An Activatable 19 F MRI Molecular Probe for Sensing and Imaging of Norepinephrine.

Norepinephrine (NE), acting as both a neurotransmitter and hormone, plays a significant role in regulating the action of the brain and body. Many studies have demonstrated a strong correlation between mental disorders and aberrant NE levels. Therefore, it is of urgent demand to develop in vivo analytical methods of NE for diagnostic assessment and mechanistic investigations of mental diseases. Herein, we report a 19 F MRI probe (NRFP) for sensing and imaging NE, which is constructed by conjugating a gadolinium chelate to a fluorine-containing moiety through a NE-responsive aromatic thiocarbonate linkage. The capacity and specificity of NRFP for detecting NE is validated with in vitro detecting/imaging experiments. Furthermore, the feasibility of NRFP for visualizing NE in animals is illustrated by ex vivo and in vivo imaging experiments, demonstrating the promising potential of NRFP for selective detection and specific imaging of NE in deep tissues of living subjects.

Introduction

Norepinephrine (NE), a crucial neurotransmitter and hormone in vertebrates, plays a fundamental role in various physiological processes to maintain the homeostasis and facilitate the tackling of various changes during metabolism.[ , , ] NE is synthesized by a series of enzymatic reactions starting from L‐tyrosine with levodopa and dopamine (DA) as the intermediates, which can be further converted to epinephrine (E) in the neurons of the locus coeruleus (LC) in the brain (Figure 1a). In biological systems, NE is critical for nervous system development, memory refreshment, heart rate regulation and blood pressure stabilization.[ , , ] It is generally accepted that there is a strong correlation between abnormal metabolic levels of NE and psychiatric conditions or neurodegenerative diseases.[ , , , ] In recent decades, with the rapid pace of modern life and the increased pressure it brings, the number of people who are suffering from mental disorders such as depression has exceeded more than 260 million all around the world. Many studies concentrating on depression have demonstrated that this disease is mainly related to the decreased levels of NE in the brain.[ , ] Though the interactions between NE and its receptors usually lasts for a very short period of time, they are capable of exerting a wide range of effects to the body for several hours or even days. Therefore, in vivo detection of NE is of great importance for sensitive diagnostic assessment and in‐depth mechanistic studies of mental disorders such as depression. Traditional methods, like electrochemical analysis and mass spectrometry, have been widely used for detection and measurement of neurotransmitters in organism. However, they are facing substantial challenges for in vivo analysis and sensing.[ , ] Recently, a variety of fluorescence probes for depression assessment based on detecting neurotransmitter in complex physiological environments have been reported.[ , , ] Unfortunately, several obstacles need to be overcome for the further applications of these approaches, which include the shallow tissue penetration of fluorescence and the low specificity towards NE due to the structural similarity of three catecholamine neurotransmitters (DA, NE, and E).

Figure 1

(a) In vivo metabolic pathways of catecholamines, including dopamine (DA), norepinephrine (NE), and epinephrine (E). MAO: monoamine oxidase; PMNT: phenylethanolamine N‐methyltransferase. (b) A schematic illustration showing the functional mechanism of our norepinephrine‐responsive 19F MRI probe (NRFP). 19F NMR/MRI signals of NRFP are “OFF” due to the strong paramagnetic relaxation enhancement (PRE) effect exerted by the chelated Gd3+ ion. The effect is significantly weakened when norepinephrine (NE) selectively cleaves the aromatic thiocarbonate linkage, which releases 3,5‐bis(trifluoromethyl)benzene‐thiol (BTFBT), leading to the switching “ON” of 19F NMR/MRI signals.

Among different molecular imaging techniques, magnetic resonance imaging (MRI) has attracted significant attention from researchers for its advantages of deep penetration, high resolution, non‐ionizing radiation, and multi‐parameter imaging.[ , , ] As a real‐time and deep‐tissue imaging method, 1H MRI is widely used in clinic, which is rendering the difference in relaxation time for protons in various soft tissues, and therefore offering high‐resolution anatomical images for diagnosis.[ , ] Nevertheless, the interferences from high background and artifacts obstruct the application of 1H MRI for detecting bioactive molecules in organisms that are of relatively low concentration.[ , ] 19F, which is of 100 % natural abundance, a wide range of chemical shifts, a gyromagnetic ratio close to 1H (94 % relative to 1H), is regarded as an ideal type of nuclei complementary to 1H for MRI. Most importantly,[ , ] the 19F content in the human body is extremely low (<10−6  m),[ , ] which allows 19F MRI to detect low concentration species with negligible background and render their distributions as “hot‐spot” images.[ , , , , , , ] Therefore, 19F MRI is gaining momentum in the field of sensing and imaging low concentration bioactive species.[ , , , , , , , , , , ]

In light of these considerations, we envision that detecting NE with 19F MRI might offer a potential means to circumvent the obstacles encountered by fluorescent approaches. We design a norepinephrine‐responsive 19F MRI probe (NRFP) by linking a Gd chelate (Gd‐DOTA) to a fluorine‐containing small molecule 3,5‐bis(trifluoromethyl)benzenethiol (BTFBT) via an aromatic thiocarbonate linkage, which is specifically responsive to NE. In NRFP, because of the paramagnetic relaxation enhancement (PRE) effect exerted by the Gd chelate, the longitudinal and transverse relaxation times (T 1 and T 2) of 19F nuclei in BTFBT are significantly shortened, resulting in the “OFF” state of 19F signals. Once NE cleaves the aromatic thiocarbonate linkage and releases BTFBT from NRFP, the PRE effect is substantially attenuated due to the increased distance between the Gd chelate and BTFBT, leading to the recovery of the relaxation times and the “ON” state of 19F signals (Figure 1b), which allows for the detection and imaging of NE. We quickly validate the working principle of the probe by measuring the T 1 and T 2 of 19F before and after NRFP’s response to NE. Additionally, we illustrate the feasibility and specificity of NRFP for the detection and imaging of NE via in vitro detecting/imaging experiments. Finally, we achieve ex vivo and in vivo imaging of NE by 19F MRI with NRFP, demonstrating the promising potential of our probe for selective detection and specific imaging of NE in the deep tissues of living subjects.

Results and Discussion

Synthesis of NRFP

NRFP was synthesized in a facile and modular approach (Scheme S1, Supporting Information). Briefly, we first synthesized Compound 3, which serves as a moiety to chelate Gd3+ in NRFP. 3,5‐Bis(trifluoro‐methyl)benzenethiol (BTFBT), which serves as a fluorine‐bearing moiety in NRFP was then coupled to 3 through a NE‐responsive aromatic thiocarbonate linkage to afford Ligand 7. The synthetic protocols and characterization data for compounds 1–7 are included in the Experimental Section and the Supporting Information. Chelation of 7 with Gd3+ ions afforded the final probe NRFP (8), which was confirmed by high‐resolution mass spectrometry (HR‐MS). The broad and short peak in the 19F nuclear magnetic resonance (NMR) spectrum of NRFP indicates the significant PRE effect within the probe.

Measurements of 19F Relaxation Times

We then measured the longitudinal and transverse relaxation times (T 1 and T 2) of 19F in NRFP on a NMR spectrometer (564 MHz for 19F). As shown in Table S1, the T 1 and T 2 of 19F in NRFP were substantially shortened (less than 10 ms) compared to those in 7 (1287.0 and 787.8 ms, respectively), confirming the pronounced PRE effect with NRFP. Upon incubation with NE, the T 1 and T 2 of 19F were considerably extended (to 621.0 and 232.2 ms, respectively), indicating the significant weakening of the PRE effect. These results confirm our design of NRFP and illustrate the feasibility of our probe for NE detection.

In vitro Sensing of NE with NRFP using 19F NMR Spectroscopy

We further explored the detection of NE with NRFP in vitro by 19F NMR. NRFP was incubated with NE and the reaction was monitored by 19F NMR spectroscopy. The intensity of the peak in 19F NMR spectra increased along with the extension of incubation time and reached a plateau after 60 min (Figure S1a), which is affirmed with signal‐to‐noise ratio (SNR) analysis (Figure S1b). This response was not affected by different pH conditions (Figure S2a). However, without NE, the intensity of the peak in 19F NMR spectra remained low even after 48 h incubation (Figure S2b) or under different pH conditions (Figure S2c). These results indicate the successful response of NRFP to NE. We then studied the specificity of our probe. NRFP were incubated with a panel of analytes and the reactions were monitored by 19F NMR. As shown in Figure 2a, NRFP did not respond to common biomolecules such as glucose and those in FBS, reducing agents such as glutathione (GSH), active sulfur substances such as l(+)‐cysteine and homocysteine, or reactive oxygen species/reactive nitrogen species (ROS/RNS) such as alkylperoxy radicals (RO⋅), hydrogen peroxide (H2O2), hypochlorous acid (HClO) and peroxynitrite (ONOO−). NRFP did respond to epinephrine and dopamine, but the responding efficiencies were insignificant compared to that of NRFP to NE. These observations are confirmed with SNR analysis (Figure 2b), indicating the excellent selectivity of NRFP for the detection of NE with a limit of detection (LOD) of 2.1 mm in 19F NMR (Figure S3). We also investigated the response of NRFP to NE by high‐performance liquid chromatography (HPLC). Incubation of NRFP with NE resulted in two new peaks in the HPLC chromatogram (Figure 2c), which were confirmed to be corresponding to the expected products Gd‐DOTA‐Tyramine (GDTy) and BTFBT (Figure 1b), respectively. These results indicate the successful and specific response of NRFP towards NE, which allows for selective detection of NE by 19F NMR.

Figure 2

(a) Representative 19F NMR spectra of NRFP (−62.7 ppm) in 500 μL 10 % D2O/H2O (final concentration: 0.4 mm), which were incubated with various analytes for 3 h at 25 °C. PBS: 50 mm, pH 7.4; FBS: 10 % in PBS; dopamine (DA), norepinephrine (NE) and epinephrine (E): 10 mm in PBS; Glu: 5.0 mm glucose in PBS; GSH: 2.0 mm glutathione in PBS; all other analytes: 0.40 mm in PBS for 3 h at 25 °C. CF3COONa (at −75.4 ppm) was used as an internal reference. (b) Signal‐to‐noise ratio (SNR) analysis on the results of the experiment as indicated in (a) (n=5). The SNR for the peak of NRFP in PBS was set as 1.0 and the other SNRs were normalized accordingly. (c) HPLC chromatograms showing NRFP (0.2 mm) alone and the products of NRFP incubated with 50 equiv. NE for 3 h. The expected products, Gd‐DOTA‐Tyramine (GDTy) and 3,5‐bis(trifluoromethyl)benzene‐thiol (BTFBT) were also analyzed. See Figure S4 for HR‐ESI‐MS analysis of the fractions corresponding to the peaks indicated by an asterisk and a triangle.

In vitro Imaging of NE using 19F MRI with NRFP

We next investigated the imaging of NE by 19F MRI with NRFP. A series of solutions containing NRFP at indicated concentrations were imaged on a 9.4 T MRI scanner with commercially available 1H/19F MRI coils before and after incubation with excess NE. The solutions showed no 19F MRI signals before incubation with NE. In contrast, after incubation, these solutions exhibited significant signals, which intensified as the increase of NRFP concentration (Figure 3a). The specificity of NRFP was also evaluated. Incubation of NRFP with common bioactive molecules (glucose, GSH, and H2O2) showed no 19F MRI signals while much stronger signals were observed for NRFP incubated with NE (Figure 3b). Incubation of NRFP with dopamine or epinephrine also showed some signals, but the intensities of these signals were relatively weak compared to those of NRFP incubated with NE. SNR analysis further confirms the significant enhancement in signal intensity for NRFP after incubation with NE, the positive correlation between signal intensity after response and NRFP concentration, and the specific response of NRFP towards NE. These results demonstrate that it is feasible to image NE with NRFP by 19F MRI.

Figure 3

(a) Representative 19F MR images of NRFP (at indicated concentrations) before and after specific activation towards NE (0.3 m). (b) Representative 19F MR images of NRFP (60 mm 19F) before and after incubated with: dopamine (DA), norepinephrine (NE) and epinephrine (E): 0.3 m in PBS; glucose (Glu), H2O2 and glutathione (GSH): 10 mm in PBS for 3 h at 25 °C. (c) SNR analysis on the results of the experiment as indicated in (a) (n=5). The SNR for the 19F MR image of 0.83 mm NRFP (i. e., 10 mm 19F) in PBS was set as 1.0 and the other SNRs were normalized accordingly. (d) SNR analysis on the results of the experiment as indicated in (b) (n=5). The SNR for the 19F MR image of NRFP with glucose in PBS was set as 1.0 and the other SNRs were normalized accordingly.

Ex vivo Imaging of NE by 19F MRI with NRFP

Before imaging experiments on animals, we quickly assessed the biocompatibility of NRFP. NRFP did not show significant cytotoxicity to L02 or HepG2 cells even at 20 mm (Figure S5a). No appreciable microscopic lesions were observed for hematoxylin and eosin (H&E) stained tissue section of all major organs collected from the mice at 7 d after intravenous injection of NRFP (Figure S5b). These results indicate the good biocompatibility of NRFP, which permits further imaging experiment on animals. We firstly performed imaging experiments with porcine muscle tissues. NRFP was subcutaneously injected to two different spots of a piece of pork, one of which was then further injected with NE (Figure 4a). The two sites of injection could be clearly seen in 1H MRI images due to the Gd chelate in NRFP, which could act as a T 1 contrast agent, leading to signal enhancement of nearby tissues (Figure 4b). As revealed by 19F MRI (Figure 4b), the tissues injected with NRFP and NE showed strong signals, the intensity of which peaked at 60 min after injection and decreased gradually. In contrast, no significant signals were observed for the tissues injected with NRFP alone. There observations are further confirmed by SNR analysis (Figure 4c). These results illustrate the feasibility of imaging NE in deep tissues by 19F MRI with NRFP.

Figure 4

(a) A schematic illustration showing the protocol for ex vivo 19F MRI. The center frequency corresponding to the 19F chemical shift at −62.7 ppm was chosen for 19F MRI. (b) Representative 1H and 19F MR images of a piece of pork at pre‐injection (0 min) or indicated time points after subcutaneous injection of prepared NRFP solutions (200 μL), which were 10 mm NRFP in 1 × PBS incubated with 50 equiv. NE (Top) and 10 mm NRFP in 1 × PBS alone (Bottom). The sites of injection are indicated by white circles. (c) SNR analysis on the results of the experiment as indicated in (b) (n=3). The SNR for the 19F MR image of NRFP without NE treatment at 0 min was set as 1.0 and the other SNRs were normalized accordingly.

In vivo Imaging of NE via 19F MRI with NRFP

Encouraged by the promising results of ex vivo imaging experiments, we further investigated the in vivo imaging of NE with NRFP. The left hinder limbs of BALB/c mice were injected with a solution of NRFP and the right hinder limbs were injected with a solution of NRFP and NE. Then the mice were subjected to MRI (Figure 5a). Due to the contrast‐enhancing effect of the Gd chelate in NRFP, bright 1H MRI signals were observed in both hinder limbs (Figure 5b). The bladder region also showed strong 1H MRI signals, which could be ascribed to the renal clearance of unreacted NRFP and GDTy cleaved from NRFP (Figure 5b). As revealed by 19F MRI, right hinder limbs showed strong signals, the intensity of which decreased over time, while left hinder limbs did not, indicating the successful in vivo imaging of NE. The bladder also exhibited strong signals, the intensity of which increased gradually. This phenomenon could be attributed to the presence of BTFBT that is cleaved from NRFP and accumulate in the bladder through renal clearance. There observations are further confirmed with SNR analysis (Figure 5c). To study the specificity of NRFP towards NE for in vivo imaging, we also subjected some BALB/c mice to MRI, whose left hinder limbs and right hinder limbs were injected with NRFP and GSH, and NRFP and H2O2, respectively (Figure S6). Similar bright 1H MRI signals could be seen in both hinder limbs and the bladder. More importantly, no significant 19F MRI signals were observed for both hinder limbs, indicating the excellent selectivity of NRFP towards NE for in vivo imaging. These results demonstrate the successful application of NRFP for in vivo deep‐tissue imaging of NE with anatomical details provided by 1H MRI and specific distribution offered by 19F MRI.

Figure 5

(a) A schematic illustration showing the protocol for in vivo 19F MRI. The center frequency corresponding to the 19F chemical shift at −62.7 ppm was chosen for 19F MRI. (b) Representative 1H and 19F MR images of living mice (BALB/c) at pre‐injection (0 min) or indicated time points after subcutaneous injection of prepared NRFP solution (200 μL), which contains 10 mm NRFP in 1 × PBS alone (left hinder limbs) and 10 mm NRFP in 1 × PBS incubated with 50 equiv. NE (right hinder limbs) for 3 h. (c) SNR analysis on the results of the experiment as indicated in (b) (n=3). The average SNR for the left thigh regions of the 19F MR images at 0 min was set as 1.0 and the other SNRs were normalized accordingly.

Conclusion

In summary, we have developed a small‐molecular 19F MRI probe (NRFP) by conjugating a fluorine‐containing moiety and a paramagnetic Gd chelates via a NE‐responsive aromatic thiocarbonate linkage, which allows for detection and imaging of NE via 19F NMR/MRI. Its capacity and specificity for sensing and imaging NE have been clearly illustrated by a series of in vitro, ex vivo, and in vivo detecting/imaging experiments. It is also noteworthy that NRFP and its cleaved products could undergo prompt renal clearance, which significantly minimizes their potential side effects. Though the previously reported fluorescence‐based methods have lower detection limits for NE, they are restricted to cellular imaging or ex vivo imaging due to the shallow tissue penetration and the interference of in vivo autofluorescence.[ , ] In contrast, our work demonstrates a promising means for selective detection and specific imaging of NE, a representative neurotransmitter involved in many biological processes, in deep tissues of living subjects in a noninvasive manner with low background interference. More importantly, the design strategy of this probe could be extended to the development of high‐performance probes aiming at sensing and imaging other bioactive species (e. g., biomarkers) in different biological systems.

Experimental Section

Synthesis

All chemicals were purchased commercial sources and used as received without further purification. Unless noted otherwise, all reactions were performed under inert (N2 or Ar) environments. Please see the Supporting Information for details.

Tri‐tert‐butyl 2,2’,2’’‐(1,4,7,10‐tertraazacyclododecane‐1,4,7‐triyl)tri‐acetate (1, tBu‐DO3A)

1, 2, and 3 were synthesized according to a previous report by our group. Briefly, cyclen (10.336 g, 60.0 mmol) and anhydrous sodium acetate (14.765 g, 180.0 mmol, 3.0 equiv.) were dissolved in 180.0 mL N,N‐dimethylacetamide (DMA). A solution of tert‐butyl bromoacetate (35.110 g, 29.1 mL, 180.0 mmol, 3.0 equiv.) in 40.0 mL DMA was added dropwise to the mixture. After vigorously stirred at room temperature (RT) for 48 h, the suspension was quenched by saturated sodium bicarbonate aqueous solution. The resulting mixture was concentrated. The residue was purified by flash chromatography (100 % CH2Cl2 to 15 % v/v CH3OH/CH2Cl2) on silica gel to give 1 (21.87 g, 42.5 mmol, 71 %) as a white powder.

1H NMR (600 MHz, CD3OD): δ 3.41 (4 H, s), 3.36 (2 H, s), 3.14 (4 H, t, J=6.5 Hz), 2.98 (4 H, t, J=6.6 Hz), 2.79 (4 H, t, J=5.5 Hz), 2.69 (4 H, t, J=6.5 Hz), 1.48 (9 H, s), 1.48 (18 H, s), the peak for –NH– on the ring could not be observed; 13C NMR (151 MHz, CD3OD): δ 171.18, 81.20, 56.23, 50.37, 49.09, 45.62, 27.03. HR‐ESI‐MS (m/z) calculated for C26H51N4O6 [M+H]+: 515.3803, found: 515.3820.

Tri‐tert‐butyl 2,2′,2′′‐(10‐(2‐methoxy‐2‐oxoethyl)‐1,4,7,10‐tetraazacy‐clododecane‐1,4,7‐triyl)triacetate (2)

1 (6.172 g, 12.5 mmol) and K2CO3 (3.455 g, 25.0 mmol, 2 equiv.) were dissolved in 40 mL anhydrous acetonitrile. Subsequently, methyl chloroacetate (2.984 g, 2.40 mL, 27.5 mmol, 2.2 equiv.) was added to the mixture. The suspension was vigorously stirred at room temperature for 12 h. After filtration, the filtrate was concentrated and purified by flash chromatography (100 % CH2Cl2 to 20 % v/v CH3OH/CH2Cl2) to give 2 (6.412 mg, 10.9 mmol, 87 %) as a yellow oil.

HR‐ESI‐MS (m/z) calculated for C29H55N4O8 [M+H]+: 587.4014, found: 587.4047.

2‐(4,7,10‐Tris(2‐(tert‐butoxy)‐2‐oxoethyl)‐1,4,7,10‐tetraazacyclodo‐decan‐1‐yl)acetic acid (3)

2 (5.864 mg, 10.0 mmol) was dissolved in a solution (45 mL) of NaOH (1.2 g) in dioxane/H2O (v/v=2 : 1). The mixture was vigorously stirred at 50 °C overnight and concentrated in vacuo. The residue was added to 30 mL DI water and extracted with dichloromethane (DCM, 3×60 mL). The organic phases were collected, dried with Na2SO4, and concentrated to give 3 (5.372 g, 9.0 mmol, 90 %) as a white foamy solid.

HR‐ESI‐MS (m/z) calculated for C28H53N4O8 [M+H]+ and C28H52N4O8Na [M+Na]+: 573.3858 and 595.3677, found: 573.3873 and 595.3699.

‐Butyl (4‐((((3,5‐bis(trifluoromethyl)phenyl)thio)carbonyl)oxy)‐phenethyl)carbamate (4)

3,5‐Bis(trifluoromethyl)thiophenol (2.462 g, 1.69 mL, 10 mmol) and triphosgene (1.484 g, 5.0 mmol) were dissolved in 20 mL DCM and stirred at 0 °C. Dry pyridine (827 μL, 10 mmol) in 5.0 mL DCM was added dropwise within 15 min. The mixture was stirred vigorously for 1 h and washed with 100 mL ice water. The organic phase was collected and dried with Na2SO4. Subsequently, N‐Boc‐tyramine (2.372 g, 10.0 mmol) and Et3N (2.77 mL, 20.0 mmol) in 30 mL N,N‐dimethylformamide (DMF) were quickly added to the organic phase and the mixture was continuously stirred overnight at RT. The crude product was concentrated and purified by flash chromatography (100 % hexane to 15 % v/v EtOAc/hexane) to give 4 (1.92 g, 3.77 mmol, 75 %) as an off‐white powder.

1H NMR (600 MHz, CDCl3): δ 8.08 (2 H, s), 7.95 (1 H, s), 7.25 (2 H, d, J=4.2 Hz), 7.15 (2 H, d, J=4.3 Hz), 4.56 (1 H, br s), 3.38 (2 H, s), 2.82 (2 H, t, J=6.8 Hz), 1.45 (9 H, s); 13C NMR (151 MHz, CDCl3): δ 166.98, 155.81, 149.60, 137.69, 134.42, 132.91‐132.24 (m), 130.69, 129.99, 123.60‐123.50 (m), 121.81, 121.05, 79.36, 41.68, 35.63, 28.39; 19F NMR (564 MHz, CDCl3): δ −62.95 (6 F, s). HR‐ESI‐MS (m/z) calculated for C22H21F6NO4SNa [M+Na]+: 532.0988, found: 532.1005.

O‐(4‐(2‐Aminoethyl)phenyl) S‐(3,5‐bis(trifluoromethyl)phenyl) carbonothioate (5)

4 (916 mg, 1.8 mmol) was dissolved in 5 mL of DCM/trifluoroacetic acid (TFA) (v:v=2 : 1). The reaction was stirred for 3 h at RT. Once the reaction was completed, the reaction mixture was concentrated in vacuo to give crude 5 as a viscous liquid, which was directly used in the next step without further purification.

HR‐ESI‐MS (m/z) calculated for C17H14F6NO2S [M+H]+ and C17H13F6NO2SNa [M+Na]+: 410.0644 and 432.4063, found: 410.0667 and 432.4067.

Tri‐tert‐butyl 2,2′,2′′‐(10‐(2‐((4‐((((3,5‐bis(trifluoromethyl)phenyl)‐thio)carbonyl)oxy)phenethyl)amino)‐2‐oxoethyl)‐1,4,7,10‐tetraaza‐cyclododecane‐1,4,7‐triyl)triacetate (6)

3 (714 mg, 1.2 mmol), EDC⋅HCl (1.15 g, 5.0 equiv.), NHS (414 mg, 3.0 equiv.) and a catalytic amount of 4‐dimethylaminopyridine (DMAP) were dissolved in 40 mL DCM and stirred for 3 h at RT. The resulting mixture was washed with water (30 mL) and saturated saline (30 mL) for two times. The organic phase was collected and dried over Na2SO4 to give crude 6 as a yellow liquid, which was directly added to a solution of 20 mL DCM containing crude 5 from last step and a catalytic amount of Et3N. The mixture was stirred overnight and was concentrated in vacuo to give crude 7 as a yellow liquid, which was directly used in the next step without further purification.

HR‐ESI‐MS (m/z) calculated for C45H63F6N5O9SNa [M+Na]+: 986.4143, found: 986.4151.

2,2′,2′′‐(10‐(2‐((4‐((((3,5‐Bis(trifluoromethyl)phenyl)thio)carbonyl)‐oxy)phenethyl)amino)‐2‐oxoe‐thyl)‐1,4,7,10‐tetraazacyclododecane‐1,4,7‐triyl)triacetic acid (7, Ligand)

Crude 6 was dissolved in 5 mL trifluoroacetic acid (TFA) and the mixture was stirring at RT for 12 h. The solution was concentrated in vacuo to give crude 7 as a yellow viscous liquid, which was directly used in the next step without further purification.

19F NMR (564 MHz, 10 % D2O/H2O): δ −62.69 (6 F, s). HR‐ESI‐MS (m/z) calculated for C33H40F6N5O9S [M+H]+ and C33H39F6N5O9SNa [M+Na]+: 796.2445 and 818.2665, found: 796.2449 and 818.2665.

Norepinephrine‐responsive 19F MRI Probe (8, NRFP)

Crude 7 was dissolved in 30 mL CH3OH: H2O (v:v=1 : 1) and GdCl3⋅6H2O (1.947 g, 3.0 mmol, 3.0 equiv.) was added. The pH was adjusted to ≈3 with 0.1 m NaOH and the resulting mixture was stirred overnight at 45 °C. The mixture was concentrated in vacuo, dissolved in 40 mL CH3CN and further purified by gradient elution using HPLC (ZORBAX SB‐18 column from 2 % CH3CN/98 % H2O to 70 % CH3CN/30 % H2O 0‐60 min). The fraction between 24.5‐28.0 min was collected. Lyophilization gave 8 (215 mg, 19 % from 4) as an off‐while solid.

19F NMR (564 MHz, 10 % D2O/H2O): δ −62.64 (6 F, br s, decoupled); HR‐ESI‐MS (m/z) calculated for C33H36F6GdN5O9SNa [M+Na]+: 973.1277, found: 973.1279.

NMR/MRI Instrumentation

All 1H NMR and 13C NMR experiments were carried out on a Bruker AVANCE III HD Ascend spectrometer (600 MHz for 1H, 151 MHz for 13C) with tetramethylsilane as an internal referencing standard. All 19F NMR experiments were carried out on the same instrument (564 MHz for 19F) using a 5 mm BBFO cryoprobe and the spectra were acquired with 18 μs delay and 64 scans. 19F NMR samples were prepared in 10 % D2O/H2O containing sodium trifluoroacetate (CF3COONa, chemical shift at −75.4 ppm) as an internal standard. All 19F magnetic resonance imaging (19F MRI) with corresponding 1H MRI were performed on a Bruker BioSpec 94/20 system (400 MHz for 1H and 376 MHz for 19F) equipped with a 40 mm (inner diameter) volume coil. Image acquisition, SNR analysis and pseudocolor rending were carried out with ParaVision 5.1 (Bruker BioSpin).

Preparation of NRFP used for 19F NMR/MRI

The final concentration of NRFP in stock solution was determined by 19F NMR using 0.1 mm CF3COONa as an internal reference. Various analytes were incubated in NRFP solutions as indicated and the resulting solutions were subjected to 19F NMR. Among the analytes, ONOO− was generated by mixing NaNO2 with H2O2, and ROO⋅ was made by dissolving 2,2′‐azo‐bis(2‐amidinopropane)dihydrochloride (AAPH) in water. All other analytes were purchased from commercial sources and used as received.

1H/19F MRI

For acquiring 1H MR images, a RARE sequence was used with the following parameters: T R/T E=1000 ms/8.5 ms, flip angle=180°, FOV=4×4 cm2, slice thickness=1 mm, matrix=256×256, average=4 (NEX=4). The total acquisition time for each time point was about 3.2 min.

For acquiring 19F MR images, a RARE sequence was used with the following parameters: T R/T E=800 ms/8.5 ms, flip angle=180°, FOV=4×4 cm2, slice thickness=20 mm, matrix=32×32, average=196 (NEX=196). The total acquisition time for each time point was about 15.68 min.

Cell Culturing

L02 cells and HepG2 cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10 % fetal bovine serum (FBS, Hyclone). All cells were maintained in a humidified atmosphere containing 5 % CO2.

Cytotoxicity Evaluation

3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide (MTT) assays were used for cytotoxicity evaluation. L02 cells and HepG2 were firstly seeded into a 96‐well plate at a density of 1×104 cells/well and incubated for 24 h. After washed twice with PBS, the cells were incubated with fresh media containing NRFP at various concentrations (0, 0.5, 1, 2, 4, 6, 8, 10, 15 and 20 mm) for another 24 h. The concentration of NRFP was determined by ICP‐MS. Subsequently, the media of each well was replaced with 100 μL of fresh media containing MTT (0.5 mg mL−1) and the plate was incubated for 4 h at 37 °C. Then, the media in each well were replaced with 200 μL DMSO. The absorbance at 490 nm of each well was measured on a MultiSkan FC microplate reader immediately. Cell viabilities were calculated accordingly.

Animal Ethics

Animal experiments were conducted according to the protocols approved by the Institutional Animal Care and Use Committee of Xiamen University.

10 mm NRFP in 1×PBS solutions were subcutaneously injected to two spots of a piece of pork. 50 equiv. NE dissolved in PBS was subcutaneously injected into one of the two spots as indicated. Then the pork was subjected to 19F MRI.

10 mm NRFP in 1×PBS buffer (pH 7.4) were incubated alone or together with 50 equiv. analytes (NE, GSH or H2O2) for 3 h. Then, the solutions were subcutaneously injected into the hinder limbs of BALB/c mice as indicated. Then the mice were subjected to 19F MRI.

Conflict of interest

The authors declare no conflict of interest.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Click here for additional data file.