Real-time imaging of alkaline phosphatase activity of diabetes in mice via a near-infrared fluorescent probe

Article abstract of DOI:10.1039/d0cc07292c

A novel water-soluble near-infrared fluorescent probe named QX-P with simple synthesis is developed. QX-P has high sensitivity and selectivity to ALP. Moreover, the probe can not only visualize ALP activity in four cell lines, but also real-time image ALP activity during the diagnosis and treatment of diabetes in mice.

Full text of DOI:10.1039/d0cc07292c

ChemComm
View Article Online
View Journal
COMMUNICATION
Real-time imaging of alkaline phosphatase activity
of diabetes in mice via a near-infrared fluorescent
probe†
Cite this:DOI: 10.1039/d0cc07292c
Received 4th November 2020,
Accepted 1st December 2020
a
a
a
ab
a
Wen-Xin Wang, Wen-Li Jiang, Hong Guo, Yongfei Li
and Chun-Yan Li
*
DOI: 10.1039/d0cc07292c
rsc.li/chemcomm
A novel water-soluble near-infrared fluorescent probe named QX-P Although these provide effective methods for ALP detection, they
with simple synthesis is developed. QX-P has high sensitivity and usually require expensive equipment and complicated operation,
selectivity to ALP. Moreover, the probe can not only visualize ALP and thus are unsuitable for application in organisms. Fluores-
activity in four cell lines, but also real-time image ALP activity cent probes have attracted widespread attention due to their
during the diagnosis and treatment of diabetes in mice.
prominent characteristics such as low cost, high spatial resolu-
7
tion, high sensitivity and good biocompatibility. To date, several
8
Diabetes is a kind of metabolic disease characterized by chronic fluorescent probes have been developed to detect ALP activity,
hyperglycemia, which is caused by insulin secretion disorders. including tetraphenylethylene, naphthalimide, rhodamine B,
It shows many typical symptoms such as polyuria, polydipsia, coumarin and xanthene. However, their application in biological
polyphagia, weight loss and fatigue. If the condition is not systems is hindered due to the short emission wavelength
treated in time, various tissues and organs will be damaged, (o650 nm) of these probes. Based on the above discussion,
especially the kidneys, heart, blood vessels and nervous the development of suitable fluorescent probes to monitor ALP
1
system. Thus, the prevention and diagnosis of diabetes are activity in vivo still has great prospects.
essential for the treatment of the disease. Studies have shown
In recent years, near-infrared (NIR) fluorescent probes have
that an increase in the alkaline phosphatase (ALP) level may be been vigorously developed and researched due to their deep
2
involved in the pathological process of diabetes. ALP is a tissue penetration, little light-damage and low auto-fluorescence
9
hydrolase that can promote the dephosphorylation of a variety interference. Some hemicyanine-based NIR fluorescent probes
1
0
of molecules, and is widely distributed in the liver, kidneys, have been reported for biological imaging of endogenous ALP.
3
bones and placenta of mammals. ALP activity is generally Despite the great efforts, three aspects still need improvement.
4
considered as an important biomarker in clinical diagnosis.
Firstly, the synthetic routes of these reported probes are complex
A large amount of evidence shows that abnormally elevated ALP and depend on the degradation of expensive cyanine dyes
1
0a
level is closely related to the occurrence of many diseases (Fig. 1A).
Secondly, although the emission wavelengths of
including breast cancer, prostate cancer, osteoporosis, hepatic these reported probes are located in the NIR region, they are
5
10a–d
disease, and especially, diabetes. Therefore, it is urgent to basically around 700 nm.
develop an analytical means for real-time monitoring of ALP emission wavelength, the deeper the penetration depth and the
activity during the diagnosis and treatment of diabetes. smaller the auto-fluorescence interference. Thirdly, some probes
So far, many analytical methods for the assay of ALP activity have poor water solubility, which limits their biological applica-
Generally speaking, the longer the
1
0d,e
have been reported, such as chromatography, surface enhanced tion to a certain extent.
Raman scattering, colorimetric methods and electrochemistry.
Therefore, developing a water-soluble
6
NIR fluorescent probe with long emission wavelength is of great
significance for the imaging of ALP activity in vivo.
a
Herein, quinolinium, having a larger conjugated structure and
good water solubility, is used to replace indole in the hemicyanine
to obtain the fluorophore QX, which shows longer absorption and
emission wavelength than the classical hemicyanine dye. The
fluorophore is synthesized by cheap raw materials (Fig. 1B). Then,
the water-soluble NIR fluorescent probe called QX-P is synthe-
sized, which links a phosphate on the hydroxyl group of QX-OH
(Fig. 1C). QX-P itself is non-fluorescent because the intramolecular
Key Laboratory for Green Organic Synthesis and Application of Hunan Province,
Key Laboratory of Environmentally Friendly Chemistry and Applications of
Ministry of Education, College of Chemistry, Xiangtan University,
Xiangtan, 411105, P. R. China. E-mail: lichunyan79@sina.com
b
College of Chemical Engineering, Xiangtan University, Xiangtan, 411105,
P. R. China
†
Electronic supplementary information (ESI) available: Experimental section,
synthesis of probes, additional spectral data, response mechanism and supple-
mental biological assay. See DOI: 10.1039/d0cc07292c
This journal is © The Royal Society of Chemistry 2020
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 1 (A) Previous works: the synthesis of hemicyanine dyes. (B) This
work: the synthesis of quinolone–xanthene dyes. (C) Design of probe
QX-P to ALP.
charge transfer (ICT) effect is blocked by the phosphate group.
After reacting with ALP, the phosphate moiety of QX-P is hydro-
lyzed and the hydroxyl group is released, causing the ICT process
to resume and a strong fluorescent signal at 770 nm appears.
Fig. 2 (A) Absorption spectra of QX-P (10 mM) in the absence (1) and
presence (2) of ALP (1.0 U mL ). Inset: The corresponding color image.
À1
(
B) Fluorescence spectra of QX-P (10 mM) in the presence of various
QX-P shows excellent selectivity and sensitivity to ALP in vitro, and activities of ALP (0, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9,
À1
1
.0 U mL ). (C) Linearity plots of the fluorescence intensity of QX-P versus
thus is applied for visualizing the activity of ALP in different cells.
Furthermore, based on its water solubility and NIR emission
characteristic, QX-P has been successfully applied to real-time
detection of ALP activity in a diabetic mice model. To our knowl-
various activities of ALP. (D) Time-dependent fluorescence response of
À1
QX-P (10 mM) upon the addition of ALP (0, 0.1, 0.5, 1.0 U mL ).
(
(
E) Fluorescence spectra of QX-P (10 mM) in the presence of ALP
À1
1.0 U mL ) at different Na
3
VO
4
concentrations (0, 50, 100, 200, 300,
edge, QX-P is the first reported fluorescent probe to monitor ALP 400, 500, 600, 700, 800 mM). (F) The inhibition efficiency versus Na VO
3
4
concentration. All spectra were measured in Tris–HCl buffer solution
activity in the diagnosis and treatment of diabetes in vivo.
QX-P was synthesized following the synthetic route displayed in
Scheme S1 (ESI†). The detailed synthetic steps and characterization
(
50 mM, pH 8.0) at 37 1C. lex = 720 nm.
are given in the ESI,† (Fig. S1–S6). Firstly, the absorption profiles of solution with pH 8.0 for subsequent testing. Then, the influence
QX-P in the absence and presence of ALP were investigated in Tris– of experimental temperature on the fluorescence response of
HCl buffer solution. As shown in Fig. 2A, the absorption band of QX-P before and after the reaction with ALP was measured
QX-P is at 568 nm. After the addition of ALP, the peak at 568 nm (Fig. S8, ESI†). The results show that the fluorescence intensity
significantly decreases and a new absorption band appears at reaches the maximum at 37 1C. Hence, 37 1C is chosen as the
7
20 nm accompanied by a color variance from purple to blue. optimum temperature for the reaction.
Furthermore, the fluorescence titration experiments between QX-P
The incubation time is another key factor in enzymatic
and ALP were studied. As shown in Fig. 2B, probe QX-P displays a reaction. The time-dependent fluorescence intensity of QX-P
very weak emission at 770 nm. When ALP is added, obvious with various ALP activities was tested. As shown in Fig. 2D, the
fluorescence enhancement at 770 nm is observed. With the fluorescence intensity of QX-P gradually increases with time
À1
increase of ALP activity from 0.0 to 1.0 U mL , the fluorescence and reaches the maximum after 10 min. Therefore, an incuba-
intensity of QX-P enhances around 7-fold. As shown in Fig. 2C, it tion time of 10 min for ALP testing is accepted. Moreover, to
was found that the fluorescent intensity has a linear relationship optimize the final concentration of QX-P used in the experi-
À1
with ALP activity in the range 0.05–1.0 U mL . And through ment, probes of various concentrations were incubated with
calculation (ESI†), the detection limit (LOD = 3d/k) for ALP is unchanging ALP activity. The enzyme kinetics experiment was
À1
0
.017 U mL . The above experimental results indicate that QX-P described by the Michaelis–Menten equation and the result is
can quantitatively detect ALP activity with high sensitivity.
displayed in Fig. S9A (ESI†). The kinetic parameters were
The effect of pH on the fluorescence of QX-P was investigated calculated by converting the curve to a linear plot by Line-
as pH is a crucial factor in the enzymatic reaction. As shown in weaver–Burke analysis (Fig. S9B, ESI†). According to the slope
Fig. S7 (ESI†), the pH of maximum fluorescence response for and intercept of the fitted line, the maximum velocity (Vmax
)
À1
ALP is about pH 8.0. Therefore, we chose Tris–HCl buffer and Michaelis constant (K
m
) were 0.491 mM min and 8.90 mM.
Chem. Commun.
This journal is © The Royal Society of Chemistry 2020

View Article Online
ChemComm
Communication
As we all know, the substrate concentration must be higher
than its K value. Therefore, 10 mM is selected as the optimum
m
QX-P concentration for the ALP assay.
The selectivity of the probe to ALP was studied. From
Fig. S10 (ESI†), the potential analytes are tested, such as
+
2+
2+
À
À
À
1
00 mM metal ions (Na , Ca , Mg ), anions (Cl , Br , OH ),
À
À
reactive oxygen species (ClO , O
Cys), amino acids (Tyr, Pro, Phe, Leu), and 1.0 U mL enzymes
2
, H
2
O
2
), biothiols (GSH, Hcy,
À1
(acetylcholinesterase, AChE; g-glutamyl transpeptidase, GGT;
phosphodiesterase, PDE; acid phosphatase, ACP). Only ALP can
display obvious change of fluorescence intensity, while the
other analytes cannot change the fluorescence signal of the
probe. This may be owing to the fact that only ALP can catalyze
the dephosphorylation process of the probe, and thus a strong
fluorescent signal is released. The results indicate that QX-P
possesses high selectivity to ALP.
Fig. 3 (A) Fluorescence images of HeLa cells. Control group: the cells
The probe (QX-P) could be used for the screening of ALP were untreated with QX-P; probe group: the cells were treated with QX-P
only; inhibitor + probe group: the cells were treated with Na VO and
QX-P. (B) Relative pixel intensity in (A). (C) Flow cytometry analysis.
ex = 640 nm, lem = 700–775 nm. Scale bar: 10 mm.
3
4
inhibitors. To prove this probability, sodium orthovanadate
Na VO ) was studied (Fig. 2E). The fluorescence intensity
declines significantly with the increase of Na VO4 concen-
(
3
4
l
3
tration. For comparison, the influence of Na VO4 on the
3
fluorescence intensity of QX-OH was also tested, and negligible was performed. Meanwhile, flow cytometry analysis was performed
changes in fluorescence were observed (Fig. S11, ESI†). The (Fig. 3C and Fig. S18, ESI†) and the results were consistent with
3 4
above results indicate that Na VO
can effectively inhibit ALP the above confocal microscopy images. The above experimental
activity. As shown in Fig. 2F, the IC50 value (the concentration phenomenon indicates that QX-P can well monitor the activity of
of an inhibitor reaching 50% of inhibition efficiency) of Na
3
VO
4
ALP in cancer cells.
was estimated to be 109.6 mM. The results strongly suggest that
Motivated by the excellent performance of living cells, the
our method can be used to monitor ALP and screen the ability of probe QX-P for the fluorescence imaging of ALP
potential inhibitors of ALP. Moreover, a possible response activity in vivo was investigated. As shown in Fig. 4A and B,
mechanism of QX-P to ALP was proposed in Scheme S2 (ESI†) the fluorescence intensity of normal group mice slowly
and proved by the MS, HPLC and DFT calculation (Fig. S12–S14, increases with time, which indicates that QX-P can visualize
ESI†).
ALP activity in normal mice. The diabetes mice are obtained
Next, the cytotoxicity of QX-P was investigated in HeLa, by intraperitoneally injecting streptozotocin, an analog of
HepG2, HCT116 and 4T1 cells by MTT assay (Fig. S15, ESI†). N-acetylglucosamine, which is a specific toxin for the pan-
1
2
More than 90% cell viability is found after the cells are creatic beta cells. At the same incubation time, the fluores-
incubated for 24 h with different concentrations of QX-P from cence signal of the diabetic mice group is significantly stronger
0
to 30 mM, which indicates that the probe has low cytotoxicity than that of the normal group, indicating that ALP activity in
in the cells. As we all know, ALP is overexpressed in HeLa, diabetic mice is significantly upregulated. Then, the diabetic
1
1
HepG2, HCT116 and 4T1 cells. Thus, the four types of cells mice were treated with metformin, which is a biguanide
were used as models for intracellular imaging of ALP activity. As derivate used as an oral hypoglycaemic drug in diabetics, and
shown in Fig. 3A and Fig. S16 (ESI†), the four types of cells has a protective effect on the organ of streptozotocin-diabetic
1
3
untreated with QX-P show hardly any fluorescence signal, which mice. It is observed that the fluorescence signal in the mice
shows that the possibility of auto-fluorescence interference is reduces significantly after the treatment, indicating the down-
ruled out. After the cells are incubated with QX-P, a significant regulation of ALP in the treatment group due to the remission
red fluorescence is observed, which is caused by the presence of of symptoms by the drug.
ALP in the four cells. Meanwhile, the fluorescence signal gradually
Subsequently, the mice were dissected and the main organs
enhances with time and reaches the maximum at 30 min, which were acquired. As shown in Fig. 4C, the liver and kidneys of the
shows that QX-P has good cell permeability and can react with diabetic mice present relatively stronger fluorescence than the
ALP in the cells (Fig. S17, ESI†). In order to prove the fact that the other organs, whereas the liver and kidneys of the normal mice
fluorescence signal was activated by the endogenous ALP in the exhibit weak fluorescence. Additionally, the liver and kidneys of
four cells, the cells were pretreated with Na VO (ALP inhibitor) the treatment group mice display weaker fluorescence signal
3
4
and then treated with QX-P. As expected, a negligible fluorescence compared with diabetic mice. This is because the liver and
signal in the cells was observed. The above results indicate that kidneys are the major metabolic organs and a large amount of
the enhancement of fluorescence signal in the four cells indeed ALP accumulates in those two organs under diabetic condi-
resulted from endogenous ALP. To more intuitively reflect the tions. Inspired by this fact, QX-P was used to study the ALP
changes in the intracellular fluorescence, 3D imaging on each cell activity in mouse blood through fluorescence imaging (Fig. 4D).
This journal is © The Royal Society of Chemistry 2020
Chem. Commun.

View Article Online
Communication
ChemComm
Notes and references
1
2
P. Dayhaw, Optom. Vis. Sci., 1995, 72, 417–424.
(a) B. Chua and E. Shrago, J. Nutr., 1978, 108, 196–202; (b) J. St ˘e p ´a n,
T. Havr ´a nek, J. Form ´a nkov ´a , J. Skrha, F. Skrha and V. Pacovsk ´y , Clin.
Chim. Acta, 1980, 105, 75–81.
3
4
(a) J. E. Coleman, Annu. Rev. Biophys. Biomol. Struct., 1992, 21,
4
41–483; (b) J. L. Millan, Purinergic Signalling, 2006, 2, 335–341.
(a) P. H. Lange, J. L. Millan, T. Stigbrand, R. L. Vessella, E. Ruoslahti
and W. H. Fishman, Cancer Res., 1982, 42, 3244–3247; (b) J. B.
Warshaw, J. W. Littlefield, W. H. Fishman, N. R. Inglis and
L. L. Stolbach, J. Clin. Invest., 1971, 50, 2137–2142; (c) J. Stebbing,
L. C. Lit, H. Zhang, R. S. Darrington, O. Melaiu, B. Rudraraju and
G. Giamas, Oncogene, 2014, 33, 939–953.
5
(a) H. G. Wada, J. E. Shindelman, A. E. Ortmeyer and H. H. Sussman,
Int. J. Cancer, 1979, 23, 781–787; (b) J. D. Schmidt, N. H. Slack,
T. M. Chu and G. P. Murphy, J. Urol., 1982, 127, 457–459;
(
c) P. A. Kyd, K. D. Vooght, F. Kerkhoff, E. Thomas and A. Fairney,
Ann. Clin. Biochem., 1998, 35, 717–725; (d) U. Domar, A. Danielsson,
K. Hirano and T. Stigbrand, Scand. J. Gastroenterol., 1988, 23,
7
93–800; (e) D. B. Maxwell, E. A. Fisher, H. A. Ross-Clunis and
H. L. Estep, J. Am. Coll. Nutr., 1986, 5, 55–59.
Fig. 4 (A) Fluorescence images of mice from the normal group, diabetes
group, and treatment group after injection of QX-P (200 mM) with time
6
7
(a) T. Hasegawa, M. Sugita, K. Takatani, H. Matsuura, T. Umemura
and H. Haraguchi, Bull. Chem. Soc. Jpn., 2006, 79, 1211–1214;
(
0, 10, 20, 40, 60 min). (B) Relative fluorescence intensity in (A). The
fluorescence intensity of the normal group at 60 min is defined as 1.0.
C) Fluorescence images of major organs. 1: liver, 2: heart, 3: spleen, 4:
lungs, 5: kidneys. (D) Fluorescence images of blood from the mice.
ex = 640 nm, lem = 680–780 nm.
(
3
b) C. Ruan, W. Wang and B. Gu, Anal. Chem., 2006, 78,
379–3384; (c) C. M. Li, S. J. Zhen, J. Wang, Y. F. Li and
(
C. Z. Huang, Biosens. Bioelectron., 2013, 43, 366–371; (d) K. Ino,
Y. Kanno, T. Arai, K. Y. Inoue, Y. Takahashi, H. Shiku and T. Matsue,
Anal. Chem., 2012, 84, 7593–7598.
(a) H. W. Liu, L. L. Chen, C. Y. Xu, Z. Li, H. Y. Zhang, X. B. Zhang and
W. H. Tan, Chem. Soc. Rev., 2018, 47, 7140–7180; (b) J. N. Liu, W. B. Bu
and J. L. Shi, Chem. Rev., 2017, 117, 6160–6224; (c) H. Singh, K. Tiwari,
R. Tiwari, S. K. Pramanik and A. Das, Chem. Rev., 2019, 119,
l
An obvious fluorescence signal is seen in the diabetes group
compared to the normal group, and a significantly weakened
fluorescent signal is observed in the treatment group, which is
consistent with the above experimental phenomenon. This
result indicates that QX-P can be used to monitor the activity
of ALP in the diagnosis and treatment of the diabetic mice.
In conclusion, we have, for the first time, reported a water-
soluble NIR fluorescent probe (QX-P) based on quinolinium–
xanthene dye for detecting ALP activity. Compared with classic
hemicyanine dye, the fluorophore QX not only has longer absorp-
tion and emission wavelengths, but also has simple synthesis
steps and cheap raw materials. QX-P displays excellent sensitivity
and selectivity to ALP. Owing to the low cytotoxicity, the probe
QX-P can visualize the activity of ALP in different cell lines with
outstanding performance. Furthermore, QX-P is used to prove
that the ALP level of the diabetic mice is higher than that
1
1718–11760; (d) X. F. Wu, W. Shi, X. H. Li and H. M. Ma, Acc. Chem.
Res., 2019, 52, 1892–1904; (e) Y. H. Tang, Y. Y. Ma, J. L. Yin and
W. Y. Lin, Chem. Soc. Rev., 2019, 48, 4036–4048.
8 (a) J. Liang, R. T. K. Kwok, H. Shi, B. Z. Tang and B. Liu, ACS Appl.
Mater. Interfaces, 2013, 5, 8784–8789; (b) X. F. Hou, Q. X. Yu, F. Zeng,
J. H. Ye and S. Z. Wu, J. Mater. Chem. B, 2015, 3, 1042–1048;
(
c) L. Dong, Q. Miao, Z. Hai, Y. Yuan and G. Liang, Anal. Chem.,
2015, 87, 6475–6478; (d) T. Kim, H. J. Kim, Y. Choi and Y. Kim,
Chem. Commun., 2011, 47, 9825–9827; (e) H. M. Zhang, C. L. Xu,
J. Liu, X. H. Li, L. Guo and X. M. Li, Chem. Commun., 2015, 51,
7031–7034; ( f ) F. Ma, M. Liu and C. Y. Zhang, Chem. Commun.,
2019, 55, 8963–8966.
(a) J. Huang, Y. L. Wu, F. Zeng and S. Z. Wu, Theranostics, 2019,
9
9
, 7313; (b) W. Fu, C. X. Yan, Z. Q. Guo, J. J. Zhang, H. Y. Zhang,
H. Tian and W. H. Zhu, J. Am. Chem. Soc., 2019, 141, 3171–3177;
(c) M. Gao, R. Wang, F. B. Yu and L. X. Chen, Biomaterials, 2018, 160,
1
–14; (d) S. Y. Gong, E. B. Zhou, J. X. Hong and G. Q. Feng, Anal.
Chem., 2019, 91, 13136–13142; (e) Z. Q. Guo, S. Park, J. Yoon and
I. Shin, Chem. Soc. Rev., 2014, 43, 16–29.
of normal mice. After treatment with a hypoglycemic drug 10 (a) Y. Tan, L. Zhang, K. H. Man, R. Peltier, G. C. Chen, H. T. Zhang,
L. Y. Zhou, F. Wang, D. Ho, S. Q. Yao, Y. Hu and H. Y. Sun, ACS Appl.
(
metformin), the production of ALP in the diabetic mice is
Mater. Interfaces, 2017, 9, 6796–6803; (b) C. S. Park, T. H. Ha,
M. Kim, N. Raja, H. S. Yun, M. J. Sung, O. S. Kwon, H. Yoon and
C. S. Lee, Biosens. Bioelectron., 2018, 105, 151–158; (c) Y. Y. Li,
H. Song, C. H. Xue, Z. J. Fang, L. Q. Xiong and H. X. Xie, Chem.
Sci., 2020, 11, 5889–5894; (d) H. W. Liu, X. X. Hu, L. M. Zhu, K. Li,
Q. M. Rong, L. Yuan, X. B. Zhang and W. H. Tan, Talanta, 2017, 175,
421–426; (e) L. B. Xu, X. He, Y. B. Huang, P. Y. Ma, Y. X. Jiang, X. Liu,
S. Tao, Y. Sun, D. Q. Song and X. H. Wang, J. Mater. Chem. B, 2019, 7,
significantly reduced. These results indicate that QX-P could
serve as a powerful tool for further studying the clinical value
of ALP in the diagnosis and treatment of diabetes.
This work was supported by the National Natural Science
Foundation of China (21775133), Hunan Provincial Natural
Science Foundation (2018JJ2385), Scientific Research Fund of
Hunan Provincial Education Department (19A479), Degree &
Postgraduate Education Reform Project of Hunan Province
1284–1291; ( f ) X. T. Gao, G. C. Ma, C. Jiang, L. L. Zeng, S. S. Jiang,
P. Huang and J. Lin., Anal. Chem., 2019, 91, 7112–7117; (g) S. J. Li,
C. Y. Li, Y. F. Li, J. J. Fei, P. Wu, B. Yang, J. Ou-Yang and S. X. Nie,
Anal. Chem., 2017, 89, 6854–6860.
(2019JGYB113) and Degree & Postgraduate Education Reform
1
1 (a) D. Di Lorenzo, A. Albertini and D. Zava, Cancer Res., 1991, 51,
Project of Hunan Province (CX20190483, XDCX2020B115).
4470–4475; (b) F. Herz, Experientia, 1985, 41, 1357–1361;
(
c) H. Matsumoto, R. H. Erickson, J. R. Gum, M. Yoshioka,
E. Gum and Y. S. Kim, Gastroenterology, 1990, 98, 1199–1207.
2 V. Burkart, Z. Q. Wang, J. Radons, B. Heller, Z. Herceg, L. Stingl,
E. F. Wagner and H. Kolb, Nat. Med., 1999, 5, 314–319.
3 D. Stepensky, M. Friedman and I. Raz, Drug Metab. Dispos., 2002, 30,
861–868.
1
1
Conflicts of interest
There are no conflicts to declare.
Chem. Commun.
This journal is © The Royal Society of Chemistry 2020