Synthesis and biological evaluation of rhein-based MRI contrast

Oct 31, 2018 - Here, we reported three novel MRI contrast agents, GdL1, GdL2 and ... 256 breast carcinoma injected with a vascular disrupting agent CA...
0 downloads 0 Views 5MB Size
Download PDF
Article Cite This: Anal. Chem. 2018, 90, 13249−13256

pubs.acs.org/ac

Synthesis and Biological Evaluation of Rhein-Based MRI Contrast Agents for in Vivo Visualization of Necrosis Li Bian,†,‡ Meng Gao,‡ Dongjian Zhang,‡ Aiyan Ji,‡,§ Chang Su,‡,§ Xinghua Duan,‡,§ Qi Luo,‡,§ Dejian Huang,†,‡ Yuanbo Feng,†,‡ Yicheng Ni,‡,∥ Zhiqi Yin,§ Qiaomei Jin,*,‡ and Jian Zhang*,‡

Anal. Chem. 2018.90:13249-13256. Downloaded from pubs.acs.org by UNIV OF WINNIPEG on 12/02/18. For personal use only.

†

Afliated Hospital of Integrated Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing 210028, Jiangsu Province, P. R. China ‡ Laboratories of Translational Medicine, Jiangsu Province Academy of Traditional Chinese Medicine, Nanjing 210028, Jiangsu Province, P. R. China § Department of Natural Medicinal Chemistry & State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing 210009, Jiangsu Province, P. R. China ∥ Theragnostic Laboratory, Campus Gasthuisberg, KU Leuven, 3000 Leuven, Belgium S Supporting Information *

ABSTRACT: Early and accurate assessment of therapeutic response to anticancer therapy plays an important role in determining treatment planning and patient management in clinic. Magnetic rseonance imaging (MRI) of necrosis that occurs after cancer therapies provides chances for that. Here, we reported three novel MRI contrast agents, GdL1, GdL2, and GdL3, by conjugating rhein with gadolinium 2-[4,7,10-tris(carboxymethyl)-1,4,7,10tetraazacyclododec-1-yl]acetic acid (Gd-DOTA) through different linkers. The T1 relaxivities of three probes (7.28, 7.35, and 8.03 mM−1 s−1) were found to be higher than that of Gd-DOTA (4.28 mM−1 s−1). Necrosis avidity of GdL1 was evaluated on the rat models of reperfused liver infarction (RLI) by MRI, which showed an increase of T1-weighted contrast between necrotic and normal liver during 0.5−12 h. Besides, L1 was also labeled with 64Cu to assess its necrosis avidity on rat models of RLI and muscle necrosis (MN) by a γ-counter. The uptakes of 64CuL1 in necrotic liver and muscle were higher than those in normal liver and muscle (P < 0.05). Then, the ability of GdL1 to assess therapeutic response was tested on rats bearing Walker 256 breast carcinoma injected with a vascular disrupting agent CA4P by MR imaging. The signal intensity of tumoral necrosis was strongly enhanced, and the contrast ratio between necrotic and viable tumor was 1.63 ± 0.11 at 3 h after administration of GdL1. Besides, exposed DNA in necrosis cells may be an important mechanism of three probes targeting to necrosis cells. In summary, GdL1 may serve as a promising MRI contrast agent for accurate assessment of treatment response.

A

objectively and quantitatively.5−7 Many studies suggested that tumor treatment response could be identified earlier by changes in the 18F-FDG signal than those in measured size.8 However, imaging of tumor metabolic status is not specific due to the fact that uptake of 18F-FDG may also be influenced by inflammation, which could be mistaken for cancer, especially after treatment.9 Molecular imaging that can visualize cellular processes should possess high specificity and sensitivity for assessment of therapy response. Clinically, many cancer therapies are aimed at inducing tumor necrosis10,11 so that visualization of necrosis in tumor could provide a direct method to assess efficacy of anticancer therapies. Necrosis leads to the rupture of cell membrane and the release of cell contents; thus, numerous

ssessment of treatment response can help to identify those nonresponsive patients to prevent them from unnecessary side effects of ineffectual treatment. Meanwhile, alternative therapeutic strategies can be suggested to the patients in time to achieve beneficial effects. Accordingly, early assessment of therapeutic response to anticancer therapy plays a crucial role in determining whether to continue or alter the therapeutic strategy.1,2 Current evaluation of the treatment response in the clinic is based on the traditional WHO and RECIST criteria for decades, which mainly rely on changes in tumor size.3,4 Nevertheless, the changes in the physical dimension of a solid tumor often occur too late to identify patients who do not respond to treatments in a timely manner. Furthermore, dimensional measurement can be hampered by scar formation and necrotic tissue. As a functional imaging technique, 18F-FDG PET can assess the treatment response by measurement of tumor metabolism © 2018 American Chemical Society

Received: April 26, 2018 Accepted: October 31, 2018 Published: October 31, 2018 13249

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256

Article

Analytical Chemistry

The relaxivities (r1 and r2) were then determined as the inverse of the relaxation time per mM. Transmetalation. Possible transmetalation by zinc(II) ions was used as a stability test for the three complexes. Transmetalation determination of GdLn is described in the Supporting Information. Log P Determination and in Vitro Stability. Log P values and in vitro stability determination of GdLn are described in the Supporting Information. In Vitro Cell Cytotoxicity. Methods of in vitro human lung cancer A549 cells cytotoxicity of GdLn are described in the Supporting Information. Interaction with DNA. The interactions of compounds with DNA were examined by fluorescence and relaxometry. Methods are described in the Supporting Information. Imaging of Cells. The human lung cancer A549 cells and human hepatocellular carcinoma HepG2 cells were cultured in RPMI-1640 supplemented with 10% fetal bovine serum maintained at 37 °C, 5% CO2. As is well-known in humans, mature red blood cells (RBCs) lack a cell nucleus that contains most of the cell’s genetic material DNA molecules. Here, RBCs were used as a blank control. Blood samples from health examination were provided by clinical laboratory. The study was approved by the local ethic committee at the Jiangsu Province Hospital of Integrated Traditional Chinese and Western Medicine. The whole blood was centrifuged at 2000 rpm for 10 min to separate the RBCs from plasma. The RBCs were then washed three times with 10 mM PBS (pH = 7.4). The washed RBCs were resuspended in an isotonic solution of PBS at a concentration of 10% by volume. Cell necrosis was induced after culture for 1 h under intense hyperthermia at 57 °C according to Perek et al.21 A549 and HepG2 cells were put into 2 mL eppendorf tubes at a density of 2 × 106 cells/tube for six tubes. Three tubes of A549 and HepG2 cells were incubated for 1 h under intense hyperthermia at 57 °C to induce necrosis, and another three tubes of nontreated A549 and HepG2 cells were used as control. Two types of A549 and HepG2 cells were then respectively incubated at 37 °C for 30 min with the 3 × 10−3 mM GdL1 in 300 μL of PBS (pH 7.4) and then washed twice with PBS and resuspended with 1% heated agar. The processing procedures of RBCs were similar to that of A549. MRI of cells was carried out on 1.5 T MR magnet (Echo speed; GE Co., NY), and 1% agar was used for data calibration. Animal Models of Necrosis. Adult male Sprague−Dawley (SD) rats weighing 280−300 g were provided by the Experimental Animal Center, Jiangsu Province Academy of Traditional Chinese Medicine. All the animal experiments were implemented with the approval of the institutional animal care and research advisory. Rat models of reperfused liver infarction (RLI) and muscular necrosis (MN) were generated as previously described.22 Briefly, rats were anesthetized with 40 mg/kg sodium pentobarbital. Under laparotomy, hepatic ischemia was induced by ligating the hilum of the right liver lobe for 3 h followed by removal of the ligature for reperfusion. Each rat with RLI was also intramuscularly injected with 0.2 mL of absolute alcohol in the right leg to induce muscle necrosis. After the above procedure, each model animal was allowed to recover for 6−8 h. The peritoneal cavity of a SD rat was injected with Walker 256 rat breast carcinoma cell line. After incubating for 2 weeks, the animal was sacrificed and the ascites tumor cells were

biomarkers for necrosis after cancer therapies provide chances for molecular imaging. As a common biomarker for necrosis, DNA becomes accessible after breakdown of the cell plasma membrane,12 which can be a potential target for molecular imaging of necrosis-related diseases and therapies. Anthraquinones as well-known intercalators of DNA have been studied to show prominent targetability to necrotic tissues. 13−15 Many studies showed that the targeting mechanism of anthraquinone-based necrosis-avid compounds may be through the binding with exposed DNA.16,17 Among these anthraquinones, rhein with excellent necrosis targetability and better pharmacokinetic properties has been labeled with 131I, 18F, and 99mTc for imaging necrosis in myocardial infarction (MI) with demonstrated favorable diagnostic capacity.15−17 However, nuclear imaging techniques are associated with radiation risk and poor anatomical resolution. On the contrary, magnetic resonance imaging (MRI) can offer higher spatial resolution on anatomical images.18−20 Therefore, this study first investigated Gd complexes of rhien via thermodynamically and kinetically stable macrocyclic chelator 2-[4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododec-1yl]acetic acid (DOTA) through three linkers of different lengths (Figure 1) and evaluated their necrosis targetability in

Figure 1. Schematic structures of GdLn (n = 1, 2, 3).

rats with reperfused liver infarction and muscular necrosis (RLI and MN) and explored their necrosis avidity mechanism by DNA binding studies and in vitro magnetic resonance imaging (MRI) of cells. Finally, this study preliminarily evaluated the potential of GdL1 to assess the treatment efficacy of vascular disrupting agents on rats bearing W256 breast carcinoma.

■

EXPERIMENTAL SECTION Synthesis of GdLn. Figure S1 shows the synthesis scheme for GdL1, GdL2, and GdL3. The detailed synthetic procedures and structural characterization are described in the Supporting Information (Figure S2−S10). Relaxivities. The relaxivity of MRI contrast agents is defined as its ability to increase the relaxation rates of the surrounding water proton spins. For each complex, five samples were prepared separately with a concentration varying between 0 and 2 mM Gd3+ in PBS (pH 7.4) solution. The relaxation time (T1 and T2) measurements of each sample were performed at 0.5 T (21.25 MHz, Niumag Analytical Instrument Corporation, Shanghai). The temperature was controlled at 25 °C. 13250

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256

Article

Analytical Chemistry collected and diluted in saline to 5 × 106 cell/mL. W256 cells of 200 μL suspension were injected subcutaneously to the left flank of SD rats. The experiments were performed once the tumor reached a size of approximately 600 mm3. In Vivo MR Imaging of Rat Model with Reperfused Liver Infarction. Six RLI model rats were randomly divided into group A and B, three in each group. Model rats in group A were intravenously injected with 0.1 mmol/kg GdL1. Rats in group B as control group were intravenously injected with 0.1 mmol/kg gadolinium 2-[4,7,10-tris(carboxymethyl)-1,4,7,10tetraazacyclododec-1-yl]acetic acid (Gd-DOTA). All rats were scanned with T1-weighted imaging (T1WI). Before and after administration, T1WI was performed at different time points. The model rats were intravenously injected with 0.2 mL of 0.5% evens blue 6 h in advance of sacrifice. When MRI scan at 12 h was completed, the model rats were sacrificed, and 3 mm sections were frozen/cut at the axial plane, which could be matched with MR images and photographed. Preparation of 64CuL1. 64CuL1 was prepared by incubating 64 CuCl2 (37−222 MBq, provided by Beijing Cancer Hospital) in 0.5 M HCl (1−6 μL) with L1 (30 μg, 100 μL of 0.1 M ammonium acetate, pH 5.5) at 50 °C for 30 min. The labeled compound was purified through a Sep-Pak C18 cartridge using water (5 mL) and eluted with ethanol (2 mL). Then the solvent was removed, and the residue was dissolved in 0.9% saline. Quality control analysis was carried out using RP-HPLC equipped with a radio detector (Berthold Technologies, Germany) and an Eclipse Plus C18, Analytical (4.6 × 250 mm, 5 μm). The mobile phase was 0.1% TFA in acetonitrile/ 0.1% TFA in water (28:72, v/v), and the flow rate of 1 mL/ min was used as elution condition. Biodistribution Study and Histochemical Staining. Twelve RLI and MN model rats were randomly divided into three groups (n = 4 each). Rats were intravenously administered with 64CuL1 under anesthesia and sacrificed at 3, 12, and 24 h post injection (p.i.), respectively. The major organs were removed, weighed, and counted for radioactivity by the γ counter (PerkinElmer 1480; Waltham, MA). Physical decay and background radiation were corrected during counting. The percentage of the injected dose per gram of tissues (%ID/g) was calculated, and the values were expressed as mean ± standard deviation (SD). Tissues of normal and necrotic liver were cut into 10 μm frozen sections, which were stained by hematoxylin and eosin (H and E). Stained sections were digitally photographed to distinguish the necrotic tissue from normal tissue. Photomicrographs with magnification at ×200 were acquired using an optical microscope (Axioskop; Zeiss, Oberkochen, Germany). CA4P Treatment and in Vivo MR Imaging. CA4P was dissolved in 0.9% saline. Six rats bearing Walker 256 breast carcinoma were randomly divided into group A and B. The rats in group A were injected intravenously with 10 mg/kg CA4P, and the rats in group B were treated with equal volume of 0.9% saline (control group). MRI studies were performed at 24 h after CA4P administration. All rats received T1WI. After contrast administration, T1WI was performed at 0 and 3 h. All parameters are as follows. When MRI scanning was completed, tumor tissues were cut into 10 μm frozen sections, which were stained with H and E. MRI Sequences and Imaging Analyses. MRI was performed on a clinical 1.5 T MR magnet (Echo speed; GE Co., NY) by using a birdcage-like rat coil with orthogonal axes

and 5 cm inner diameter (Shanghai Chenguang Medical Technology Co., SH). All animals were anesthetized with isoflurane in a mixture of 80% room air and 20% oxygen. Spin−echo (SE) T1-weighted imaging (T1WI) was performed with the following parameters: TR = 550 ms, TE = 24 ms, FOV = 100 × 100 mm2, imaging acquisition matrix = 224 × 192, a total examination time of 3 min 5 s. Contract agents of Gd-DOTA and GdL1 were used at the same dosage of 0.1 mmol/kg. MRI analyses were performed by using the built-in software of the system (GE, AW 4.3, GE Healthcare Bio-Sciences, Pittsburgh, PA). The signal intensities (SIs) of cells and tissue were measured with an operator-defined circular region of interest (ROI) with 3 mm2 on T1WI. Signal intensity (SI) of cells were corrected by SI′cells = SIcells − SIagar. Contrast ratio was used to express the visual conspicuity of adjacent component on the one image and calculated with the following formula: CR = SIA/SIB. Relative enhancement (RE) was used to describe the degree of SI enhancement, using the following formula: RE% = (SIpostcontrast − SIprecontrast)/SIprecontrast × 100. Statistical Analysis. All numerical data were expressed as mean ± standard deviation (SD). Statistical analysis was performed with SPSS 13.0 (SPSS Inc., Chicago, IL). Differences were analyzed using the Student t test. P < 0.05 was considered significant.

■

RESULTS AND DISCUSSION Synthesis of GdLn. GdLn was prepared according to Figure S1. The purity of all synthetic compounds was greater than 95% after purification. The identification of all compounds was determined by 1H NMR, 13C NMR, and ESI-MS. The schematic structures of GdL1, GdL2, and GdL3 are shown in Figure 1. MRI contrast agents should have low toxicity due to high dose. Monomeric anthraquinones have been used for food, cosmetics, dye, and pharmaceuticals,23 which may be safe substrates for necrosis-avid contrast agents. Relaxivities. The longitudinal relaxivities (r1) at 21.25 MHz and 25 °C in PBS (PH 7.4) were determined by plotting 1/T1 against the concentration of GdLn, and the longitudinal relaxivities of GdL1, GdL2, GdL3, and Gd-DOTA were 7.28, 7.35, 8.03, and 4.28 mM−1 s−1, respectively (Table 1). The Table 1. Characterization of GdLn (n = 1, 2, 3) and GdDOTA 21.25 MHz, 25 °C, PBS 7.4 GdL1 GdL2 GdL3 Gd-DOTA

r1 (mM−1 s−1)

r2 (mM−1 s−1)

log P

7.28 7.35 8.03 4.28

14.05 14.39 17.24 8.82

−2.15 ± 0.01 −1.32 ± 0.01 −0.71 ± 0.01 NA

results showed that conjugations of Gd-DOTA to rhein through three different lengths of linkers all possess fairly higher relaxivity than Gd-DOTA does, which could be interpreted by their heavier molecular weight that slowed down their tumbling rate. This slower rate matched well with the Larmor frequency of water molecules and posed a higher relaxivity, compared to those lower molecular weight complexes with fast tumbling rates.24,25 Such probes with higher relaxivity can shorten the T1 and/or T2 value more at equivalent concentrations than those agents with a lower relaxivity do. The advantage of this difference can be utilized to 13251

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256

Article

Analytical Chemistry

study possess high thermodynamic stability and kinetic inertness, making these compounds safe in MRI as contrast agents. In Vitro Cell Cytotoxicity. GdL1, GdL2, GdL3, and GdDOTA were evaluated against human lung cancer A549 cells using the MTT assay after 24 h incubations. As shown in Figure S11, the proliferation and viability of A549 cells were not affected. Even at the maximum concentration (1 mM) of GdL1, GdL2, and GdL3, the viability of A549 cells retained 90.78%, 87.27%, and 87.71%, respectively, relative to control. These observations indicate that GdL1, GdL2, and GdL3 showed negligible cytotoxicity in the concentration range required for MR imaging. DNA Binding Experiment. The ethidium bromide (EB) is a well-known intercalator. The DNA-EB fluorescence quenching experiments were employed to further investigate the interaction mode between the compound and DNA. Seven compounds (rhein, L1, L2, L3, GdL1, GdL2, and GdL3) were studied as quenchers. The fluorescence profile of DNA-EB after addition of GdL1 as a representative example is shown in Figure 4A. The KSV (Stern−Volmer quenching constant) values for rhein, Ln and GdLn were 1.01, 1.36, 1.32, 1.40, 2.28, 2.19, and 2.30 × 104 M−1 in the same condition. This indicated that the partial replacement of EB bound to DNA by these seven compounds. Consequently, seven rhein derivatives might bind to DNA in intercalative mode. The quenching constants of GdLn were greater than those of four other compounds; this can be explained by the fact that positive Gd3+ interact with negative phosphate backbone through electrostatic interaction, which can strengthen the interaction between GdL1 and CtDNA. According to the results of DNA-EB fluorescence quenching experiments, we can conclude that three contrast agents (GdL1, GdL2, and GdL3) can bind to DNA with similar binding ability. Consequently, we only chose GdL1 for the following study as a representative example. The interaction of GdL 1 with DNA was simultaneously determined by relaxometry. Figure 4B shows that addition of DNA with increasing concentration decreased the T1 (GdL1) from 695 to 564 ms. R1 increased from 1.43 ± 0.02 to 1.78 ± 0.02 (mM s)−1. The increased R1 likely reflected a slower tumbling time (τR) when GdL1 bound to DNA, and similar effects have been remarked with many protein-binding Gd contrast agents.28 These data indicated that GdL1 can bind to DNA and exhibited an increased relaxivity. In Vitro MRI of Cells. The MR image of hyperthermiatreated A549 and HepG2 cells (N-A549 and N-HepG2) showed hyperintense (bright), while nontreated A549 and HepG2 cells (V-A549 and V-HepG2) and RBCs (N-RBCs and V-RBCs) showed relatively hypointense (Figure 5A). The signal intensities of N-A549 and N-HepG2 were significantly higher than that of V-A549 and N-HepG2 (P < 0.01, Figure 5B), and the contrast ratios were 1.55 ± 0.02 and 2.11 ± 0.05,

lower the dose or to generate more signal at equivalent doses, potentially lowering toxicity and/or improving detection or delineation of lesions. Transmetalation. Possible transmetalation by zinc(II) ions was used as a stability test for GdL1, GdL2, GdL3, and GdDOTA. This transmetalation process is monitored by measuring the relaxation rate of the solution. The relative water proton paramagnetic longitudinal relaxation rate R1P(t)/ R1P(0), which is defined as the ratio of relative value of R1P at any time t [R1P(t)] versus its initial value [R1P(0)], is used to estimate the extent of transmetalation. As observed in Figure 2,

Figure 2. Evolution of the relative water proton paramagnetic longitudinal relaxation rate R1P(t)/R1P(0) versus time (from 0 to 72 h) for GdL1, GdL2, GdL3, and Gd-DOTA. Initial concentrations of gadolinium complexes and ZnCl2 were 2.5 mM in phosphate buffer at pH = 7.4, T = 37 °C, B = 0.5 T.

GdL1, GdL2, and GdL3 all processed very little transmetalation in 72 h and showed high kinetic stability (>92% of the relaxation rate) the same as Gd-DOTA. This results are in accordance with other previously reported Gd-DOTA complexes.26 Log P Determination. The results of octanol−water partition coefficients (log P) are presented in Table 1. GdL1 shown the most negative log P value (−2.15 ± 0.01), and GdL3 with six carbon atom linker was the most hydrophobic (log P value −0.71 ± 0.01). GdL1 has the best water solubility, which meets the requirements of MRI contrast agents in clinic. In Vitro Stabilities of GdLn. The gadolinium-based MRI contrast agents consist of Gd ions (Gd3+) bound tightly by chelating agents to form a stable complex, which can mitigate the substantial natural toxicity of the free Gd ion.27 Stability refers to the ability of chelating agent to avoid losing the Gd ion. Therefore, it is necessary to determine the stability of contrast agent. In this study, the HPLC profiles of stability studies are shown in Figure 3. Three contrast agents all turned out to be stable after incubation in rat plasma at 37 °C for 24 h. DOTA is an octadentate ligand that can coordinate with lanthanide ions with four carboxylate and four amino groups. Consequently, contrast agents with DOTA synthesized in this

Figure 3. HPLC chromatograms of GdLn after incubation in rat plasma at 37 °C for 24 h. GdL1 (A), GdL2 (B), and GdL3 (C). 13252

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256

Article

Analytical Chemistry

respectively. The signal intensities showed no significant difference between N-RBCs and V-RBCs (P > 0.05). These results indicated that the uptake of GdL1 by necrotic A549 or HepG2 cells was greater than that by viable A549 or HepG2 cells, which meant that GdL1 can image necrotic cells specifically. However, the uptake of GdL1 by necrotic RBCs and by viable RBCs showed no significant difference. This can be explained that A549 and HepG2 cells contain DNA while RBCs do not. Therefore, the necrosis avidity of GdL1 may due to the interaction between GdL1 and DNA. Reliable biomarkers are critical to evaluate the response to treatment in an early phase. Phosphatidylserine (PS) that exposed on the outer leaflet of the plasma membrane in early apoptosis for evaluation of therapeutic effect is realized by using 99mTc-HYNIC-Annexin A5.29,30 However, phosphatidylserine externalization is not restricted to apoptosis but also found in activated endothelium of tumor vasculature, activated megakaryocytes, macrophages, necrosis, and autophagy.31,32 Dissimilarly, E-DNA as a specific biomarker of necrosis emergesin high concentrations in the necrosis tissue. Although E-DNA can be detected in the blood, the concentration of EDNA in the blood should be significantly lower than that in necrotic tissue due to the large blood volume and the presence of serum nucleases.33 These can be used to explain that EDNA possess high specificity as a reliable biomarker for imaging necrotic tissue. Therefore, the MRI contrast agents that target E-DNA will become a potential trend in the evaluation of tumor therapeutic effects. Radiochemistry. The radiochemical purity of 64CuL1 was greater than 95% as shown in Figure S12. The radiolabeling of 64 CuL1 was quantitative, on the basis of RP-HPLC showing only one peak representing 64CuL1, with a retention time of 7.25 min. On this RP-HPLC, free ligand L1 showed a retention time of 5.46 min. In Vivo MRI and Biodistribution Study of Rat Models with RLI and MN. The results shown in Figure 6 demonstrated that by plain scans necrotic livers in two groups appeared typically slightly hyperintense relative to normal livers on T1WI with CR of 1.29 ± 0.07 and 1.40 ± 0.07 in GdL1 and Gd-DOTA group, respectively. However, the boundary of necrotic tissue was not legible, which made lesion delineation difficult. Immediately after administration of GdL1, the signal intensity (SI) of normal liver was strongly enhanced (RE% = 55.38 ± 7.18%) but not the necrotic liver, leaving a reduced necrosis-to-liver contrast ratio (CR = 1.04 ± 0.12). The SIs of normal livers decayed gradually afterward, but the SIs of the necrotic liver were enhanced significantly from 0.5 to 12 h (RE% = 90.60 ± 25.11%, at 0.5 h; RE% = 114.29 ± 26.02%, at 3 h; RE% = 68.73 ± 8.63%, at 12 h), which made CR remain strong and stable over a persistent period (CR = 1.62 ± 0.09, at 0.5 h; CR = 2.09 ± 0.22, at 3 h; CR = 1.95 ± 0.15, at 12 h). The normal liver of Gd-DOTA group showed the same trend compared to GdL1 group, but the degree of enhancement was lower than that of normal liver of GdL1 group. The SIs of the necrotic livers were also moderately enhanced from 10 min to 3 h, with the maximum RE% appearing at 0.5 h (RE% = 25.66 ± 4.80%) and the maximum CR appearing at 1 h (CR = 1.57 ± 0.05) after Gd-DOTA administration, which were significantly lower than that shown in GdL1 group (P < 0.05). The results above indicated that only at delayed phase could necrotic liver be enhanced at discernible degrees in both groups. However, the timing for visualizing necrotic liver with GdL1 was much longer than that

Figure 4. Fluorescence emission spectra (excited at 520 nm) in the EB (0.60 × 10−5 M) and Ct-DNA (2.40 × 10−5 M) system with increasing amounts of GdL1. The arrows show the intensity changes upon the addition of increasing concentrations of GdL1 (up to down: 0, 1.13, 2.21, 3.26, 4.26, and 5.23 × 10−5 M, respectively) (A). The relaxation rates of GdL1 bound to DNA with different concentrations ([GdL1] = 0.25 mM, [DNA] = 0, 0.02, 0.04, 0.08, 0.15, 0.31 mM) (B).

Figure 5. MR images of human lung cancer A549 cells, human Hepatocellular carcinoma HepG2 cells, and red blood cells (RBC) after incubation with GdL1 for 30 min (A). The signal intensities of A549, HepG2, and RBC (B). N- indicates necroticheated at 57 °C for 1 h and then treated with GdL1 for 30 min, V- indicates viable treated with GdL1 for 30 min. 13253

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256

Article

Analytical Chemistry

Figure 6. Representative transversal T1-weighted MR images of rat model of reperfused liver infarction (RLI) taken at different time points and corresponding digital photographs of macroscopic section. White arrow indicates necrotic liver, S indicates stomach (A). Contrast ratios between necrotic liver and normal liver after the use of GdL1 and Gd-DOTA in rat with RLI. N liver indicates necrotic liver, V liver indicates viable liver (n = 3) (B). Relative enhancement of liver after the use of GdL1 and Gd-DOTA in rats with RLI (n = 3). N liver in GdL1 showed P < 0.05 versus in GdDOTA at 0.5 h (C). H and E images and micrograph of liver from rat model of RLI at 3 h after injection (H and E, ×200) (D).

the nontarget tissues was rapid, and the uptake was lower than 0.7%ID/g at 12 h p.i. H and E staining was used to distinguish the necrotic tissue from normal tissue. Histopathologic outcomes (Figure 6D) from H and E stained slices showed that pink areas correspond to necrotic tissue while purple areas correspond to viable tissue. In Vivo MRI of Rats Bearing W256. To noninvasively monitor the therapy response of Walker 256 breast carcinoma, CA4P was used to treat tumor by blocking tumor vessels in rats. T1-weighted images of preinjection and 0, 3 h postinjection were acquired. Before contrast enhancement, the implanted subcutaneous W256 tumors in both groups appeared isointense on T1WI (Figure 7A1, 7B1), which made lesion delineation difficult. Immediately after administration of GdL1, the central tumor in group A showed hypointense, whereas the tumor periphery strongly enhanced (Figure 7A2). The whole tumor looked hyperintense in group B (Figure 7B2) at the same time. At 3 h after administration of GdL1, the signal intensity at the center of tumor in group A further increased with a CR of 1.63 ± 0.11 (Figure 7A3). The signal intensity of whole tumor decreased in group B at 3 h after administration of GdL1. These results can be explained by the fact that GdL1 can target and visualize the necrotic tumor tissue. Histopathologic outcomes (Figure 7A4) from H and E stained slices revealed that necrosis was induced in tumor by the treatment of CA4P. Consequently, GdL1 showed the potential of assessing the response of malignant tumor to the treatment with CA4P. Transcatheter arterial chemoembolization (TACE) has been considered the best choice of palliative treatment for inoperable liver cancer patients.34,35 The aim of TACE was to occlude the arterial blood supply to the tumor, resulting in ischemia and tumor necrosis. Frequently, tumors located in the liver cannot be completely necrotized by TACE due to the collateral blood flow through individual surrounding blood vessels.36 Therefore, early assessing therapeutic response and treating residual tumor in time are important for achieving

with Gd-DOTA, which is presumably due to the necrosis affinity of GdL1. The biodistribution results of 64CuL1 in rat RLI and MN models are shown in Table 2. The uptake of 64CuL1 by Table 2. Biodistribution Analysis of 64CuL1 in Rat Models of Reperfused Liver Infarction and Muscular Necrosis at 3, 12, and 24 ha tissue blood lung heart spleen stomach kidney bladder S intestine L intestine brain normal liver necrotic liver normal muscle necrotic muscle ratios NL/L NM/M

3h 0.45 0.22 0.09 0.36 0.14 3.35 0.29 0.51 0.49 0.02 0.33 2.26 0.10 0.61

± ± ± ± ± ± ± ± ± ± ± ± ± ±

12 h 0.03 0.06 0.00 0.03 0.00 0.05 0.04 0.08 0.07 0.00 0.06 0.21 0.03 0.05

6.85 ± 0.51 6.08 ± 0.64

0.29 0.07 0.05 0.15 0.12 0.60 0.24 0.26 0.13 0.02 0.14 0.56 0.07 0.26

± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.08 0.01 0.00 0.01 0.01 0.15 0.06 0.04 0.02 0.00 0.03 0.07 0.01 0.05

4.00 ± 0.73 3.70 ± 0.46

24 h 0.24 0.03 0.04 0.12 0.09 0.51 0.22 0.10 0.11 0.01 0.10 0.34 0.06 0.17

± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.05 0.00 0.01 0.01 0.01 0.03 0.03 0.02 0.02 0.00 0.01 0.03 0.01 0.05

3.57 ± 0.47 2.90 ± 0.36

a Date are presented as %ID/g ± SD (n = 4). S intestine, small intestine. L intestine, large intestine. NL/L, necrotic liver/liver. NM/ M, necrotic muscle/muscle.

necrotic tissue (liver and muscle) was significantly higher than that by normal tissue (liver and muscle) at each time point (P < 0.05). Necrotic to viable liver and muscle ratios were 6.85 ± 0.51 and 6.08 ± 0.64 at 3 h; 4.00 ± 0.73 and 3.70 ± 0.46 at 12 h. These results showed that 64CuL1 was specifically avid to necrotic tissues. The clearance of 64CuL1 from the blood and 13254

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256

Article

Analytical Chemistry

Figure 7. MR images of rats bearing W256 tumor treated with CA4P at preinjection, 0, 3 h postinjection of GdL1, and corresponding H and E stained tumor section (A1−A4); white arrows indicate the necrotic areas of tumor. MR images of nontreated control rats bearing W256 tumor at preinjection, 0, 3 h postinjection of GdL1, and corresponding H and E-stained tumor section (B1−B4), (H and E, ×200).

more satisfactory treatment efficacy. Vascular disrupting agents (VDA) are new anticancer drugs that selectively blockade the tumor vessels, deprive the blood supply of malignant cells, and subsequently cause tumor necrosis.37 Combretastatin A-4 disodium phosphate (CA4P) is a promising VDA, which has shown remarkable antitumor effects in extensive preclinical and clinical studies.38 The therapeutic mechanism of TACE is similar to that of VDAs. Therefore, GdL1 could be used for assessing early efficacy of both TACE and VDAs for the treatment of cancers.

Jian Zhang: 0000-0002-8402-9753 Author Contributions

The manuscript was written through the contributions of all the authors. All authors have given their approval to the final version of this manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study was funded by the National Natural Science Foundation of China (No. 81473120, 81501536, 81771870) and the Hospital Pharmaceutical Research Foundation of Changzhou Siyao Pharm (No. 2014YX011).

■

CONCLUSION This study has developed three MRI contrast agents based on rhein that all showed good characteristics in vitro. We delightedly found that these three MRI contrast agents still showed good intercalation ability with DNA in vitro. Among these three contrast agents, GdL1 presented favorable necrosis targeting properties and provided accurate and early measures for assessing the therapy efficacy of CA4P. In summary, GdL1 may serve as a promising necrosis-avid contrast agent for early and accurate assessment of tumor response after treatment by MRI imaging.

■

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.8b01868. Synthesis and characterization of GdLn (n = 1, 2, 3); transmetalation; log P determination of GdLn, in vitro stability of GdLn, in vitro cell cytotoxicity, radiochemical purity of 64CuL1 (PDF)

■

REFERENCES

(1) Juweid, M. E.; Stroobants, S.; Hoekstra, O. S.; Mottaghy, F. M.; Dietlein, M.; Guermazi, A.; Wiseman, G. A.; Kostakoglu, L.; Scheidhauer, K.; Buck, A. J. Clin. Oncol. 2007, 25, 571. (2) Wahl, R. L.; Jacene, H.; Kasamon, Y.; Lodge, M. A. J. Nucl. Med. 2009, 50 (Suppl 1), 122S. (3) Miller, A. B.; Hoogstraten, B.; Staquet, M.; Winkler, A. Cancer 1981, 47, 207. (4) Watanabe, H.; et al. Gan to Kagaku Ryoho. 2009, 36, 2495− 2501. (5) Schmidt, T.; Lordick, F.; Herrmann, K.; Ott, K. J. Natl. Compr. Cancer Network 2015, 13, 239−247. (6) Avril, S.; Muzic, R. F.; Plecha, D.; Traughber, B. J.; Vinayak, S.; Avril, N. J. Nucl. Med. 2016, 57 (Suppl 1), 34S. (7) Bang, J. I.; Lim, Y.; Paeng, J. C.; Han, S. W.; Park, S.; Lee, J. M.; Kim, H. J.; Cheon, G. J.; Lee, D. S.; Chung, J. K.; KIm, T. Y.; Kang, K. W. Nucl. Med. Mol. Imaging 2017, 51, 147. (8) Ueda, S.; Saeki, T.; Takeuchi, H.; Shigekawa, T.; Matsuura, K.; Nakamiya, N.; Sano, H.; Shimada, H.; Hirokawa, E.; Osaki, A. J. Breast Cancer. 2013, 16, 342−344. (9) Rosenbaum, S. J.; Lind, T.; Antoch, G.; Bockisch, A. Eur. Radiol. 2006, 16, 1054. (10) Thorpe, P. E. Clin. Cancer Res. 2004, 10, 415−427. (11) Ni, Y.; Chen, F.; Mulier, S.; Sun, X.; Yu, J.; Landuyt, W.; Marchal, G.; Verbruggen, A. Eur. Radiol. 2006, 16, 1031−1040. (12) Smith, B. A.; Smith, B. D. Bioconjugate Chem. 2012, 23, 1989− 2006. (13) Rescifina, A.; Zagni, C.; Varrica, M. G.; Pistarà, V.; Corsaro, A. Eur. J. Med. Chem. 2014, 74, 95−115.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Fax: +86-25-85637817. *E-mail: [email protected]. Fax: +86-25-85637817. ORCID

Dongjian Zhang: 0000-0003-4821-6620 Qiaomei Jin: 0000-0002-8559-9125 13255

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256

Article

Analytical Chemistry (14) Malik, E. M.; Müller, C. E. Med. Res. Rev. 2016, 36, 705−748. (15) Wang, Q.; Yang, S.; Jiang, C.; Li, J.; Wang, C.; Chen, L.; Jin, Q.; Song, S.; Feng, Y.; Ni, Y. Sci. Rep. 2016, 6, 21341. (16) Luo, Q.; Jin, Q.; Su, C.; Zhang, D.; Jiang, C.; Fish, A. F.; Feng, Y.; Ni, Y.; Zhang, J.; Yin, Z. Anal. Chem. 2017, 89, 1260. (17) Ji, A. Y.; Jin, Q. M.; Zhang, D. J.; Zhu, H.; Su, C.; Duan, X. H.; Bian, L.; Sun, Z. P.; Zhang, J.; Yang, Z.; Yin, Z. Q.; Ni, Y. C. ACS Med. Chem. Lett. 2017, 8, 191−195. (18) Zeng, J. F.; Jing, L. H.; Hou, Y.; Jiao, M. X.; Qiao, R. R.; Jia, Q. J.; Liu, C. Y.; Fang, F.; Lei, H.; Gao, M. Y. Adv. Mater. 2014, 26, 2694−2698. (19) Gao, Z. Y.; Hou, Y.; Zeng, J. F.; Chen, L.; Liu, C. Y.; Yang, W. S.; Gao, M. Y. Adv. Mater. 2017, 29, 1701095. (20) Chen, L.; Chen, J. Y.; Qiu, S. S.; Wen, L.; Wu, Y.; Hou, Y.; Wang, Y.; Zeng, J. F.; Feng, Y.; Li, Z.; Shan, H.; Gao, M. Y. Small 2018, 14, 1702700. (21) Perek, N.; Sabido, O.; Le Jeune, N.; Prevot, N.; Vergnon, J. M.; Clotagatide, A.; Dubois, F. Eur. J. Nucl. Med. Mol. Imaging 2008, 35, 1290−1298. (22) Zhang, D.; Huang, D.; Ji, Y.; Jiang, C.; Li, Y.; Gao, M.; Yao, N.; Liu, X.; Shao, H.; Jing, S.; Ni, Y.; Yin, Z.; Zhang, J. J. Drug Target. 2015, 23, 180−90. (23) Dave, H.; Ledwani, L. Indian J. Nat. Prod Resour. 2012, 3, 291− 319. (24) Lauffer, R. B. Chem. Rev. 1987, 87, 901−927. (25) Powell, D. H.; Dhubhghaill, O. M. N.; Pubanz, D.; Helm, L.; Lebedev, Y. S.; Schlaepfer, W.; Merbach, A. E. J. Am. Chem. Soc. 1996, 118, 9333−9346. (26) Gu, S.; Kim, H. K.; Lee, G. H.; Kang, B. S.; Chang, Y.; Kim, T. J. J. Med. Chem. 2011, 54, 143−152. (27) Gibby, W. A.; Gibby, K. A.; Gibby, W. A. Invest. Radiol. 2004, 39, 138−142. (28) Lauffer, R. B. Magn Reson Q. 1990, 6, 65−84. (29) Hoebers, F. J. P.; Kartachova, M.; Bois, J. D.; Brekel, M. W. M. V. D.; Tinteren, H. V.; Herk, M. V.; Rasch, C. R. N.; Olmos, R. A. V.; Verheij, M. Eur. J. Nucl. Med. Mol. Imaging 2008, 35, 509−518. (30) Belhocine, T.; Steinmetz, N.; Hustinx, R.; Bartsch, P.; Jerusalem, G.; Seidel, L.; Rigo, P.; Green, A. Clin. Cancer Res. 2002, 8, 2766−2774. (31) van Genderen, H. O.; Kenis, H.; Hofstra, L.; Narula, J.; Reutelingsperger, C. P. Biochim. Biophys. Acta, Mol. Cell Res. 2008, 1783, 953−963. (32) Ran, S.; Thorpe, P. E. Int. J. Radiat. Oncol., Biol., Phys. 2002, 54, 1479−1484. (33) Dasari, M.; Lee, S.; Sy, J.; Kim, D.; Lee, S.; Brown, M.; Davis, M.; Murthy, N. Org. Lett. 2010, 12, 3300−3303. (34) Wang, B.; Xu, H.; Gao, Z. Q.; Ning, H. F.; Sun, Y. Q.; Cao, G. W. Acta Radiol. 2008, 49, 523. (35) Lencioni, R.; Crocetti, L. Radiology 2012, 262, 43−58. (36) Miyayama, S.; Yamashiro, M.; Okuda, M.; Yoshie, Y.; Nakashima, Y.; Ikeno, H.; Orito, N.; Matsui, O. Cardiovasc Intervent Radiol. 2010, 33, 513−522. (37) Mason, R. P.; Zhao, D.; Liu, L.; Trawick, M. L.; Pinney, K. G. Integr Biol. (Camb). 2011, 3, 375−387. (38) Siemann, D. W.; Chaplin, D. J.; Walicke, P. A. Expert Opin. Invest. Drugs 2009, 18, 189−197.

13256

DOI: 10.1021/acs.analchem.8b01868 Anal. Chem. 2018, 90, 13249−13256