19F-MRS/1H-MRI dual-1 of β-galactosidase activity

Article abstract of DOI:10.1039/c3sc21099e

Because of the importance of lacZ gene in various applications ranging from molecular biology to clinical trials, the development of non-invasive bimodal techniques for improving precision and accuracy to assay gene expression has attracted much attention. In this paper, we propose a dual-function probe for synergistic combination of ¹⁹F-MRS/¹H-MRI to simultaneously detect β-gal activity. Based on this strategy, we have designed, synthesized and characterized a series of ¹⁹F-MRS/ ¹H-MRI reporters, and demonstrated the feasibility of 1-O-(β-d-galactopyranosyl)-3-fluorocatechol MGD-3-FCAT for assessing β-gal activity in solution and in vitro with lacZ transfected tumor cells as well, by the characterization of β-gal responsive ¹⁹F- chemical shift changes Δδ_F, hydrolytic kinetics, and T₁, T₂ relaxation mapping.

Full text of DOI:10.1039/c3sc21099e

Chemical Science
View Article Online
View Journal | View Issue
EDGE ARTICLE
19F-MRS/1H-MRI dual-function probe for detection of
b-galactosidase activity†
Cite this: Chem. Sci., 2013, 4, 2132
a
Jian-Xin Yu, Vikram D. Kodibagkar,b Li Liu,c Zhongwei Zhang,a Li Liu,a
*
Jennifer Magnussona and Yuting Liud
Because of the importance of lacZ gene in various applications ranging from molecular biology to clinical
trials, the development of non-invasive bimodal techniques for improving precision and accuracy to assay
gene expression has attracted much attention. In this paper, we propose a dual-function probe for
synergistic combination of 19F-MRS/1H-MRI to simultaneously detect b-gal activity. Based on this
strategy, we have designed, synthesized and characterized a series of 19F-MRS/1H-MRI reporters, and
demonstrated the feasibility of 1-O-(b-D-galactopyranosyl)-3-fluorocatechol MGD-3-FCAT for assessing b-
gal activity in solution and in vitro with lacZ transfected tumor cells as well, by the characterization of b-
gal responsive 19F-chemical shift changes DdF, hydrolytic kinetics, and T1, T2 relaxation mapping.
Received 28th July 2012
Accepted 25th February 2013
DOI: 10.1039/c3sc21099e
www.rsc.org/chemicalscience
high gyromagnetic ratio (g ¼ 40.05 MHz Tꢀ1), 100% natural
isotopic abundance, large chemical-shi dispersion, impor-
Introduction
tantly, essentially no intrinsic 19F signals detectable in living
tissue and thus no interference from background signals.18,24 So
far aside from the lacZ gene 19F-MRS/MRI reporters, there have
been varieties of other functional 19F-NMR probes developed for
studies of diverse biological events.28–38
As a reporter, the lacZ gene encoding b-galactosidase (b-gal) has
been widely applied in molecular biology, small animal inves-
tigations and clinical trials, including assays of clonal insertion,
transcriptional activation and protein expression, as well as
protein interaction.1–3 Therefore, its detection has been exploi-
ted with techniques from colorimetric,4 uorescence,5–7 chem-
iluminescence,8,9 positron emission tomography or single-
photon emission computed tomography,10,11 to magnetic reso-
nance imaging (MRI)12–17 and 19F-MRS/MRI.18–26
However, unlike the mechanism of 1H-MRI, which visualizes
numerous water molecules around the probes, 19F-MRI inten-
sity is only attributed from the probe signals or content,
resulting in poor sensitivity,18,24,37 which is the major obstacle of
this technique for further applications. Different from 19F-MRI,
19F-MRS has been widely involved in in vivo studies on drug
absorption, distribution, metabolism and excretion for reasons
of its favorable MR properties, simplicity and high sensitivity.39
In this paper, we propose a novel dual-function probe for
synergistic combination of 19F-MRS/1H-MRI to detect b-gal
activity, in which 19F-MRS recognizes its presence and
progression by dening the 19F chemical shi change
and hydrolytic kinetics, whereas 1H-MRI reveals its location and
magnitude by high spatial resolution and 3-dimensional
anatomical details.
Comparatively, 1H-MRI is a superior imaging modality in
living systems because it is non-invasive, able to differentiate
between so tissues, and has high lateral and depth resolu-
tions. Currently, 1H-MRI is widely applied for medical diag-
nosis, and becomes one of the most promising techniques for
investigation of physiological events.27 However, contrast agent
enhanced 1H-MRI oen suffers from low contrast-to-noise ratio
because of the large background signals. To avoid this limita-
tion, 19F-MRI recently has been attracting considerable atten-
tion as an alternative technique, since the 19F nucleus has a
aDepartment of Radiology, The University of Texas Southwestern Medical Center at
Dallas, 5323 Harry Hines Blvd., E6.118, Dallas, Texas 75390-9058, USA. E-mail:
Jian-Xin.Yu@UTSouthwestern.edu; Fax: +1 214-648-4538; Tel: +1 214-648-2716
Probe design
bCurrently at School of Biological and Health Systems Engineering, Arizona State
Cancer cells exhibit increased uptake and utilization of more
University, Tempe, AZ 85287-9709, USA
cXinjiang Institute of Chemistry and Physics, Chinese Academy of Sciences, 40 Beijing Fe3+ because of their rapid replication.40,41 By chelating with
cancer cellular Fe3+ to reduce its levels, various critical cellular
South Road, Urumqi, Xinjiang 830011, China
dDepartment of Chemistry, Xinjiang University, 14 Victory Road, Urumqi, Xinjiang
processes dependent on Fe3+ in cancer are disrupted. Through
830046, China
this mechanism, chelators are employed to inhibit tumor
growth as iron chelation therapy. Pyridoxal iso-
† Electronic supplementary information (ESI) available: Fig. S1–S5, syntheses of
ligands and their corresponding Fe-complexes as well as molecular
nicotinoylhydrazone (PIH), salicylaldehyde benzoylhydrazone
(SBH), salicylaldehyde isonicotinoylhydrazone (SIH) and
characterization of 3- and 4-FCAT b-D-galactopyranosides. See DOI:
10.1039/c3sc21099e
2132 | Chem. Sci., 2013, 4, 2132–2142
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
salicylaldehyde nicotinoylhydrazone (SNH) (Fig. 1 and S1†) are
in clinical trials for the treatment of various metastatic and
solid cancers, and uorinated chelators have shown
pronounced anticancer activities, even more effective than each
of their parent chelators.40,41
Results and discussion
1H-MRI verication
The clinically applied tridentate chelators PIH, SBH, SIH and
SNH, and their corresponding 2 : 1 Fe-complexes [Fe(PIH-H)2]+
(PIH/Fe3+), [Fe(SBH-H)2]+ (SBH/Fe3+), [Fe(SIH-H)2]+ (SIH/Fe3+),
[Fe(SNH-H)2]+ (SNH/Fe3+) were synthesized according to the
described methods (see ESI†).40,41 T1/T2/T2*-weighted 1H-MRI
evaluations were investigated by using a spin echo pulse
sequence at varying repetition times (TRs) and echo times (TEs).
As we expected, the results showed that the complexes PIH/Fe3+,
SBH/Fe3+, SIH/Fe3+ or SNH/Fe3+ each produced a substantial T1-
weighted contrast (Fig. 3), which opens a new era of an Fe-
depletion complex functioning as a 1H-MRI agent, and also
suggests a potential for clinical theranostic application.
From the structural similarities of the Fe-complexes of the
clinically applied chelators and the known Fe-based 1H-MRI
contrast agents,42 we speculated that the clinically applied Fe-
1
chelators could act as Fe-based H-MRI contrast agents. When
using uorinated chelators as aglycons, their b-D-galactosides
will work as prodrugs, in which the sugar residue b-D-gal-
actopyranosyl will act as a carrier and masking unit. Upon
cleavage at a lacZ transfected tumor, the blocked uorinated
chelators will be lacZ (or b-gal)-specically released and acti-
vated to capture the tumor abundant Fe3+. The in situ formed
paramagnetic Fe-complexes will (1) generate 1H-MRI contrast
and (2) localize and accumulate 19F-MRS signals at the site of b-
gal activity. Fig. 2 illustrates the design principle of 19F-MRS/1H-
19F-MRS/1H-MRI dual-function conrmation
MRI dual-function probe for detecting b-gal activity, using SBH Following the strategy described in Fig. 2, we rstly prepared
as a model chelator.
tridentate 2-hydroxyl-5-uorobenzaldehyde benzoylhydrazone
However, the paramagnetic metal ions like Gd3+, Fe3+ and (p-FSBH) according to the previously described method (see
Mn2+ can shorten the relaxation time of 19F nucleus by the ESI†).25,40,41 F-NMR tests showed that p-FSBH in 1 : 1 (v/v0)
19
paramagnetic relaxation enhancement (PRE) effect,43 in which DMSO/PBS (0.1 M, pH ¼ 7.4) at 25 ꢁC gave a single narrow 19F-
the distance between the 19F nucleus and paramagnetic center MRS signal with respect to sodium triuoroacetate (NaTFA, dF ¼
is vital for the PRE modulation.44 Based on this mechanism, 0 ppm) at dF ¼ ꢀ49.60 ppm, which was attenuated to vanished
Kikuchi et al. developed off-on switching 19F-MRI probes such very quickly upon the addition of a half equivalent ferric
as Gd-DFP-Gal, Gd-DOTA-DEVD-T, Gd-DOTA-DEVD-AFC and ammonia citrate (FAC), due to the strong intramolecular PRE
Gd-FC-Lac to detect b-gal, protease or b-lactamase activities in effect on the 5-19F nucleus from the in situ generated para-
in vitro.26,34,35
magnetic 2 : 1 Fe-complex [Fe(p-FSBH-H)2]+ (p-FSBH/Fe3+), it
In contrast to Gd3+ with very strong relaxivity, Fe3+ has a indicates that the 19F nucleus in the para-position to 2-hydroxyl
much shorter electronic relaxation time, and limits the relaxa- is still in the range of PRE. The distance from 19F to para-
tion enhancing efficiency on 19F nucleus. Importantly, other magnetic Fe3+ r ¼ 7.01 A was calculated from the computed
˚
than Gd3+-complexes which are isotropic electron systems, the minimized energy conformation of Fe-complex p-FSBH/Fe3+
Solomon–Bloembergen theory assumes that Fe3+-complexes using Chem3D 8.0 (MM2 as the force-eld) and Visual Molec-
represent anisotropic paramagnetic systems, in which the Fe3+
-
ular Dynamics (VMD 1.9). The T1-weighted images showed that
complex PRE interaction vectors are rigid in the molecular the contrast enhancement of Fe-complex p-FSBH/Fe3+ in Fig. 4
frame.43 This offers us a possibility of constructing a novel was similar to SBH/Fe3+ in Fig. 3. However, when the 19F-atom
probe with an 19F-atom closer to the paramagnetic Fe3+ center was introduced in the para-position to benzoylhydrazone
in a conjugated molecule as designed in Fig. 2, which is (SBHF-p) with the distance to paramagnetic Fe3+ of r ¼ 8.10 A,
˚
expected to have a large 19F-NMR chemical shi difference upon its 19F-MRS signal intensity was reduced by 22% from a sharp
b-gal cleavage, but without an Fe3+–19F PRE effect.
peak to a broad singlet at dF ¼ ꢀ32.66 ppm upon the formation
This concept of probe design is based on tumor biology and of Fe-complex [Fe(SBHF-p-H)2]+ (SBHF-p/Fe3+). Also, signicant
synergizes the advantages of 19F-MRS and H-MRI. The mech- contrast change in T1-weighted MR images was observed
1
anism proceeds in the b-gal responsive “turn-on” way to (Fig. 5). The identical T1-weighted images compared with the
simultaneously generate 1H-MRI contrast and 19F-MRS signal prepared [Fe(SBHF-p-H)2]+ (SBHF-p/Fe3+) conrmed the attri-
shi, which are selectively and specically conned and gath- bution to the intramolecular Fe3+–19F PRE effect.
ered at the site of b-gal activity, offering increases of the
Inspired by these results, we turned to the bidentate catechol
magnitudes of both signals and reliability of detection, as well analogues, which also are the most widely used Fe3+ chela-
as better precision and accuracy.
tors,40,41 and identied recently as the endogenous ligands for
siderocalin to deliver Fe3+ to the kidney.45 Catechol moieties
possess a high affinity for Fe3+ to form thermodynamically
stable 1 : 3 Fe3+-complexes at physiological pH, and discrim-
inately coordinate to other biologically critical metal ions such
as Zn2+, Cu2+ and Ca2+. This unique specicity is very essential
in the development of clinically useful metal–ligand systems.
Moreover, catecholate ligands are of particular interest in the
development of Fe3+-based 1H-MRI contrast agents because
Fig. 1 Structures of PIH, SBH, SIH and SNH.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 2132–2142 | 2133

View Article Online
Chemical Science
Edge Article
Fig. 2 Proposed mechanism of 19F-MRS/1H-MRI probe for detection of b-gal activity.
˚
these ligands have the potential to maintain coordinative and T2-weighted images with the distance r ¼ 6.53 A between
saturation while providing the opportunity for water to form 19F and Fe3+
hydrogen bonds with the catecholate oxygen atoms, which is
fundamentally related to the effectiveness of the Fe3+-based 1H-
.
b-D-Galactopyranosides synthesis
MRI contrast agents.46–49
For proof of principle, we started with the simple molecule 3-
uorocatechol (3-FCAT). 19F-MRS of 3-FCAT in PBS gave a sharp
19F-signal at dF ¼ –60.80 ppm, which showed neither changes of
19F chemical shi nor 19F-signal intensity upon the addition of
FAC during the formation of the corresponding 3 : 1 Fe-complex
[Fe(3-FCAT-2H)3]3ꢀ (3-FCAT/Fe3+) but a color change of the
solution from clear to dark purple then black, verifying the
complex formation. This result was validated by the same 19F-
MRS data from direct acquisition of the prepared 3-FCAT/Fe3+.
Very interestingly, 1H-MRI experiments of 3-FCAT/Fe3+ clearly
showed signicant differences in both T1- and T2-weighted
images (Fig. 6), which suggested the potential for combining T1
and T2 responses as concerted effects for the addition of
certainty to the imaging assessment, where tissue heterogeneity
may otherwise be misinterpreted. These results are consistent
with our previous observations with catecholate-coumarins as
Escherichia coli (lacZ) b-gal catalyzes the hydrolysis of b-D-gal-
actopyranosides by cleavage of the C–O bond between D-galac-
tose and the aglycones with overall retention of b-anomeric
conguration. Aer the demonstration of simultaneous 19F-
MRS/1H-MRI dual-function, b-D-galactopyranosyl group was
introduced to 3- and 4-FCAT for generating their b-D-gal-
actopyranosides in order to screen b-gal 19F-MRS/1H-MRI dual-
function reporters.
The phenol hydroxyl pKa values of 3- and 4-FCAT corre-
sponding to their positions were calculated using the Advanced
Chemistry Development Soware (http://www.acdlabs.com)
(Table 1). The noticeably lower pKa of the hydroxyl group in the
meta-position to the 19F-atom suggested that phase-transfer-
catalysis at pH ¼ 8–9 could provide regioselective synthesis of
mono-b-D-galactopyranosides, as we exploited previously for b-
19
gal F-MRS/MRI reporters.19,20,25,48 Reaction of 2,3,4,6-tetra-O-
1
Fe3+-based H-MRI contrast agents.16,48 In this situation the
acetyl-a-D-galactopyranosyl bromide with an equimolar amount
of 3- or 4-FCAT at room temperature catalyzed by tetrabutyl-
ammonium bromide (TBAB) in a dichloromethane-aqueous
biphasic system (pH 8–9) under N2 atmosphere afforded 1-O-
(20,30,40,60-tetra-O-acetyl-b-D-galactopyranosyl)-3-uorocatechol
(MG-3-FCAT) (95%) and 2-O-(20,30,40,60-tetra-O-acetyl-b-D-gal-
actopyranosyl)-4-uorocatechol (MG-4-FCAT) (88%), respec-
tively (Fig. 7). NOESY spectra displaying the correlations
between anomeric H-10 and H-6 in MG-3-FCAT or H-3 in MG-4-
FCAT showed the mono-b-D-galactopyranosylations occurred at
calculated distance from 19F to paramagnetic Fe3+ is only r ¼
4.82 A, which indicates that 19F-atoms in bidentate 3-FCAT/Fe3+
˚
structure are not distributed within the region of Fe3+ aniso-
tropic PRE interaction. These results were further conrmed by
using the 3-FCAT isomer, 4-uorocatechol (4-FCAT), with the
resulting consistency through the identical tests, as shown in
Fig. S2,† in which 4-FCAT and its Fe-complex [Fe(4-FCAT-
2H)3]3ꢀ (4-FCAT/Fe3+) both appeared as sharp peaks at dF
–47.10 ppm in PBS, accompanying substantial differences of T1-
¼
Fig. 3 The comparison of 1H-MRI effect of the Fe-complexes from ligands PIH, SBH, SIH and SNH with ferric ammonium citrate (FAC). (A) Control, FAC (8 mM) in 1 : 1
(v/v0) DMSO/PBS (0.1 M, pH ¼ 7.4); (B) complex, PIH, SBH, SIH or SNH each (16 mM), FAC (8 mM) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼ 7.4); conditions: T1-weighted 1H-
MRI, 200 MHz, TR ¼ 300 ms, TE ¼ 20 ms, 1.5 mm slice, 128 ꢂ 128, 50 mm ꢂ 50 mm.
2134 | Chem. Sci., 2013, 4, 2132–2142
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
Fig. 4 The 19F-MRS/1H-MRI and molecular characterization of p-FSBH to p-FSBH/Fe3+
.
19F-MRS: (1) p-FSBH (1.3 mg, 5.0 mmol) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼
7.4, 600 mL) at 25 ꢁC; (2) p-FSBH (1.3 mg, 5.0 mmol) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼ 7.4, 500 mL), FAC (663 mg, 2.5 mmol) in PBS (0.1 M, pH ¼ 7.4, 100 mL) at 25 ꢁC,
19F-NMR parameter sets: relaxation delay: 1 s, pulse width (90ꢁ): 16.6 ms, number of points acquired: 59 966, filter bandwidth: 28 000 Hz, each spectrum acquisition
time: 0.6 s, number of acquisitions: 128, and enhanced with an exponential line broadening of 40 Hz. 1H-MRI: (A) control, p-FSBH (10.0 mM) in 1 : 1 (v/v0) DMSO/PBS
(0.1 M, pH ¼ 7.4) at 20–22 ꢁC; (B) complex, p-FSBH (10.0 mM), FAC (5.0 mM) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼ 7.4) at 20–22 ꢁC; conditions: T1-weighted 1H-MRI:
using the same parameters as in Fig. 3; molecular modelling calculation: the minimized energy conformation of p-FSBH/Fe3+ was obtained by the molecular calculation
program Chem3D 8.0, using MM2 as the force-field, and the distance of F- to paramagnetic Fe3+ r ¼ 7.01 A was calculated by using Visual Molecular Dynamics (VMD
˚
1.9), white balls: H atoms, grey balls: C atoms, red balls: O atoms, blue balls: N atoms, yellow balls: F atoms, dark red ball: Fe atom.
the hydroxyl group in meta-position to the 19F-atom, as pre- furnished the fully galactopyranosylated uorocatechols: 1,2-di-
dicted. Subsequent deacetylation with NH3/MeOH from 0 ꢁC to O-(20,30,40,60-tetra-O-acetyl-b-D-galactopyranosyl)-3-uorocatechol
room temperature gave the free mono-galactopyranosides (FG-3-FCAT) (93%) and 1,2-di-O-(20,30,40,60-tetra-O-acetyl-b-D-
MGD-3-FCAT and MGD-4-FCAT in nearly quantitative yields. galactopyranosyl)-4-uorocatechol (FG-4-FCAT) (86%), respec-
The anomeric b-D-conguration of MGD-3-FCAT and MGD-4- tively (Fig. 7). Deacetylation of FG-3-FCAT and FG-4-FCAT in
4
1
ꢁ
FCAT in the C1 chair conformation was conrmed by the H- NH3/MeOH from 0 C to room temperature accomplished the
0
NMR chemical shis dH-1 ¼ 4.62–4.67 ppm of the anomeric free di-b-D-galactopyranosides FGD-3-FCAT and FGD-4-FCAT in
0
0
0
0
protons, and the coupling constants of J1 ,2 ¼ 8 Hz and J2 ,3
¼
high yields. The ESI-MS of FGD-3-FCAT and FGD-4-FCAT dis-
0
10 Hz. The anomeric carbon resonances appeared at dC-1 played the expected molecular ions, corresponding to the fully
102.68–105.75 ppm are in accordance with the b-D- galactopyranosylated derivatives. The structures of FGD-3-FCAT
conguration.19,20
and FGD-4-FCAT were conrmed by their 1H and 13C-NMR
Molecular MRI of intracellular targets and processes must spectra. The anomeric protons H-10, H-100 of D-galactoses linked
rely on the effective uptake of contrast agents by cells in vivo, to 1,2-positions of uorocatechols 3- and 4-FCAT at 4.53–4.83
0
0
00 00
0
0
requiring sufficient solubility and the capability to enter cells. ppm with the well resolved doublets (J1 ,2 ¼ J1 ,2 ¼ 8 Hz, J2 ,3
¼
00 00
As demonstrated in clinical applications, carbohydrate-associ- J2 ,3 ¼ 10 Hz) conrmed both D-galactoses in the b-congura-
ated prodrugs showed improved water solubility and perme- tion. As expected, di-b-D-galactopyranosides FGD-3-FCAT and
ability, leading to better selectivity and efficacy of FGD-4-FCAT are soluble in PBS (0.1 M, pH ¼ 7.4) in high
chemotherapy.50 Accordingly, we sought to conjugate an addi- concentrations (FGD-3-FCAT: 402 mg mLꢀ1 and FGD-4-FCAT:
tional b-D-galactopyranosyl unit to MGD-3-FCAT and MGD-4- 337 mg mLꢀ1 in PBS at room temperature), unlike MGD-3-FCAT
FCAT. Condensation of 3- or 4-FCAT directly with 2.2 equiva- and MGD-4-FCAT, which required the addition of DMSO for
lents of 2,3,4,6-tetra-O-acetyl-a-D-galactopyranosyl bromide in such high concentrations, as previously successfully applied to
anhydrous CH2Cl2/MeCN catalyzed by Hg(CN)2 as a promoter, b-gal F-NMR and 1H-MRI reporters.20,48
19
Fig. 5 The 19F-MRS/1H-MRI and molecular characterization of SBHF-p to SBHF-p/Fe3+
.
19F-MRS: (1) SBHF-p (1.2 mg, 4.5 mmol) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼
7.4, 600 mL) at 25 ꢁC; (2) SBHF-p (1.2 mg, 4.5 mmol) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼ 7.4, 500 mL), FAC (596 mg, 2.3 mmol) in PBS (0.1 M, pH ¼ 7.4, 100 mL) at 25 ꢁC,
19F-NMR: using the same parameters as in Fig. 4. 1H-MRI: (A) control, SBHF-p (9.0 mM) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼ 7.4) at 20–22 ꢁC; (B) complex, SBHF-p (9.0
mM), FAC (4.5 mM) in 1 : 1 (v/v0) DMSO/PBS (0.1 M, pH ¼ 7.4) at 20–22 ꢁC; conditions: T1-weighted 1H-MRI: using the same parameters as in Fig. 3; molecular modelling
˚
calculation: using the same programs as in Fig. 4, r ¼ 8.10 A.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 2132–2142 | 2135

View Article Online
Chemical Science
Edge Article
Fig. 6 The 19F-MRS/1H-MRI and molecular characterization of 3-FCAT to 3-FCAT/Fe3+
.
19F-MRS: (1) 3-FCAT (1.2 mg, 9.0 mmol) in PBS (0.1 M, pH ¼ 7.4, 600 mL) at
25 ꢁC; (2) 3-FCAT (1.2 mg, 9.0 mmol) in PBS (0.1 M, pH ¼ 7.4, 500 mL), FAC (796 mg, 3.0 mmol) in PBS (0.1 M, pH ¼ 7.4, 100 mL) at 25 ꢁC, 19F-NMR: using the same
parameters as in Fig. 4. 1H-MRI: (A) control, 3-FCAT (12.0 mM) in PBS (0.1 M, pH ¼ 7.4) at 20–22 ꢁC; (B) complex, 3-FCAT (12.0 mM), FAC (4.0 mM) in PBS (0.1 M, pH ¼
7.4) at 20–22 ꢁC; conditions: T1-weighted 1H-MRI: using the same parameters as in Fig. 3; T2-weighted 1H-MRI: 200 MHz, TR ¼ 2000 ms, TE ¼ 80 ms, 1.5 mm slice, 128 ꢂ
˚
128, 50 mm ꢂ 50 mm; molecular modelling calculation: using the same programs as in Fig. 4, r ¼ 4.82 A.
Table 1 The hydroxyl pKa values of 3- and 4-fluorocatechols
Table 2 19F-MRS chemical shifts (ppm) and hydrolytic rates (mM minꢀ1 per unit)
of MGD-3-FCAT, MGD-4-FCAT and FGD-3-FCAT, FGD-4-FCAT with b-gal (5 units,
E801A) in PBS (0.1 M, pH ¼ 7.4, 600 mL) at 37 ꢁC
Fluorocatechols
pKa(OH-1)
pKa(OH-2)
3-Fluorocatechol (3-FCAT)
4-Fluorocatechol (4-FCAT)
8.54 ꢃ 0.10
11.54 ꢃ 0.10
8.80 ꢃ 0.10
No.
MGD-3-FCAT MGD-4-FCAT FGD-3-FCAT FGD-4-FCAT
12.55 ꢃ 0.18
dF(substrate) ꢀ54.55
dF(aglycone) ꢀ60.80
ꢀ46.29
ꢀ47.10
0.81
ꢀ52.85
ꢀ60.80
7.95
ꢀ43.13
ꢀ47.10
3.97
Dd
n
6.25
b-Gal hydrolysis
328.57
283.93
10.51
38.33
Kinetic 19F-MRS were applied to analyze the reactions of b-gal
with mono-galactopyranosides MGD-3-FCAT, MGD-4-FCAT and
di-b-D-galactopyranosides FGD-3-FCAT, FGD-4-FCAT.18–25 MGD-
3-FCAT, MGD-4-FCAT and FGD-3-FCAT, FGD-4-FCAT in PBS (0.1
M, pH ¼ 7.4) each gave a narrow 19F-signal at dF ¼ ꢀ54.55,
ꢀ46.29, ꢀ52.85 and ꢀ43.13 ppm, respectively. Following the
reaction with b-gal (E801A) in PBS (0.1 M, pH ¼ 7.4) at 37 ꢁC, the
kinetic 19F-MRS measurement of 19F chemical shi and signal
intensity changes at each time point (0–60 min) indicated that
MGD-3-FCAT, FGD-3-FCAT converted to 3-FCAT (dF ¼ ꢀ60.80
ppm), and MGD-4-FCAT, FGD-4-FCAT to 4-FCAT (dF ¼ ꢀ47.10
ppm) with various hydrolytic rates (Table 2 and Fig. 8). Fig. 9
showed the 19F-MRS time course spectra of MGD-3-FCAT with b-
gal (E801A) in PBS (0.1 M, pH ¼ 7.4) at 37 ꢁC. The hydrolysis of
MGD-3-FCAT, FGD-3-FCAT and MGD-4-FCAT, FGD-4-FCAT
proceeded smoothly, indicating that these four substrates and
the liberated 3-FCAT and 4-FCAT had no inhibitory effects on b-
Fig. 8 The kinetic hydrolysis time courses of mono-b-D-galactopyranosides
MGD-3-FCAT / 3-FCAT (,), MGD-4-FCAT / 4-FCAT (D), and di-b-D-gal-
actopyranosides FGD-3-FCAT / 3-FCAT (C), FGD-4-FCAT / 4-FCAT (A) (23.0
mM each) with b-gal (5 units, E801A) in PBS (0.1 M, pH ¼ 7.4, 600 mL) at 37 ꢁC.
Fig. 7 The structures of mono-b-D-galactopyranosides MG-3-FCAT, MG-4-FCAT, MGD-3-FCAT, MGD-4-FCAT, and di-b-D-galactopyranosides FG-3-FCAT, FG-4-FCAT,
FGD-3-FCAT, FGD-4-FCAT.
2136 | Chem. Sci., 2013, 4, 2132–2142
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
gal (E801A). Given that MGD-3-FCAT and MGD-4-FCAT both
were much more reactive substrates to b-gal, MGD-3-FCAT also
showed larger 19F chemical shi change (DdF ¼ 6.25 ppm),
making it favored for the further evaluation. To verify the
specicity, MGD-3-FCAT was incubated with other similar
enzymes a-galactosidase (Sigma G7163) and b-glucuronidase
(Sigma G8295) in PBS (0.1 M, pH ¼ 7.4) at 37 ꢁC, however, MGD-
3-FCAT showed the resistance to both, and maintained its 19F-
signal at dF ¼ ꢀ54.55 ppm over the period of 60 min.
Fig. 9 19F-MRS spectra of time course. Mono-b-D-galactopyranoside MGD-3-
FCAT (4.0 mg, 13.8 mmol), b-gal (5 units, E801A) in PBS (0.1 M, pH ¼ 7.4, 600 mL) at
37 ꢁC, 19F-MRS: using the same parameters as in Fig. 4, number of acquisitions: 64,
pre-acquisition delay: 0 s, red peaks: MGD-3-FCAT at dF ¼ ꢀ54.55 ppm, blue
peaks: 3-FCAT at dF ¼ ꢀ60.80 ppm.
19F-MRS/1H-MRI dual detection of b-gal activity in solution
Having a b-gal 19F-MRS/1H-MRI substrate in hand, we moved
on to the detection of b-gal activity using 19F-MRS/1H-MRI
dual-function. 19F-MRS of the mixture of MGD-3-FCAT and
Fig. 10 (A) 19F-MRS detection of b-gal activity. MGD-3-FCAT (4.0 mg, 13.8 mmol), FAC (1.2 mg, 4.5 mmol), without and with b-gal (5 units, E801A) in PBS (0.1 M, pH ¼
7.4, 600 mL) at 37 ꢁC, 19F-MRS: using the same parameters as in Fig. 4, number of acquisitions: 64, MGD-3-FCAT at dF ¼ ꢀ54.55 ppm, 3-FCAT/Fe3+ at dF ¼ ꢀ60.80 ppm.
(B) 1H-MRI detection of b-gal activity. Conditions: 1H-MRI, 200 MHz, 1 mm slice, 128 ꢂ 256, 32 mm ꢂ 64 mm. (C) Control, MGD-3-FCAT (3.0 mg, 10.3 mmol), FAC (10.0
mmol), without b-gal (E801A), TFA (2.0 mmol), PBS (1.5 mL); (1) MGD-3-FCAT (1.0 mg, 3.5 mmol), FAC (10.0 mmol), b-gal (E801A, 5 units), TFA (2.0 mmol), PBS (1.5 mL); (2)
MGD-3-FCAT (2.0 mg, 6.9 mmol), FAC (10.0 mmol), b-gal (E801A, 5 units), TFA (2.0 mmol), PBS (1.5 mL); (3) MGD-3-FCAT (3.0 mg, 10.3 mmol), FAC (10.0 mmol), b-gal
(E801A, 5 units), TFA (2.0 mmol), PBS (1.5 mL); (4) MGD-3-FCAT (4.0 mg, 13.8 mmol), FAC (10.0 mmol), b-gal (E801A, 5 units), TFA (2.0 mmol), PBS (1.5 mL); (5) MGD-3-
FCAT (5.0 mg, 17.2 mmol), FAC (10.0 mmol), b-gal (E801A, 5 units), TFA (2.0 mmol), PBS (1.5 mL) at 37 ꢁC in 30 min. For FAC (10.0 mmol) alone in PBS (0.1 M, pH ¼ 7.4, 1.5
mL), R1 ¼ 3.0 sꢀ1, R2 ¼ 11.1 sꢀ1
.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 2132–2142 | 2137

View Article Online
Chemical Science
Edge Article
Fig. 11 (A) In vitro 19F-MRS detection of lacZ gene expression in MCF7-lacZ cells. (1) Control: MGD-3-FCAT (2.6 mg, 9.0 mmol), FAC (797.4 mg, 3.0 mmol), with MCF7-WT
cells (5 ꢂ 106) in PBS (0.1 M, pH ¼ 7.4, 600 mL) at 37 ꢁC; (2) MGD-3-FCAT (2.6 mg, 9.0 mmol), FAC (797.4 mg, 3.0 mmol), with MCF7-lacZ cells (5 ꢂ 106) in PBS (0.1 M, pH ¼
7.4, 600 mL) at 37 ꢁC at different time points, t ¼ 240 and 300 minutes, 19F-MRS: using the same parameters as in Fig. 9, MGD-3-FCAT at dF ¼ ꢀ54.55 ppm, 3-FCAT/Fe3+
at dF ¼ ꢀ60.80 ppm. (B) b-gal activity validation in MCF7-lacZ cells with deep blue (bottom) and no activity in MCF7-WT cells (top), X-gal staining (200ꢂ). (C) In vitro 1H-
MRI detection of lacZ gene expression in MCF7-lacZ cells after 25 hour incubation at 37 ꢁC. Conditions: 1H-MRI, 200 MHz, 1 mm slice, 128 ꢂ 128, 25.6 mm ꢂ 25.6 mm.
(A) Control, MGD-3-FCAT (3.0 mg, 10.3 mmol), FAC (5.0 mmol), TFA (2.0 mmol), MCF7-WT cells (5 ꢂ 106) in PBS (0.1 M, pH ¼ 7.4, 850 mL); (B) MGD-3-FCAT (3.0 mg, 10.3
mmol), FAC (5.0 mmol), TFA (2.0 mmol), MCF7-lacZ cells (5 ꢂ 106) in PBS (0.1 M, pH ¼ 7.4, 850 mL); (C) MGD-3-FCAT (5.8 mg, 20.0 mmol), FAC (5.0 mmol), TFA (2.0 mmol),
MCF7-lacZ cells (5 ꢂ 106) in PBS (0.1 M, pH ¼ 7.4, 850 mL); (D) MGD-3-FCAT (3.0 mg, 10.3 mmol), without FAC, TFA (2.0 mmol), MCF7-lacZ cells (5 ꢂ 106) in PBS (0.1 M, pH
¼ 7.4, 850 mL).
FAC in PBS (0.1 M, pH ¼ 7.4) displayed a sharp 19F-signal at dF of MGD-3-FCAT (3.5–17.2 mmol) and a xed concentration of
¼ ꢀ54.55 ppm. Upon the addition of b-gal (E801A) as the FAC (10.0 mmol), without or with b-gal (E801A, 5 units).
reaction proceeded at 37 ꢁC, the appearance of a new single Addition of b-gal to the mixture solution of MGD-3-FCAT and
19F-peak of 3-FCAT/Fe3+ was observed in a time-dependent FAC at 37 ꢁC for 30 min generated greater relaxation rate
manner at dF ¼ ꢀ60.80 ppm (Fig. 10A). The formation of the 3- differences DR1 and DR2 (Fig. 10B), which are increasingly
FCAT/Fe3+ complex during the enzymatic reaction of MGD-3- associated with the accumulation of Fe-complex 3-FCAT/Fe3+
FCAT in the presence of FAC was conrmed by LC/MS(ESI+) formed in situ (Fig. S3†). By the comparison with the relaxation
data, in which the substrate MGD-3-FCAT and complex 3- rates R1 ¼ 3.0 and R2 ¼ 11.1 sꢀ1 of FAC solution, we found that
FCAT/Fe3+ were detected in HPLC chromatograms at 4.4 and the interaction of b-D-galactopyranoside MGD-3-FCAT and Fe3+
16.1 min, corresponding to m/z 313.9 [M + Na]+ of MGD-3- (R1 ¼ 3.2 and R2 ¼ 14.0 sꢀ1, Fig. 10B) also produced small
FCAT (calculated m/z 313.2), and m/z 437.8 [M]+ of [Fe(3-FCAT- relaxivity, which suggested that the weaker complexity of
H)3] (calculated m/z 437.1), respectively. The relaxation times carbohydrate with trivalent Fe3+ (ref. 51) results in the increase
T1 and T2 were measured for vials containing various amounts of nonlinear DR.
2138 | Chem. Sci., 2013, 4, 2132–2142
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
19F-MRS/1H-MRI simultaneous assessments of lacZ gene
expression in transfected cancer cells
shows the location and magnitude of b-gal. Both signals are
simultaneously generated and enhanced in the “turn-on” mode
from the same site and identity, resulting in better assessing
precision and reliability.
In order to demonstrate the potential for in vivo application,
19F-MRS/1H-MRI dual functional detections of lacZ gene
expression in human MCF7 breast and PC3 prostate cancer
cells were conrmed. The cytotoxicity of MGD-3-FCAT was
examined for both the conjugate and the cleavage product
using both wild-type and lacZ expressing MCF7 and PC3 cells
(Fig. S4†). Because of releasing acute toxic 3-FCAT (Category 4,
GHS Classication) in MCF7- and PC3-lacZ cells, Fig. S4†
showed that MGD-3-FCAT is more toxic in these b-gal
expressing cells. Aer incubation of MGD-3-FCAT with wild
type (WT) and lacZ transfected human MCF7 breast cancer
Along with this bimodal reporter design, we have success-
fully synthesized and characterized a series of 19F-MRS/1H-MRI
probes, and demonstrated the feasibility of MGD-3-FCAT
assessing b-gal activity in solution and in vitro with lacZ trans-
fected tumor cells in the presence of Fe3+ ions. 19F-MRS/1H-MRI
clearly revealed WT versus lacZ-transfected cells in culture upon
incubation with MGD-3-FCAT based on the signicant differ-
ences in 19F-chemical shi change and both T1 and T2, in which
the concerted responses for adding certainty show most
promising for validation of b-gal detection. We demonstrate the
rst example using anisotropic Fe3+–19F PRE effect to assay b-gal
activity by real-time 19F-MRS/1H-MRI.
ꢁ
cells in the presence of FAC in PBS (0.1 M, pH ¼ 7.4) at 37 C
under 5% CO2 in air with 95% humidity, the mixtures were
subjected to 19F-MRS analysis at different time points. At the
starting point (0 min), MGD-3-FCAT in MCF7-WT/FAC or
MCF7-lacZ/FAC both exhibited a single narrow 19F-signal at dF
¼ –54.55 ppm. As the time of incubation went on, the 19F-peak
of MGD-3-FCAT in MCF7-WT/FAC remained unchanged. But in
MCF7-lacZ/FAC cell solution, a new single 19F-peak of 3-FCAT/
Fe3+ at dF ¼ –60.80 ppm was seen and progressively grown in a
smooth monotonic manner with the rate of n ¼ 5.48 mM minꢀ1
per million cells (Fig. 11A), whereas the rate was 1.90 times
higher when incubated with equal numbers of lysed MCF7-lacZ
cells. The b-gal activity in MCF7-lacZ cells was validated by X-
gal staining, which turned to deep blue, whereas MCF7-WT
cells showed no blue color (Fig. 11B). Signicant differences in
relaxation times T1 and T2 were observed between the lacZ
transfected and WT MCF7 breast cancer cells aer incubation
for 25 hours. In MCF7-WT cells relaxation rates were deter-
mined to be T1 ¼ 249 ꢃ 22 ms and T2 ¼ 58 ꢃ 8 ms, while in
MCF7-lacZ cells T1 ¼ 94 ꢃ 12 ms and T2 ¼ 36 ꢃ 3 ms were
observed [compared with the same concentration of MGD-3-
FCAT (12.12 mM), the T1 and T2 values with various concen-
trations are represented as bars at the right sides adjacent to T1
and T2 maps] (Fig. 11C).
In the present studies, we introduced exogenous Fe3+, ulti-
mately, we hope to develop this approach to scavenge the
1
elevated Fe3+ of tumors for H-MRI signal generation. Indeed,
aroyl hydrazones and catechols are in clinical use as Fe3+ scav-
enging agents suggesting a potential for clinical theranostic
application.
Experimental
General methods
NMR spectra were recorded on a Varian Unity INOVA 400
1
spectrometer (400 MHz for H, 100 MHz for 13C, 376 MHz for
1
19F), H and 13C chemical shis are referenced to TMS as the
internal standard with CDCl3, or DMSO-d6 as solvents. 19F uses
a dilute solution of NaTFA in a capillary as the external stan-
dard, chemical shis are given in ppm. All compounds were
ꢁ
characterized by NMR at 25 C. Microanalyses were performed
on a Perkin-Elmer 2400CHN microanalyser. Mass spectra were
obtained by positive and negative ESI-MS using a Micromass Q-
TOF hybrid quadrupole/time-of-ight instrument (Micromass
UK Ltd). LC/MS data were obtained using a HP Agilent 1100
Series LC/MSD on an Eclipse XDB-C18 column with 70%/30%
CH3CN/H2O as eluents.
In similarly incubated intact PC3-lacZ cells, the rate of
conversion from MGD-3-FCAT (dF ¼ –54.55 ppm) to 3-FCAT (dF
¼ –60.80 ppm) is n ¼ 862.86 mM minꢀ1 per million cells, and
1.86 times higher with the same numbers of lysed PC3-lacZ
cells. These differences between intact and lysed cells indicate
that membrane permeation is a big obstruction to b-gal reaction
in MCF7- and PC3-lacZ cells. The relaxation times were T1 ¼ 466
ꢃ 62 ms and T2 ¼ 143 ꢃ 4 ms in PC3-WT cells, whereas for PC3-
lacZ cells T1 ¼ 204 ꢃ 60 ms and T2 ¼ 101 ꢃ 3 ms aer 25 hour
incubation were obtained [with MGD-3-FCAT (15.13 mM)],
respectively (Fig. S5†).
Solutions in organic solvents were dried with anhydrous
sodium sulfate, and concentrated in vacuo below 45 ꢁC. 2,3,4,6-
Tetra-O-acetyl-a-D-galactopyranosyl bromide was purchased
from the Sigma Chemical Company. b-Gal (E801A) was
purchased from Promega (Madison, WI, USA) and enzymatic
reactions were performed at 37 ꢁC in PBS solution (0.1 M, pH ¼
7.4). Column chromatography was performed on silica gel (200–
300 mesh) and silica gel GF254 used for analytical TLC was
purchased from the Aldrich Chemical Company. Detection was
effected by spraying the plates with 5% ethanolic H2SO4 (fol-
ꢁ
lowed by heating at 110 C for 10 min) or by direct UV illumi-
nation of the plate. The purity of the nal products was
determined by HPLC with $95%.
Conclusion
In summary, we present here a novel bimodal approach of
synergizing 19F-MRS/1H-MRI for the detection of b-gal activity,
in which 19F-MRS is used to dene b-gal responsive 19F-chem-
Mono-b-D-galactopyranosides MG-3-FCAT and MG-4-FCAT
ical shi changes and hydrolytic kinetics for identication of General procedure – a solution of 2,3,4,6-tetra-O-acetyl-a-D-
the presence and progression of b-gal. Concurrently, 1H-MRI galactopyranosyl bromide (1.62 g, 3.93 mmol) in CH2Cl2
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 2132–2142 | 2139

View Article Online
Chemical Science
Edge Article
(50 mL) was added dropwise to a vigorously stirred biphasic included for routine culture. The cells were maintained in
mixture (pH 8–9) of 3- or 4-uorocatechol (0.50 g, 3.93 mmol) Dulbecco's modied Eagle's medium (DMEM, Mediatech Inc.,
and tetrabutylammonium bromide (TBAB) (160 mg, 0.5 mmol) Herndon, VA, USA) containing 10% fetal bovine serum (FBS,
in CH2Cl2–H2O (80 mL, 1 : 1 v/v0) over a period of 1 h at room Atlanta Biologicals, Inc., Lawrenceville, GA, USA) with 100
temperature under an N2 atmosphere, and the stirring units mLꢀ1 penicillin, 100 units mLꢀ1 streptomycin, and
ꢁ
continued for additional ꢄ3 h until TLC showed that the cultured in a humidied 5% CO2 incubator at 37 C. The b-gal
reaction was completed. The products were extracted with activity of lacZ-transfected tumor cells was measured using a b-
CH2Cl2 (4 ꢂ 30 mL), washed (H2O), dried (Na2SO4), and gal assay kit with o-nitrophenyl-b-D-galactopyranoside (Prom-
evaporated under reduced pressure to give a syrup, which was ega, Madison, WI, USA), and conrmed by X-gal staining.21,22
puried by column chromatography on silica gel to give the Cell lysis was achieved by a freeze/thaw method: equal
acetylated mono-b-D-galactopyranosides MG-3-FCAT and MG-4- numbers of MCF7-lacZ or PC3-lacZ cells were suspended in
FCAT.
PBS and then frozen at –80 ꢁC for 10 min before thawing at
room temperature over 3 cycles.
Di-b-D-galactopyranosides FG-3-FCAT and FG-4-FCAT
General procedure – to a vigorously stirred solution of 3- and 4- Cytotoxicity
uorocatechols (0.32 g, 2.50 mmol) and Hg(CN)2 (2.10 g, 8.25
The cytotoxicity of MGD-3-FCAT was assessed for both conju-
mmol) in anhydrous MeCN (80 mL) containing freshly activated
gate and the cleavage product in both wild-type and lacZ
expressing MCF7 and PC3 cells using a colorimetric CellTiter 96
Aqueous Nonradioactive MTS Cell Proliferation Assay (Prom-
ega). Assays were performed in triplicate using 24-well plates
seeded with 103 cells per well in 500 mL of RPMI-1640 without
phenol red and supplemented with 10% FCS and 2 mM
glutamine.19
˚
4 A molecular sieves (5.0 g) was added dropwise 2,3,4,6-tetra-O-
acetyl-a-D-galactopyranosyl bromide (2.26 g, 5.50 mmol, 2.2
equiv.) in CH2Cl2 (50 mL). The mixture was stirred continually
in the dark at room temperature under N2 atmosphere until
TLC indicated that the reaction was completed, then diluted
with CH2Cl2 (100 mL), ltered through Celite, washed, dried
(Na2SO4), and evaporated under reduced pressure to give a
syrup, which was puried by column chromatography on silica
gel to give the acetylated di-b-D-galactopyranosides FG-3-FCAT
and FG-4-FCAT.
Kinetic 19F-MRS experiments
Relative substrate efficacy of mono-galactopyranosides MGD-3-
FCAT, MGD-4-FCAT and di-galactopyranosides FGD-3-FCAT,
FGD-4-FCAT was evaluated using 19F-MRS. Enzyme reactions
Free mono- and di-b-D-galactopyranosides MGD-3-FCAT,
MGD-4-FCAT, FGD-3-FCAT and FGD-4-FCAT
ꢁ
were conducted at 37 C in PBS (0.1 M, pH ¼ 7.4) using b-gal
(E801A, Promega, Madison, WI, USA). MGD-3-FCAT, MGD-4-
FCAT and FGD-3-FCAT, FGD-4-FCAT (13.8 mmol) were dissolved
General procedure – a solution of acetylated mono-b-D-gal-
actopyranosides MG-3-FCAT or MG-4-FCAT (950 mg), or di-O-
(20,30,40,60-tetra-O-acetyl-b-D-galactopyranosyl) uorocatechols
FG-3-FCAT or FG-4-FCAT (1.50 g) in anhydrous MeOH (100
in PBS (595 mL, pH ¼ 7.4) and b-gal (E801A, 5 mL, 1 unit mLꢀ1
)
added, followed by immediate 19F-MRS data acquisition at 37 ꢁC
with subsequent spectra every 102 s providing a kinetic curve
over 60 min.
ꢁ
mL) containing 0.5 M NH3 was vigorously stirred from 0 C to
room temperature overnight until TLC showed that the reac-
tion was completed, and evaporated to dryness in vacuo.
Chromatography of the crude syrup on silica gel with ethyl MRI
acetate-methanol (1 : 2, 1 : 4) afforded the corresponding free
MRI studies were performed using a 4.7 T horizontal bore
mono-b-D-galactopyranosides MGD-3-FCAT, MGD-4-FCAT and
di-b-D-galactopyranosides FGD-3-FCAT, FGD-4-FCAT in high
yields.
magnet equipped with a Varian INOVA Unity system (Palo Alto,
CA, USA), T1 and T2 maps were acquired using a spin-echo
sequence with varying repetition times (TR) and echo times
(TE), respectively. An inversion recovery (IR) turbo fast low-
angle shot (FLASH) with magnitude reconstruction pulse
Stable lacZ transfected cancer cells
E. coli lacZ gene (from pSV-b-gal vector, Promega, Madison, WI, sequence was performed for quantitative T1 measurement. The
USA) was inserted into high expression human cytomegalo- raw data were acquired using a centric k-space reordering
virus (CMV) immediate-early enhancer/promoter vector scheme, followed by the phase encoding steps with higher
phCMV (Gene Therapy Systems, San Diego, CA, USA) giving a phase encoding gradient amplitudes. Data acquisition
recombinant vector phCMV/lacZ. This was used to transfect parameters of the FLASH readout were TR/TE/Flip angle ¼ 10
wild type MCF7 (human breast cancer) and PC3 (human ms/5 ms/10ꢁ. The standard multi-echo Carr–Purcell–Meiboom–
prostate cancer) cells (ATCC, Manassas, VA, USA) using Gene- Gill pulse sequence was used for measuring T2 from a single
PORTER2 (Gene Therapy Systems, Genlantis, Inc., San Diego, echo train. The T2 maps were obtained on a voxel-by-voxel
CA, USA), as described in detail previously.21,22 The highest b- basis using nonlinear least-squares t the equation M ¼
gal expressing colony was selected using the antibiotic G418 M0eꢀTE/T from the images taken at each echo time. All maps
2
disulfate (800 mg mLꢀ1, Research Products International Corp, were processed using in-house MatLab scripts (MathWorks
Mt Prospect, IL, USA) and G418 (200 mg mLꢀ1) was also Inc., Natick, MA, USA).
2140 | Chem. Sci., 2013, 4, 2132–2142
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
13 Y. T. Chang, C. M. Cheng, Y. Z. Su, W. T. Lee, J. S. Hsu,
G. C. Liu, T. L. Cheng and Y. M. Wang, Bioconjugate Chem.,
2007, 18, 1716.
14 K. Hanaoka, K. Kikuchi, T. Terai, T. Komatsu and T. Nagano,
Chem.–Eur. J., 2008, 14, 987.
15 F. Arena, J. B. Singh, E. Gianolio, R. Stefania and S. Aime,
Bioconjugate Chem., 2011, 22, 2625.
16 W. Cui, L. Liu, V. D. Kodibagkar and R. P. Mason, Magn.
Reson. Med., 2010, 64, 66.
17 N. E. Bengtsson, G. Brown, E. W. Scott and G. W. Walter,
Magn. Reson. Med., 2010, 63, 745.
18 J.-X. Yu, V. D. Kodibagkar, W. Cui and R. P. Mason, Curr.
Med. Chem., 2005, 12, 819.
19 (a) J.-X. Yu, P. Otten, Z.-Y. Ma, W. Cui, L. Liu and
R. P. Mason, Bioconjugate Chem., 2004, 15, 1334; (b)
W. Cui, P. Otten, Y.-M. Li, K. S. Koeneman, J.-X. Yu and
R. P. Mason, Magn. Reson. Med., 2004, 51, 616; (c) J.-X. Yu,
Z.-Y. Ma, Y.-M. Li, K. S. Koeneman, L. Liu and
R. P. Mason, Med. Chem., 2005, 1, 255; (d) J.-X. Yu, L. Liu,
V. D. Kodibagkar, W. Cui and R. P. Mason, Bioorg. Med.
Chem., 2006, 14, 326.
Abbreviations
MRI
MRS
TR
Magnetic resonance imaging
Magnetic resonance spectroscopy
Repetition time
TE
Echo time
NOESY Nuclear overhauser effect spectroscopy
b-Gal
PIH
SBH
SIH
b-Galactosidase
Pyridoxal isonicotinoylhydrazone
Salicylaldehyde benzoylhydrazone
Salicylaldehyde isonicotinoylhydrazone
Salicylaldehyde nicotinoylhydrazone
SNH
p-FSBH 2-Hydroxyl-5-uoro-benzaldehyde benzoylhydrazone
SBHF-p 2-Hydroxylbenzaldehyde 4-uorobenzoylhydrazone
PRE
DMSO
FAC
PBS
TLC
Paramagnetic relaxation enhancement
Dimethyl sulfoxide
Ferric ammonium citrate
Phosphate buffered saline
Thin layer chromatography.
20 J.-X. Yu and R. P. Mason, J. Med. Chem., 2006, 49, 1991.
21 V. D. Kodibagkar, J.-X. Yu, L. Liu, H. Hetherington and
R. P. Mason, Magn. Reson. Imaging, 2006, 24, 959.
22 L. Liu, V. D. Kodibagkar, J.-X. Yu and R. P. Mason, FASEB J.,
2007, 21, 2014.
23 J.-X. Yu, V. D. Kodibagkar, L. Liu and R. P. Mason, Magn.
Reson. Med., 2008, 21, 704.
24 J.-X. Yu, W. Cui, D. Zhao and R. P. Mason, Fluorine Health,
2008, 197.
25 J.-X. Yu, V. D. Kodibagkar, R. R. Hallac, L. Liu and
R. P. Mason, Bioconjugate Chem., 2012, 23, 596.
26 S. Mizukami, H. Matsushita, R. Takikawa, F. Sugihara,
M. Shirakawa and K. Kikuchi, Chem. Sci., 2011, 2, 1151.
27 R. Weissleder and M. J. Pittet, Nature, 2008, 452, 580.
Acknowledgements
This research was supported by the DOD Prostate Cancer New
Investigator Award W81XWH-05-1-0593. NMR experiments were
performed at the Advanced Imaging Research Center. We are
grateful to Dr Ralph Mason for valuable discussion, Jennifer
McAnally and Ya Ren for expert technical assistance.
References
1 A. Kruger, V. Schirrmacher and R. Khokha, Cancer Metastasis
Rev., 1999, 17, 285.
2 U. Haberkorn, W. Mier and M. Eisenhut, Curr. Med. Chem.,
2005, 12, 779.
3 A. Razgulin, N. Ma and J. H. Rao, Chem. Soc. Rev., 2011, 40, 28 (a) Y. Takaoka, T. Sakamoto, S. Tsukiji, M. Narazaki,
4186.
T. Matsuda, H. Tochio, M. Shirakawa and I. Hamachi, Nat.
Chem., 2009, 1, 557; (b) Y. Takaoka, Y. Sun, S. Tsukiji and
I. Hamachi, Chem. Sci., 2011, 2, 511.
4 L. Li, R. J. Zemp, G. Lungu, G. Stoica and L. V. Wang, J.
Biomed. Opt., 2007, 12, 020504.
5 C. H. Tung, Q. Zeng, K. Shah, D. E. Kim, D. Schellingerhout 29 M. Ito, A. Shibata, J. Zhang, M. Hiroshima, Y. Sako,
and R. Weissleder, Cancer Res., 2004, 64, 1579.
6 M. Kamiya, H. Kobayashi, Y. Hama, Y. Koyama, M. Bernardo,
T. Nagano, P. L. Choyke and Y. Urano, J. Am. Chem. Soc.,
2007, 129, 3918.
7 V. Josserand, I. Texier-Nogues, P. Huber, M. C. Favrot and
J. L. Coll, Gene Ther., 2007, 14, 1587.
Y. Nakano, K. Kojima-Aikawa, B. Mannervik, S. Shuto,
Y. Ito, R. Morgenstern and H. Abe, ChemBioChem, 2012,
13, 1428.
30 (a) K. Mizusawa, Y. Ishida, Y. Takaoka, M. Miyagawa,
S. Tsukiji and I. Hamachi, J. Am. Chem. Soc., 2010, 132,
7291; (b) Y. Takaoka, K. Kiminami, K. Mizusawa,
K. Matsuo, M. Narazaki, T. Matsuda and I. Hamachi, J. Am.
Chem. Soc., 2011, 133, 11725.
8 T. S. Wehrman, G. von Degenfeld, P. O. Krutzik, G. P. Nolan
and H. M. Blau, Nat. Methods, 2006, 3, 295.
9 L. Liu and R. P. Mason, PLoS One, 2010, 5, e12024.
10 S. Celen, C. Deroose, T. de Groot, S. K. Chitneni, R. Gijsbers,
31 P. Harvey, K. H. Chalmers, E. De Luca, A. Mishra and
D. Parker, Chem.–Eur. J., 2012, 18, 8748.
Z. Debyser, L. Mortelmans, A. Verbruggen and G. Bormans, 32 M. Higuchi, N. Iwata, Y. Matsuba, K. Sato, K. Sasamoto and
Bioconjugate Chem., 2008, 19, 441.
T. C. Saido, Nat. Neurosci., 2005, 8, 527.
11 M. E. Van Dort, K. C. Lee, C. A. Hamilton, A. Rebemtulla and 33 K. Yamaguchi, R. Ueki, H. Nonaka, F. Sugihara, T. Matsuda
B. D. Ross, Mol. Imaging, 2008, 7, 187. and S. Sando, J. Am. Chem. Soc., 2011, 133, 14208.
12 A. Y. Louie, M. M. Huber, E. T. Ahrens, U. Rothbacher, 34 S. Mizukami, R. Takikawa, F. Sugihara, Y. Hori, H. Tochio,
R. Moats, R. E. Jacobs, S. E. Fraser and T. J. Meade, Nat.
Biotechnol., 2000, 18, 321.
M. Walchli, M. Shirakawa and K. Kikuchi, J. Am. Chem.
Soc., 2008, 130, 794.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 2132–2142 | 2141

View Article Online
Chemical Science
Edge Article
35 (a) S. Mizukami, R. Takikawa, F. Sugihara, M. Shirakawa and
K. Kikuchi, Angew. Chem., Int. Ed., 2009, 48, 3641; (b)
H. Matsushita, S. Mizukami, Y. Mori, F. Sugihara,
M. Shirakawa, Y. Yoshioka and K. Kikuchi, ChemBioChem,
2012, 13, 1579.
36 K. Tanaka, N. Kitamura and Y. Chujo, Bioconjugate Chem.,
2011, 22, 1484.
37 M. Srinivas, A. Heerschap, E. T. Ahrens, C. G. Figdor and
I. J. M. de Vries, Trends Biotechnol., 2010, 28, 363.
38 E. T. Ahrens, R. Flores, H. Y. Xu and P. A. Morel, Nat.
Biotechnol., 2005, 23, 983.
A. M. Blamire, Magn. Reson. Med., 2011, 66, 931; (c)
P. Harvey, I. Kuprov and D. Parker, Eur. J. Inorg. Chem.,
2012, 2015; (d) F. Schmid, C. Holtke, D. Parker and
C. Faber, Magn. Reson. Med., 2012, 66, 931.
45 G. H. Bao, M. Clion, T. M. Hoette, K. Mori, S. X. Deng,
A. Qiu, M. Viltard, D. Williams, N. Paragas, T. Leete,
R. Kulkarni, X. Li, B. Lee, A. Kalandadze, A. J. Ratner,
J. C. Pizarro, K. M. Schmidt-Ott, D. W. Landry,
K. N. Raymond, R. K. Strong and J. Barasch, Nat. Chem.
Biol., 2010, 6, 602.
46 J. A. Davies, S. G. Dutremez, C. M. Hockensmith, R. Keck,
N. Richardson, S. Selman, D. A. Smith, C. W. Ulmer II,
L. S. Wheatley and J. Zeiss, Acad. Radiol., 1996, 3,
936.
39 D. G. Reid and P. S. Murphy, Drug Discovery Today, 2008, 13,
473.
40 (a) D. Zu and R. C. Hider, Coord. Chem. Rev., 2002, 232, 151;
(b) J. L. Buss, B. T. Greene, J. Turner, F. M. Torti and 47 D. D. Schwert, N. Richardson, G. Ji, B. Raduchel, W. Ebert,
S. V. Torti, Curr. Top. Med. Chem., 2004, 4, 1623.
41 (a) D. B. Lovejoy and D. R. Richardson, Curr. Med. Chem.,
P. E. Heffner, R. Keck and J. A. Davies, J. Med. Chem., 2005,
48, 7482.
2003, 10, 1035; (b) D. R. Richardson, Curr. Med. Chem., 48 J.-X. Yu, P. K. Gulaka, L. Liu, V. D. Kodibagkar and
2005, 12, 2711; (c) D. S. Kalinowski and D. R. Richardson, R. P. Mason, ChemPlusChem, 2012, 77, 370.
Pharmacol. Rev., 2005, 57, 547; (d) D. R. Richardson, 49 K. L. Haas and K. J. Franz, Chem. Rev., 2009, 109, 4921.
D. S. Kalinowski, S. Lau, P. J. Jansson and D. B. Lovejoy, 50 (a) X. Chen, B. Wu and P. G. Wang, Curr. Med. Chem.: Anti-
Biochim. Biophys. Acta, Gen. Subj., 2009, 1790, 702.
42 (a) N. Richardson and J. A. Davies, Polyhedron, 1999, 18,
2457; (b) D. D. Schwert, J. A. Davies and N. Richardson,
Top. Curr. Chem., 2002, 221, 165.
Cancer Agents, 2003, 3, 139; (b) Y. Singh, M. Palombo and
P. J. Sinko, Curr. Med. Chem., 2008, 15, 1802; (c) F. Kratz,
¨
I. A. Muller, C. Ryppa and A. Warnecke, ChemMedChem,
2008, 3, 20.
43 G. M. Clore and J. Iwahara, Chem. Rev., 2009, 109, 4108.
44 (a) K. H. Chalmers, E. De Luca, N. H. M. Hogg,
A. M. Kenwright, I. Kuprov, D. Parker, M. Botta,
J. I. Wilson and A. M. Blamire, Chem.–Eur. J., 2010, 16, 134;
(b) K. H. Chalmers, A. M. Kenwright, D. Parker and
51 O. Coskuner and C. A. Gonzalez, Metallic Systems: A Quantum
Chemists Perspective, Chapter 3, Carbohydrate and Trivalent
Iron Ion Interactions, ed. T. C. Allison, O. Coskunner and C.
A. Gonzalez, CRC Press, Taylor & Francis Group, 2011, pp.
83–106, ISBN: 1420060775, 9781420060775.
2142 | Chem. Sci., 2013, 4, 2132–2142
This journal is ª The Royal Society of Chemistry 2013