1
45%. H NMR (300 MHz, DMSO-d6), d (TMS, ppm): 9.485–9.155
(m, 2H), 7.215–7.010 (m, 6H), 6.994–6.830 (m, 4H), 6.802–6.609
(m, 4H), 6.596–6.363 (m, 4H). TPE-4OH was synthesized from
4,40-dihydroxybenzophenone (4.28 g, 20 mmol), zinc dust (1.28 g,
20 mmol) and TiCl4 (2.19 mL, 20 mmol), and purified by silica
1
gel. Yield: 50%. H NMR (300 MHz, DMSO-d6), d (TMS, ppm):
9.348–9.141 (m, 4H), 6.834–6.599 (m, 8H), 6.559–6.355 (m, 8H).
TPE-PA-ET was synthesized from TPE-OH (0.7 g, 2 mmol),
trimethylamine (0.42 mL, 3 mmol) and chlorophosphoric acid
diethyl ester (0.22 mL, 1.5 mmol) in 8 mL dry THF, and the
reaction mixture was stirred at room temperature under N2
overnight. Then the reaction was quenched with 1 mL cool
water, and the organic fraction was extracted and dried with
Na2SO4. The solvent was removed by vacuum rotary evaporation
and purified by silica gel with petroleum ether (60–90 1C)-ethyl
acetate as an eluent. Yield: 70%. 1H NMR (300 MHz, DMSO-d6),
d (TMS, ppm): 7.222–7.056 (m, 9H), 7.044–6.851 (m, 10H),
4.178–3.984 (m, 4H), 1.296–1.115 (m, 6H). TPE-2PA-ET was
synthesized from TPE-2OH (1.1 g, 3 mmol), trimethylamine
(1.26 mL, 9 mmol) and chlorophosphoric acid diethyl ester
(0.66 mL, 4.5 mmol), and purified by silica gel. Yield: 55%.
1H NMR (300 MHz, DMSO-d6), d (TMS, ppm): 7.209–7.070
(m, 6H), 7.055–6.852 (m, 12H), 4.216–3.968 (m, 8H), 1.320–1.075
(m, 12H). TPE-4PA-ET was synthesized from TPE-4OH (1.2 g,
3 mmol), trimethylamine (2.52 mL, 18 mmol) and chlorophos-
phoric acid diethyl ester (1.32 mL, 9 mmol), and purified by silica
gel. Yield: 55%. 1H NMR (300 MHz, DMSO-d6), d (TMS, ppm):
7.107–6.899 (m, 16H), 4.197–3.980 (m, 16H), 1.290–1.085 (m, 24H).
TPE-PA was synthesized from TPE-PA-ET and TMS-I. Briefly,
TPE-PA-ET (0.24 g, 0.5 mmol) was dissolved in 5 mL dry DCM,
then TMS-I (0.21 mL, 1.5 mmol) was added dropwise using a
syringe. The reaction system was tightly sealed and further
reacted for 4 h at room temperature. Then the solvent was
removed by vacuum rotary evaporation and the residue was
purified by washing with DCM several times. The product was
further purified by dissolving in deionized water and lyophilized.
Scheme 1 (A) The molecular structure of TPE-PA, TPE-2PA and TPE-4PA
probes. (B) Illustration of the TPE-nPA (n = 1, 2, 4) probes in response to
ALP in aqueous solution and living osteogenic differentiated stem cells
(osteoblast).
signal-to-background ratio was realized. Though the ALP turn-
on probes based on the phosphorylated TPE derivatives were
reported previously,31,32 their applications in detecting the ALP
activity in living cells, especially for stem cells, are still limited
and remain to be explored.
In this work, three phosphorylated TPE derivative (TPE-PA,
TPE-2PA and TPE-4PA) probes with different numbers of –PO3H2
groups were synthesized (Scheme 1A). With the –PO3H2 group,
the probes are highly water-soluble. It was found that in the
presence of ALP, the TPE-nPA (n = 1, 2, 4) probes could be
hydrolysed leaving the low water-soluble TPE-nOH (n = 1, 2, 4)
residues (Scheme 1B). And the residues could form a large
number of aggregates, resulting in significant fluorescence
increase over the background. The TPE-PA and TPE-2PA probes
were also studied for monitoring the ALP activity during osteo-
genic differentiation.
Experimental section
Synthesis and characterization of the TPE-PA, TPE-2PA and
TPE-4PA probes
1
Yield: 80%. H NMR (300 MHz, DMSO-d6), d (TMS, ppm): 7.225–
The three probes were synthesized in three steps. TPE-OH was 7.033 (m, 9H), 7.020–6.978 (m, 10H); 13C NMR (75 MHz, DMSO-d6), d
synthesized by a McMurry reaction as reported previously.33 (TMS, ppm): 150.35, 150.29, 143.54, 143.79, 140.19, 138.97, 132.06,
Briefly, a mixture of benzophenone (1.82 g, 10 mmol), 4-hydroxy- 130.97, 128.27, 128.18, 128.15, 126.92, 126.84, 119.59, 119.55;
benzophenone (1.98 g, 10 mmol) and zinc dust (1.28 g, 20 mmol) ESI-MS: calcd 428.42; found: 427.04, [M ꢀ H]ꢀ; 854.84,
was dissolved in dry THF and cooled in an ice bath under N2 for [2M ꢀ H]ꢀ; 1282.37, [3M ꢀ H]ꢀ. TPE-2PA was synthesized from
30 min, then TiCl4 (2.19 mL, 20 mmol) was added dropwise. TPE-2PA-ET (0.32 g, 0.5 mmol) and TMS-I (0.43 mL, 3 mmol).
After refluxing overnight, the reaction suspension was cooled to Yield: 70%. 1H NMR (300 MHz, DMSO-d6), d (TMS, ppm): 7.190–
room temperature, poured into 100 mL 10% K2CO3 aqueous 7.027 (m, 6H), 7.015–6.824 (m, 12H); 13C NMR (75 MHz, DMSO-
solution and vigorously stirred for 5 min. The final mixture was d6), d (TMS, ppm): 150.38, 150.32, 143.58, 140.08, 139.00, 132.06,
filtered to remove the insoluble compounds, and extracted with 130.98, 128.17, 126.87, 119.73, 119.68; ESI-MS: calcd 524.40;
ethyl acetate to get the organic fraction. After drying with found: 523.1, [M ꢀ H]ꢀ; 1046.3, [2M ꢀ H]ꢀ. TPE-4PA was
Na2SO4, the solvent was removed by vacuum rotary evaporation. synthesized from TPE-4PA-ET (0.24 g, 0.25 mmol) and TMS-I
The product was finally purified by silica gel with petroleum (0.43 mL, 3 mmol). Yield: 70%. 1H NMR (300 MHz, DMSO-d6), d
ether (60–90 1C)-ethyl acetate as an eluent. Yield: 25%. 1H NMR (TMS, ppm): 7.015–6.853 (m, 16H); 13C NMR (75 MHz, DMSO-
(300 MHz, DMSO-d6), d (TMS, ppm): 9.419–9.244 (m, 1H), d6), d (TMS, ppm): 150.39, 150.33, 139.22, 139.05, 132.06, 119.69;
7.291–7.005 (m, 9H), 7.005–6.838 (m, 6H), 6.785–6.633 (m, 2H), ESI-MS: calcd 716.36; found: 714.9, [M ꢀ H]ꢀ.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Journal of Materials Chemistry B
Paper
Determination of the solubility of TPE-nPA and TPE-nOH in
distilled water
evaluated by MTT and live–dead cell assay. BMSCs at a seeding
density of 5 ꢁ 104 mLꢀ1 were cultured in 96-well or 24-well
plates for 24 h. Then the TPE-PA and TPE-2PA probes at
different concentrations (2.5, 5, 10, 20 and 30 mM) were added
and incubated for 4 h. Then the cells were washed with PBS and
incubated with fresh medium for another 24 h. For MTT assay,
20 mL MTT (5 mg mLꢀ1) were added in each well and managed
according to the standards. The optical density (OD) values
were measured at 570 nm using a microplate reader (Model
550, USA). And the cell toxicity was calculated using the OD
value of the non-treated BMSCs as control. For dead/live cell
assay, cells were stained with fluorescein diacetate/propidium
iodide (FDA/PI) (Sigma, USA) solution for 30 min. And the
fluorescence images were taken on a confocal laser scanning
microscope (C1-Si, Japan).
To evaluate the saturated solubility of TPE-nPA and TPE-nOH,
an excess amount of each compound was added in distilled
water and kept in a shaker (ZHWY-1102C, China) at 25 1C for
24 h to reach equilibrium. Then the samples were centrifuged,
filtered using a 0.22 mm membrane filter and diluted suitably.
The quantitative concentration of each sample was analyzed by
HPLC (LC-6AD, Japan).
Fluorescence response of the probes for detecting ALP activity
in aqueous solution
To evaluate the fluorescence response of the three phosphorylated
TPE (TPE-PA, TPE-2PA and TPE-4PA) probes to ALP, each probe at
10 mM in Tris-HCl buffer (10 mM, pH 8.0) was incubated with ALP
(40 mU mLꢀ1) at room temperature for 5 min or 60 min. Then the Establishment of the osteogenic differentiation model
fluorescence intensities were measured using a LS55 luminescence
spectrometer (Perkin-Elmer, USA). The quantum yield (F) of
TPE-PA, TPE-2PA and TPE-4PA probes after incubating with ALP
For establishing an osteogenic differentiation model, BMSC
culture in the osteogenic differentiation medium (ODM) (high
glucose Dulbecco’s modified Eagle’s medium containing 10 mM
glycerolphosphate, 50 mM ascorbic acid phosphate and 100 nM
for 60 min was evaluated using quinine sulfate as a reference. For
enzyme inhibition assay, ALP (40 mU mLꢀ1) was pre-incubated
dexamethasone, 10% FBS, 1% penicillin–streptomycin) was
with EDTA (0.5 mM) for 30 min before incubating with each probe.
chosen as a model, and BMSC culture in the maintenance
The selectivity of the probes was evaluated by incubating the
medium (DMEM) (low glucose Dulbecco’s modified Eagle’s
medium, 10% FBS, 1% penicillin–streptomycin) was chosen
TPE-PA, TPE-2PA and TPE-4PA probes (10 mM) with the ALP enzyme
and a group of proteins.31,34 For ALP, papain (Pap), trypsin (Try)
as the un-differentiated control. To evaluate the osteogenic
and lysozyme (Lys), the concentration was 40 mU mLꢀ1; for bovine
differentiation, BMSCs were cultured in 24-well or 6-well plates
and the time point after cell culture in the maintenance
medium for 24 h was defined as day 0. Then in one group,
the medium was changed to the osteogenic differentiation
serum albumin (BSA) and a-amylase (Amy), the concentration was
20 nM. After incubating for 60 min, the fluorescence intensities
were measured using a LS55 luminescence spectrometer.
The kinetic enzyme assay was evaluated by monitoring the
medium (marked with ‘‘+’’), and the other group was maintained
fluorescence changes over time. Each probe (10 mM) was incubated
in the maintenance medium (marked with ‘‘ꢀ’’). At each time
with ALP (40 mU mLꢀ1) and the fluorescence intensity with
point (day 0, day 3 and day 7) the osteogenic differentiation
increasing time was recorded from 0 to 300 s. For the TPE-4PA
markers: calcium deposits, ALP and Runx2, were evaluated.
probe, the incubation time was prolonged to 60 min. To explore the
At each time point, BMSCs were fixed with 4% paraform-
mechanism for fluorescence changes, each probe at 40 mM was
aldehyde at room temperature for 15 min. Then the cells were
stained with 1% Alizarin Red S (Sigma, USA) in Tris-HCl buffer
(10 mM, pH 8.3) at 37 1C for 30 min and washed with PBS buffer
(10 mM, PH 7.4) twice. The images were captured using a low
incubated with ALP (160 mU mLꢀ1) for 5 min and the incubation
time of 60 min for the TPE-4PA probe was also evaluated. Then the
aggregation behaviours were analysed by dynamic light scattering
(DLS) (Nano-ZS ZEN3600, UK). The mechanism for the hydrolysis
magnification inverted microscope (Olympusix70, Japan).
of each probe was verified by HPLC. For HPLC analysis, each probe
The cellular ALP and Runx2 protein contents were analysed
by Western blot analysis using glyceraldehyde-3-phosphate
at 40 mM was incubated with ALP (160 mU mLꢀ1) in Tris-HCl buffer
for 60 min, then 20% (v/v) methanol (EtOH) was added to
dehydrogenase (GAPDH) as a protein loading control. Briefly,
completely dissolve the hydrolysis product.
at each time point BMSCs were lysed with RIPA lysis buffer and
To evaluate the ability of the TPE-PA, TPE-2PA and TPE-4PA
probes for quantitatively detecting the ALP activity, each probe
40 mg protein samples were suspended using 5ꢁ SDS sample
buffer. Then the samples were boiled and separated on a 10%
(10 mM) was incubated with different concentrations of ALP
SDS-PAGE. Subsequently, the samples were transferred to a
PVDF membrane, blocked for 1 h and incubated with a primary
antibody GAPDH (1 : 10 000, Abcam, UK), ALP (1 : 1000, Abcam, UK)
and Runx2 (1 : 1000, Boster, China) at 4 1C overnight. Then the
(0–80 mU mLꢀ1) for 5 min. Likewise, ALP (40 mU mLꢀ1) was
incubated with different concentrations of each probe (1–10 mM)
for 5 min. The fluorescence intensities were measured using a
LS55 luminescence spectrometer.
samples were incubated with the secondary antibody HRP–goat
anti rabbit IgG (1 : 10 000, Kirkegaard & Perry Laboratories, Inc.,
USA) at room temperature for 30 min, and the protein contents
Biocompatibility of the TPE-PA and TPE-2PA probes
The bone marrow mesenchymal stem cells (BMSCs) from male were measured using a chemiluminescence kit (Aspen, USA).
Sprague Dawley (SD) rats used in this article were purchased
The gene expressions of ALP and Runx2 were analysed by
from Cyagen Biosciences Inc. (USA). The biocompatibility was the real time reverse transcription polymerase chain reaction
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
1
(rt-PCR) and the transcription of gene encoding ALP (50-TGG Detailed characterization data of the three probes using H NMR,
ACC TCA TCA GCA TTT GG-30; 50-GAG GGA AGG GTC AGT CAG 13C NMR, ESI-MS and HPLC are shown in Fig. S1–S3 (ESI†). The
GTT-30), Runx2 (50-TGA CCA GTC TTA CCC CTC TTA TC-30; purity of TPE-PA, TPA-2PA and TPE-4PA calculated from HPLC was
50-GTC ATC ATC TGA AAT ACG CCT G-30) and the housekeeping 97.774%, 93.624% and 95.572%, respectively. For the three phos-
gene b-actin (50-CGT TGA CAT CCG TAA AGA CCT C-30; 50-TAG phorylated TPE probes, different numbers of –PO3H2 groups were
GAG CCA GGG CAG TAA TCT-30) was evaluated. Briefly, at each included. On the one hand the –PO3H2 groups could act as
time point, BMSCs were collected and lysed with TRIzol reagent hydrophilic parts to enhance the solubility of the probes, giving a
(Invitrogen Life Technologies, USA). cDNA was synthesized low fluorescence background; on the other hand, the –PO3H2
according to the manufacturer’s instruction of the first strand groups could act as specific substrates for ALP detection.
cDNA synthesis kits (Toyobo, Japan). And the gene expressions
The solubility of TPE-nPA and TPE-nOH in distilled water
were measured using a SYBRs Premix Ex Taqt kit (TaKaRa, was evaluated. The solubility of TPE-PA, TPE-2PA and TPE-4PA
Japan) using StepOnet Real-Time PCR (Life technologies, USA). was 10.34 ꢂ 0.47 ꢁ 10ꢀ3, 12.46 ꢂ 1.41, 141.34 ꢂ 1.02 mg mLꢀ1
,
And the results were analysed by the DDCT method using the respectively; the solubility of TPE-OH, TPE-2OH and TPE-4OH
gene expressions of BMSCs at day 0 as a control.
was 8.53 ꢂ 0.60 ꢁ 10ꢀ6, 1.43 ꢂ 0.12 ꢁ 10ꢀ4, 1.88 ꢂ 0.14 ꢁ
10ꢀ3 mg mLꢀ1, respectively. It was demonstrated that the
solubility of TPE-nPA was much better than their presumed
hydrolysis product, making it possible for detecting the ALP
activity by aggregation-induced emission.
Cellular fluorescence response of TPE-PA and TPE-2PA probes
during osteogenic differentiation
To evaluate the ability of TPE-PA and TPA-2PA probes for monitoring
the cellular ALP activity during osteogenic differentiation, BMSCs at
a seeding density of 5 ꢁ 104 mLꢀ1 were cultured on a glass bottomed
dish for 24 h, and this time point was defined as day 0. Then in one
group, the medium was changed to the osteogenic differentiation
medium (marked with ‘‘+’’), and the other group was maintained in
the maintenance medium (marked with ‘‘ꢀ’’). For detecting the
cellular ALP activity, at each time point (day 0, day 3 and day 7) the
cells were incubated with TPE-PA and TPE-2PA probes at 20 mM for
4 h, then the fluorescence images were taken using a confocal laser
scanning microscope. To quantitatively measure the cellular fluores-
cence intensity, at each time point (day 0, day 3 and day 7) the cells
were collected and detected by flow cytometry (BD FACSAriaTM III,
USA) and the mean fluorescence intensity was calculated using the
flowjo software.
The fluorescence response of the three phosphorylated TPE
probes to the ALP enzyme was firstly evaluated in aqueous
solution. After incubating with ALP for 5 min (Fig. 1A–C),
strong fluorescence signals were recorded for TPE-PA and
TPE-2PA probes, while little to no fluorescence increase was
recorded for the TPE-4PA probe. After the incubation time was
prolonged to 60 min (Fig. 1D–F), strong fluorescence signals
were recorded for the three probes, confirming that all the
three probes were fluorescence responsive to the ALP enzyme.
At 60 min, the quantum yield (F) for TPE-PA, TPE-2PA and TPE-
4PA probes was 0.075, 0.118 and 0.101, respectively. Therefore,
the fluorescence intensity of the TPE-2PA probe was much
higher than that of TPE-PA and TPE-4PA probes after incubation
with the ALP enzyme. The maximum excitation and emission
wavelengths of TPE-nPA probes incubated with ALP for 5 min
and 60 min are listed in Table S1 (ESI†). It was found that at
60 min the emission maximum wavelengths were blue shifted.
This may be due to complete hydrolysis of the TPE-PA, TPE-2PA
and TPE-4PA probes and the formation of a higher ordered
structure of the aggregates.35–37
Commercial methods for detecting the ALP activity
To compare our probe with the commercial methods, the ALP
activity during osteogenic differentiation was also evaluated by
BCIP/INT substrate method and ALP assay. For the BCIP/INT
method, at each time point (day 0, day 3 and day 7), BMSCs were
fixed with 4% paraformaldehyde for 15 min, and then the cells
were incubated with the BCIP/INT substrate (Ameresco, USA) at
37 1C for 30 minutes. And the images were captured using a low
magnification inverted microscope. For the ALP assay, at each
time point BMSCs were lysed with 1% Trixon X-100. Then the
content of the total protein was measured using a BCA protein
assay kit (Pierce, USA) and the cellular ALP activity was measured
according to the instruction of the ALP assay kit(Nanjing
Jiancheng Bioengineering Institute, China). The results were
calculated as units of ALP per gram protein (U gprotꢀ1).
Results and discussion
Synthesis and fluorescence response of the probes to the ALP
enzyme in aqueous solution
The synthetic route towards the three phosphorylated TPE
(TPE-PA, TPE-2PA and TPE-4PA) probes is shown in Scheme 2.
Scheme 2 Synthetic route towards TPE-PA, TPE-2PA and TPE-4PA.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Journal of Materials Chemistry B
Paper
no DLS signal was collected for TPE-4PA solution; at 60 min, a
significant DLS signal was also collected for TPE-4PA solution.
These results agreed with the fluorescence changes of the TPE-
4PA probe at 5 min and 60 min. The hydrolysis of the three
probes was verified by HPLC (Fig. 2B and Fig. S7, ESI†), and it
was shown that after incubating for 60 min, the three probes
could be hydrolysed to their corresponding phenol forms,
which confirmed that the aggregates were caused by the
hydrolysis of the phosphorylated TPE probes.
The ability of the three probes for quantitatively detecting
the ALP activity was evaluated by incubating with different
concentrations of the ALP enzyme. At 10 mU mLꢀ1, the fluores-
cence intensity of the TPE-2PA probe was not changed, while a
dramatic fluorescence increase was recorded for the TPE-PA
probe (Fig. 3A), which demonstrated that the TPE-PA probe was
more sensitive at extremely low concentrations of ALP. With the
increasing concentration of ALP, for the TPE-PA probe the
fluorescence intensities increased gradually till 60 mU mLꢀ1
(linearly increased from 10 mU mLꢀ1 to 50 mU mLꢀ1, Fig. S8A,
ESI†), while the fluorescence intensity was not increased with
further increase of the ALP enzyme. For the TPE-2PA probe, the
fluorescence intensities gradually increased till 80 mU mLꢀ1
(linearly increased from 10 mU mLꢀ1 to 40 mU mLꢀ1, Fig. S8B,
ESI†). While for the TPE-4PA probe, no obvious fluorescence
changes were detected with the increasing concentration of the
ALP enzyme. Likewise, ALP (40 mU mLꢀ1) treated with different
concentrations of the TPE-PA, TPE-2PA and TPE-4PA probes
were also studied (Fig. 3B). For the TPE-PA probe, the fluores-
cence increased linearly with the increasing concentration of
the TPE-PA probe from 1 mM to 10 mM (Fig. S8C, ESI†). For the
TPE-2PA probe, the fluorescence intensities gradually increased
till 10 mM (linearly increased from 3 mM to 6 mM, Fig. S8D,
ESI†), while no obvious fluorescence changes were observed
with the increasing concentration of the TPE-4PA probe.
From the above studies, we could conclude that the TPE-PA
and TPE-2PA probes were more sensitive to the ALP enzyme
than the TPE-4PA probe. The fluorescence responses differ
from each other as the molecule structures differ: (1) only
one –PO3H2 group is included in one TPE-PA molecule, while
there are two –PO3H2 groups included in one TPE-2PA molecule
and four –PO3H2 groups included in one TPE-4PA molecule.
Therefore, the interaction between TPE-2PA, TPE-4PA and ALP
Fig. 1 Fluorescence spectra of TPE-PA, TPE-2PA and TPE-4PA probes in the
presence of ALP and the inhibitor: EDTA. (A–C) The incubation time is 5 min;
(D–F) the incubation time is 60 min. [TPE-PA] = [TPE-2PA] = [TPE-4PA] =
10 mM; [ALP] = 40 mU; [EDTA] = 0.5 mM; lex = 335 nm.
The interaction was also confirmed by adding of an inhibitor:
EDTA. In the presence of EDTA, the catalytic ability of ALP was
completely inhibited and no significant fluorescence increase was
recorded at both 5 min and 60 min, confirming the specific
interaction between the phosphorylated TPE probes and the ALP
enzyme. The selectivity of the probes was evaluated by incubating
the TPE-nPA probes with ALP and a group of proteins (Fig. S5,
ESI†). It was demonstrated that the TPE-PA, TPE-2PA and TPE-4PA
probes showed high selectivity to ALP versus other proteins, and
could be used for detecting the ALP enzyme.
The kinetic studies for ALP were performed by monitoring
the fluorescence changes over time. As shown in Fig. 2A, for the
TPE-PA probe the fluorescence intensity increased with the
increasing time, while for the TPE-2PA probe the fluorescence
intensity did not increase before 90 s and for the TPE-4PA probe
the fluorescence intensity did not increase in 300 s. After the
incubation time was prolonged for the TPE-4PA probe, significant
fluorescence increase was not observed till 45 min. For AIE-based
probes, the fluorescence change is induced by the aggregation
of the fluorogens. Therefore, the mechanism for fluorescence
changes was explored by the analysis of the aggregation behaviours
(Fig. S6, ESI†). After incubating for 5 min, significant DLS
signals were formed for TPE-PA and TPE-2PA solutions, while
Fig. 2 (A) Time-dependent fluorescence spectra of the probes upon
addition of ALP at RT: TPE-PA (0–300 s), TPE-2PA (0–300 s), TPE-4PA
(0–300 s), TPE-4PA (15–60 min). [TPE-PA] = [TPE-2PA] = [TPE-4PA] =
10 mM; [ALP] = 40 mU; lex = 335 nm. (B) HPLC traces of the probes before
and after incubation with ALP for 60 min. [TPE-PA] = [TPE-2PA] =
[TPE-4PA] = 40 mM; [ALP] = 160 mU.
Fig. 3 Fluorescence spectra of (A) the probes incubated with different
concentrations of ALP (0–80 mU) for 5 min at RT. [TPE-PA] = [TPE-2PA] =
[TPE-4PA] = 10 mM; lex = 335 nm; lem = 470 nm. (B) ALP incubated with
different concentrations of the probes (1–10 mM) for 5 min at RT. [ALP] =
40 mU; lex = 335 nm; lem = 470 nm.
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
was much complex as there are more –PO3H2 groups that should
be cleaved. (2) With a certain concentration of the probes, the
number of –PO3H2 groups in the TPE-PA probe was only one-half as
that in the TPE-2PA probe. Therefore, the detection limit of the
TPE-PA probe was much lower than that of the TPE-2PA probe.
Considering the above results, only the TPE-PA and TPA-2PA probes
were chosen for the following cellular imaging studies.
The biocompatibility of the TPE-PA and TPE-2PA probes
The biocompatibility of TPE-PA and TPE-2PA probes for BMSCs
was evaluated before cellular imaging studies. As shown in
Fig. 4A, no obvious cell toxicity was found for the TPE-PA probe
at concentrations from 2.5 mM to 30 mM. For the TPE-2PA
probe, at different concentrations from 2.5 mM to 20 mM, no
obvious cell toxicity was observed. When the concentration was
increased to 30 mM, the cell viability decreased to 63%.
The cell toxicity was also evaluated by live/dead cell staining
assay (Fig. 4B). It was found that no PI-stained cells were
observed, confirming that whether at relatively low or high
concentrations, the TPE-PA and TPA-2PA probes would not
damage the cell membrane of BMSCs.
Fig. 5 Evaluation of the osteogenic differentiation model: (A) Alizarin red
S staining for cell mineralization. The scale bars are 50 mm; (B) Runx2 and
ALP protein expressions determined by Western blot analysis; (C) the gene
expressions of Runx2 and ALP measured by rt-PCR. Mean ꢂ SD, **p o 0.01.
Establishment of the osteogenic differentiation model
The osteogenic differentiation medium with a main component
of high glucose maintenance medium and addition of glycero-
phosphate, ascorbic acid phosphate and dexamethasone was a
commercial formulation for inducing osteogenic differentiation.
Therefore, the osteogenic differentiation model was established by
the culture of BMSCs in the osteogenic differentiation medium
(ODM) (marked with ‘‘+’’), and BMSC culture in the maintenance
medium (DMEM) (marked with ‘‘ꢀ’’) was chosen as a control.
As shown in Fig. 5A, at day 0 the cells were slightly stained
with red colour. At day 3, in ODM the cell aggregates were
formed and heavily stained; in DMEM the cells were stained
much weaker. At day 7, in ODM large cell aggregates were
formed and heavily stained, which was much stronger than that
in DMEM. The protein expressions of ALP and Runx2 were
evaluated by Western blot using GAPDH as an internal control
(Fig. 5B) and the semi-quantitative analysis is shown in Fig. S9
(ESI†). At day 3, the protein expressions of ALP and Runx2 in
ODM were 3.1 and 1.33 times that at day 0, which were much
higher than that in DMEM: 0.69 and 0.52 times, respectively. At
day 7, the protein expressions of ALP and Runx2 in ODM were
1.86 and 0.69 times that at day 0, and the protein expressions of
ALP and Runx2 in DMEM were 0.216 and 0.33 times that at
day 0. Although in ODM the protein expression of Runx2 at day
7 was lower than that at day 3, it was still 2.1 times that in
DMEM at day 7. The gene expressions of ALP and Runx2 were
evaluated by rt-PCR using b-actin as a housing keeping gene
(Fig. 5C). Similar to the protein expressions, at day 3 and day 7,
the gene expressions of ALP and Runx2 in ODM were much
higher than that in DMEM.
From the above results, it could be concluded that during
BMSC culture in the osteogenic differentiation medium, the
osteogenic markers were up-regulated, which confirmed that
the BMSCs were undergoing an osteogenesis process. More-
over, with the increasing culture time, the ALP gene and protein
expressions of BMSC culture in the osteogenic differentiation
medium were much higher than that in the maintenance
medium, which confirmed that ALP could be used as a marker
for detecting osteogenic differentiation.
Monitoring cellular ALP activity during osteogenic
differentiation
The ability of the TPE-PA and TPE-2PA probes for monitoring
the cellular ALP activity was studied using a confocal laser
scanning microscope (CLSM). At day 0, the cellular fluores-
cence was observed, which confirmed that the probes were
Fig. 4 Cell viability of BMSCs endocytosed with different concentrations
(2.5, 5, 10, 20 and 30 mM) of the TPE-PA and TPE-2PA probes. (A) Cell
viability evaluated by MTT assay; (B) merged images of live–dead cell assay.
Green: live cells. Red: dead cells. The scale bars are 200 mm.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Journal of Materials Chemistry B
Paper
endocytosed by BMSCs (Fig. 6A). With the increasing culture intensity only slightly changed. We hypothesize that this was
time, for the TPE-PA endocytosed cells though the cellular induced by the limited endocytosis of the TPE-PA probe and
fluorescence in ODM at day 3 and day 7 was slightly stronger rapid consumption of the TPE-PA substrates. For the TPE-2PA
than that in DMEM, the fluorescence was much weaker than endocytosed cells, the cellular fluorescence intensity significantly
that of the TPE-2PA endocytosed cells. For the TPE-2PA endo- increased in ODM. Therefore, we believe that the TPE-2PA
cytosed cells, with the increasing culture time, at day 3 a few probe was more suitable for monitoring the cellular ALP activity
cells were observed with a strong cellular fluorescence and the in mesenchymal stem cells.
cellular fluorescence in most cells was slightly stronger than
that in DMEM. At day 7, in ODM many cells were observed with
a strong fluorescence, which was much stronger than that of
Cellular ALP activity evaluated by commercial methods
the cells in DMEM.
To compare our probe with commercial methods, the cellular
The flow cytometry was used to analyze the percentage of ALP activity was also evaluated by commercial methods: BCIP/
fluorescence-positive cells (Fig. 6B). For the TPE-PA endocytosed INT cell staining method (Fig. 7A1–A5) and ALP assay (Fig. 7B).
cells, with the increasing culture time, at day 3 the percentage of As shown in Fig. 7A1–A5, at day 0 the cells were slightly stained.
positive cells in ODM and DMEM was 17.6% and 9.16%, At day 3, in ODM small cell aggregates were formed and heavily
respectively. At day 7, the percentage only slightly changed. For stained, although most cells were still slightly stained. At day 7,
the TPE-2PA endocytosed cells, at day 3 the percentage of the in ODM large cell aggregates were formed and heavily stained,
positive cells in ODM and DMEM increased to 49.8% and 21.8%, which was much heavier than that at day 3. And the cells in
respectively. At day 7, the percentage continuously increased to DMEM were stained much weaker than that in ODM. As shown
64.0% and 46.1%, respectively. The mean cellular fluorescence in Fig. 7B, at day 3, the cellular ALP activity in ODM and DMEM
intensity was calculated using the flowjo software (Fig. 6C). At was about 1.78 and 1.38 times that at day 0, respectively. At
day 0, the mean fluorescence intensity of TPE-PA endocytosed day 7, the cellular ALP activity in ODM and DMEM was about
cells was much lower than that of the TPE-2PA endocytosed cells, 2.12 and 1.76 times that at day 0, respectively.
which demonstrated that the cell endocytosis of the TPE-2PA
It was demonstrated that both the commercial methods and
probe was much better than the TPE-PA probe. It was found that our probe could be used for detecting the cellular ALP activity
for the TPE-PA endocytosed cells, the cellular fluorescence and similar results were obtained from the three methods. For
the BCIP/INT cell staining method, the BCIP substrate could be
hydrolysed by the cellular ALP, and the hydrolysis product
could further react with the INT substrate and form red brown
precipitates. For ALP assay, the phenyl phosphate disodium
substrate could be hydrolysed by the ALP in cell lysis, and the
hydrolysis product phenol could react with the 4-amino-
antipyrine substrate and can be oxidized by the potassium
ferricyanide substrate, producing quinone derivatives with red
colour. The final product could be quantified by measuring the
optical density. However, for these methods, two or three
substrates were used. Moreover, a process of cell fixing or lysis
was needed, making it impossible for detecting the ALP activity
in living cells.
Fig. 6 The ability of the TPE-PA and TPE-2PA probes for monitoring the
ALP activity at day 0, day 3 and day 7 in ODM (marked with ‘‘+’’) or DMEM
(marked with ‘‘ꢀ’’). (A) Fluorescence images captured by CLSM, and the
fluorescence of TPE-PA and TPE-2PA probes were marked with green
colour. The scale bars are 20 mm; (B) flow cytometry analysis of cellular
fluorescence; (C) mean fluorescence intensity calculated using the flowjo
software.
Fig. 7 (A1–A5) ALP staining of BMSCs at (A1) day 0; (A2) day 3 in ODM;
(A3) day 3 in DMEM; (A4) day 7 in ODM; and (A5) day 7 in ODM. The scale
bars are 50 mm. (B) ALP activity detected using a commercial ALP assay
kit at day 0, day 3 and day 7 in ODM (marked with ‘‘+’’) or DMEM (marked
with ‘‘ꢀ’’). Mean ꢂ SD, **p o 0.01.
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
15 Z. Zhang, M. J. Gupte, X. Jin and P. X. Ma, Adv. Funct. Mater.,
2015, 25, 350–360.
16 A. Carvalho, A. Pelaez-Vargas, D. J. Hansford, M. H. Fernandes
and F. J. Monteiro, Langmuir, 2016, 32, 1091–1100.
17 A. Higuchi, Q. D. Ling, S. T. Hsu and A. Umezawa, Chem.
Rev., 2012, 112, 4507–4540.
18 H. A. Kodama, Y. Amagai, H. Sudo, S. Kasai and S. Yamamoto,
Jpn. J. Oral Biol., 1981, 23, 899–901.
19 E. E. Golub and K. Boesze-Battaglia, Curr. Opin. Orthop.,
2007, 18, 444–448.
Conclusions
In this study, three phosphorylated TPE (TPE-PA, TPE-2PA and
TPE-4PA) probes with different numbers of –PO3H2 groups were
successfully synthesized for detecting the ALP activity in living
cells. In the presence of ALP, the TPE-PA and TPE-2PA probes
could be quickly hydrolysed and form a large number of
aggregates, giving strong fluorescence signals. Additionally, the
TPE-2PA probe was non-invasive and living-cell-permeable, and
exhibited good capability in monitoring the ALP activity in living
stem cells during osteogenic differentiation. The probe demon-
strated here provides us a novel method for imaging the cellular
ALP activity in living stem cells with potential for detecting
osteogenic differentiation.
20 Z. Chen, C. Wu, W. Gu, T. Klein, R. Crawford and Y. Xiao,
Biomaterials, 2014, 35, 1507–1518.
21 J. S. Suh, J. Y. Lee, Y. S. Choi, P. C. Chong and Y. J. Park,
Biomaterials, 2013, 34, 4347–4359.
22 F. Y. Cao, W. N. Yin, J. X. Fan, L. Tao, S. Y. Qin, R. X. Zhuo
and X. Z. Zhang, ACS Appl. Mater. Interfaces, 2015, 7,
6698–6705.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (21474077, 51303137 and 51233003).
´
´
23 J. OakaKeem and B. HyunaChung, Chem. Commun., 2015,
51, 3270–3272.
24 F. Y. Cao, J. X. Fan, Y. Long, X. Zeng and X. Z. Zhang,
Nanomedicine: NBM, 2016, 12, 1313–1322.
25 D. Ding, K. Li, B. Liu and B. Z. Tang, Acc. Chem. Res., 2013,
46, 2441–2453.
26 H. Shi, R. T. Kwok, J. Liu, B. Xing, B. Z. Tang and B. Liu,
J. Am. Chem. Soc., 2012, 134, 17972–17981.
27 H. Shi, J. Liu, J. Geng, B. Z. Tang and B. Liu, J. Am. Chem.
Soc., 2012, 134, 9569–9572.
Notes and references
1 P. Bianco, M. Riminucci, S. Gronthos and P. G. Robey, Stem
Cells, 2001, 19, 180–192.
2 S. Sun, Z. Guo, X. Xiao, B. Liu, X. Liu, P. H. Tang and N. Mao,
Stem Cells, 2003, 21, 527–535.
3 M. Soleimani and S. Nadri, Nat. Protoc., 2009, 4, 102–106.
4 R. O. Oreffo, C. Cooper, C. Mason and M. Clements, Stem 28 Y. Yuan, C. J. Zhang, M. Gao, R. Zhang, B. Z. Tang and
Cell Rev., 2005, 1, 169–178. B. Liu, Angew. Chem., Int. Ed., 2015, 54, 1780–1786.
5 H. Petite, V. Viateau, W. Bensaıd, A. Meunier, C. de Pollak, 29 D. Wu, L. Shao, Y. Li, Q. Hu, F. Huang, G. Yu and G. Tang,
¨
M. Bourguignon, K. Oudina, L. Sedel and G. Guillemin, Nat.
Biotechnol., 2000, 18, 959–963.
6 D. Baksh, L. Song and R. S. Tuan, J. Cell. Mol. Med., 2004, 8,
301–316.
Chem. Commun., 2016, 52, 541–544.
30 Y. Yuan, C. J. Zhang, S. Xu and B. Liu, Chem. Sci., 2016, 7,
1862–1866.
31 J. Liang, R. T. K. Kwok, H. Shi, B. Z. Tang and B. Liu, ACS
Appl. Mater. Interfaces, 2013, 5, 8784–8789.
7 A. I. Caplan, J. Cell. Physiol., 2007, 213, 341–347.
8 A. Freiman, Y. Shandalov, D. Rozenfeld, E. Shor, S. Segal, 32 X. Gu, G. Zhang, Z. Wang, W. Liu, L. Xiao and D. Zhang,
D. Ben-David, S. Meretzki, D. Egozi and S. Levenberg, Stem
Cell Res. Ther., 2016, 7, 1–12.
Analyst, 2013, 138, 2427–2431.
33 L. Zhao, Y. Chen, J. Yuan, M. Chen, H. Zhang and X. Li, ACS
Appl. Mater. Interfaces, 2015, 7, 5177–5186.
34 Z. G. Song, R. T. Kwok, E. G. Zhao, Z. K. He, Y. N. Hong,
J. W. Y. Lam, B. Liu and B. Z. Tang, ACS Appl. Mater.
Interfaces, 2014, 6, 17245–17254.
9 J. M. Seong, B. C. Kim, J. H. Park, I. K. Kwon, A. Mantalaris
and Y. S. Hwang, Biomed. Mater., 2010, 5, 062001.
10 J. Kim, W. G. Bae, H. W. Choung, K. T. Lim, H. Seonwoo, H. E.
Jeong, K. Y. Suh, N. L. Jeon, P. H. Choung and J. H. Chung,
Biomaterials, 2014, 35, 9058–9067.
¨
35 Z. Li, Y. Q. Dong, B. X. Mi, Y. H. Tang, M. Haussler, H. Tong,
11 E. Y. L. Waese, R. R. Kandel and W. L. Stanford, Skeletal
Radiology, 2008, 37, 601–608.
12 D. E. Discher, D. J. Mooney and P. W. Zandstra, Science,
2009, 324, 1673–1677.
13 J. E. Samorezov and E. Alsberg, Adv. Drug Delivery Rev., 2015,
84, 45–67.
Y. P. Dong, J. W. Y. Lam, Y. Ren, H. H. Y. Sung, K. S. Wong,
P. Gao, I. D. Williams, H. S. Kwok and B. Z. Tang, J. Phys.
Chem. B, 2005, 109, 10061–10066.
36 J. Huang, N. Sun, J. Yang, R. L. Tang, Q. Q. Li,
D. G. Ma, J. G. Qin and Z. Li, J. Mater. Chem., 2012, 22,
12001–12007.
14 T. Bai, A. Sinclair, F. Sun, P. Jain, H. C. Hung, P. Zhang, J. R. Ella- 37 W. Bai, Z. Y. Wang, J. Q. Tong, J. Mei, A. J. Qin, J. Z. Sun and
Menye, W. G. Liu and S. Jiang, Chem. Sci., 2016, 7, 333–338.
B. Z. Tang, Chem. Commun., 2015, 51, 1089–1091.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2016