Preliminary biological evaluation of 18 F-AlF-NOTA-MAL-Cys-Annexin V as a novel apoptosis imaging agent

A novel annexin V derivative (Cys-Annexin V) with a single cysteine residue at its C-terminal has been successfully labeled site-specifically with NOTA-maleimide aluminum [18F]fluoride complexation and evaluated it as a novel apoptosis agent in vitro and in vivo. The total synthesis time of 18F-AlF-NOTA-MAL-Cys-Annexin V from [18F]fluoride was about 65 min. The tracer was stable in vitro and it was excreted through renal in normal mice. The rate of the tracer bound to erythrocytes with exposed phosphatidylserine was 89.36±0.61% and this binding could be blocked by unlabeled Cys-Annexin V. In rats treated with cycloheximide, there were 6.23±0.23 times (n=4) increase in hepatic uptake of the tracer as compared to normal rats at 1h p.i. The uptake of the tracer in liver also could be blocked by co-injection of unlabeled Cys-Annexin V. These results indicated the favorable characterizations such as convenient synthesis and specific apoptotic cells targeting of18F-AlF-NOTA-MALCys-Annexin V were suitable for its further investigation in clinical apoptosis imaging.


INTRODUCTION
Apoptosis, also known as programmed cell death, is an important way to maintain the relative balance of the body, which is also closely related with a variety of pathological processes such as myocardial ischemia and tumor response to treatment [1,2]. Therefore, it is very important to detect and quantify the apoptosis in vivo in order to diagnose and evaluate therapeutic efficacy.
During the early phase of apoptosis, phosphatidylserine (PS) in the lipid bilayer of the cell membrane is flipped from the inner layer to the outer layer and exposed to the cell surface [3,4]. So it is a good target for the development of probe to image apoptotic cells [5]. Annexin V with high-affinity for PS is an endogenous protein with a molecular weight of about 36-kD and contains about 319 amino acids. It belongs to the calcium-dependent phospholipid-binding protein family and always used as PS targeting agent [6,7]. Flow cytometry or fluorescence microscopy examination of apoptosis using fluorescein or biotin-labeled Annexin V as a probe is a sensitive, efficient, mature laboratory testing method. Annexin V labeled with various radionuclides are also useful as radiotracers in vivo imaging of apoptosis as single photon emission computed tomography (SPECT) and positron emission tomography (PET) imaging agents. Annexin V and Annexin V derivatives were radiolabeled with 125 I, 123 I, 111 In and 99m Tc for SPECT imaging of apoptosis [8][9][10][11]. In 1998, 99m Tc-HYNIC-Annexin V was first reported by Blankenberg et al and then it became the most successful and extensively studied radiotracer of SPECT apoptosis imaging. [12]. In a wide array of preclinical [11,13] and clinical studies [14][15][16] 99m Tc-HYNIC-Annexin V imaging was useful to assess tumor response to therapy.

Research Paper
Since PET is more sensitive and quantitative than SPECT, Annexin V has been radiolabeled with positron emission radioisotopes including 124 I [17][18], 68 Ga [19], and 18 F [20][21][22] for PET imaging. Among them 18 F-labeled Annexin V was the most studied radiotracer because of the favorable properties of 18 F, such as moderate half-life of 109.8 min, high image resolution and lower radiation dose [23]. NH 2 group reactive agent N-succinimidy-4-18 F-fluorobenzoate ( 18 F-SFB) was used mostly as prosthetic group to label Annexin V [20,22,24], however the reaction between 18 F-SFB and Annexin V is nonspecific. 18 F-SFB could react with any NH 2 group of Annexin V, whereas there are 23 NH 2 groups available on Annexin V. N-substituted maleimides were used mainly as thiol reactive agents to radiolabel proteins at free thiol of cysteines [25]. We previously evaluated one fluorine-18-labeled analog of Annexin V mutant, Cys-Annexin V, having a C-terminal cysteine, prepared by radiolabeling with 18 F-FBEM [26]. 18 F-FBEM-Cys-Annexin V displayed good liver uptake in rats treated with cycloheximide, however the preparation process of the radiotracer consumed much time. To overcome this difficulty, we developed a new method to radiolabel Cys-Annexin V using aluminum [ 18 F]fluoride ( 18 F-AlF) with a maleimide monoamide NOTA (NOTA-MAL). This method was required less time than synthesis of 18 F-FBEM-Cys-Annexin V. It is important to reduce synthesis time of radiotracer, because of the physical half-life of 18 F is only 109.8 min and also it is helpful to reduce radiation exposure.
The aim of this study was to evaluate 18 F-AlF-NOTA-MAL-Cys-Annexin V as a new apoptotic imaging agent in vitro and in vivo.

In vitro stability of 18 F-AlF-NOTA-MAL-Cys-Annexin V
It is important to study the stability of radiotracer. If radiotracer is not stable, some radioactive decomposed side products will affect the imaging results. The stability of 18 F-AlF-NOTA-MAL-Cys-Annexin V was studied in (A) phosphate buffered saline, (B) human serum and (C) cell culture media, respectively. The results are presented in Figure 2. 18 F-AlF-NOTA-MAL-Cys-Annexin V was stable in PBS, human serum and cell culture media and the radiochemical purities of the tracer were also >95% with HPLC analysis after 180 min. These results suggested that the radiotracer was stable in vitro.

Bioactivity study
To determine the bioactivity of the radiotracer in a cell binding assay, the rate of it bound to erythrocytes was 89.36±0.61% as shown in Figure 3. When tubes were added with 50-fold of Cys-Annexin V, the rate of it bound to erythrocytes was 9.58 ± 1.06%. On the other hand, when the addition with 50-fold of BSA, binding of 18 F-AlF-NOTA-MAL-Cys-Annexin V was 88.84 ± 1.01 %. These values indicated that 18 F-AlF-NOTA-MAL-Cys-Annexin V was specific binding to PS of erythrocytes.

Dynamic MicroPET imaging of mice
After administration of 18 F-AlF-NOTA-MAL-Cys-Annexin V, major organ time-activity curves were obtained from 0 to 180 min dynamic microPET scans. In Figure 4, diamond represents heart, triangle represents kidney and square represents liver. The radioactivity kinetics were calculated from a region-of-interest analysis of the dynamic microPET scans. The radiotracer was excreted mostly through the kidney and the peak (67% ID/g) at about 35 min p.i. and then reduced to 34% ID/g at 180 min p.i.

Apoptotic rat liver imaging
Twelve liver apoptosis rats were induced with 10 mg/kg cycloheximide and divided into three groups, treated group(B), blocking group(C) and BSA group(D). The other four normal rats were served as control group(A). In Figure 5. four representative coronal microPET images displayed of 7.4 MBq 18 F-AlF-NOTA-MAL-Cys-Annexin V at 1h p.i. The uptakes of the radiotracer in liver (arrow) of control group, treated group, blocking group and BSA group were 0.50±0.02, 3.09±0.08, 0.76±0.04 and 3.34±0.09%ID/g, respectively, at 1h p.i. The uptake ratios (treated/control, blocking/ control, BSA/control) of liver were 6.23±0.23, 1.52±0.07, 6.72±0.21(n=4), respectively, at 1h p.i. These values means that the uptake of 18 F-AlF-NOTA-MAL-Cys-Annexin V was increased in the cycloheximide treatment liver and it also could be blocked with unlabeled Cys-Annexin V and BSA could not block the tracer uptake in liver of rats. These results indicated 18 F-AlF-NOTA-MAL-Cys-Annexin V could bind specifically to apoptotic cells.
In Figure 6, four representative images of liver TUNEL staining sections were displayed. There were little apoptotic nuclei (green nuclei) in rats' liver of control group and there were more apoptotic nuclei in that of treated group (B), blocking group (C) and BSA group (D). The rates of TUNEL-positive nuclei were 2.0±0.3%, 15.2±1.5%, 14.8±0.5% and 15.5±1.2% of control group, treated group, blocking group and BSA group, respectively. The liver uptake ratio as measured via microPET at 1 h p.i. between treated group and control group correlated well with the ratio of apoptotic nuclei in liver measured by using TUNEL staining between treated group and control group.

DISCUSSION
Site-specific labeling method is important for providing a chemically homogeneous radioactive conjugate with defined in vivo properties. Since there is no free thiol group provided by cysteine in the Annexin V molecule, we introduce a unique cysteine residue at the desired position and use thiol-mediated chemical labeling or chelator coupling. The Annexin V derivative with a single cysteine residue at its C-terminal (Cys-Annexin V) has been successfully labeled with 18 F-FBEM by this approach [26]. 18 F-FBEM-Cys-Annexin V mainly excreted through the renal pathway in nornal mice and showed high uptake in the rats' liver treated with cycloheximide. Despite the encouraging results for 18 F-FBEM-Cys-Annexin V, the radiosynthesis of [ 18 F]FBEM was time consuming, which has became the impediment of this radiotracer to widespread use.  Compared with DOTA, NOTA is a promising chelator to provide more stable complexes with a number of radiometals such as gallium and indium [27,28]. A refined method has been reported by using a complex composed of NOTA and [ 18 F]aluminum fluoride to label peptides within a short time about 10min [29][30][31]. These results suggested that NOTA is a very promising chelator for radiolabeling of proteins and peptides.
Our procedure for preparation of 18 F-FBEM-Cys-Annexin V required [ 18 F]FBEM, which was synthesized by an semiautomated synthesis device and then purified by HPLC [26]. The synthesis process of 18  Also, the synthesis of 18 F-FBEM required be anhydrous for the fist step, however, that of 18 F-AlF-NOTA-MAL did not require be anhydrous. Therefore, it is easy and convenient to synthesize 18 F-AlF-NOTA-MAL.
The yield of [ 18 F]FBEM was low, less than 10% from [ 18 F]fluoride without decay corrected and the synthesis lasted for about 100 min. And then the yield of [ 18 F]FBEM-Cys-Annexin V was less than 70% from [ 18 F]FBEM and the time of the reaction and purification of 18 F-FBEM-Cys-Annexin V was over 40 min. Thus, the total yield of 18 F-FBEM-Cys-Annexin V was about 5% from [ 18 F]fluoride without decay corrected and consumed over 2 h. In contrast, the total yield of 18 F-AlF-NOTA-MAL-Cys-Annexin V was about 15% from [ 18 F]fluoride without decay corrected. And the preparation time of 18 F-AlF-NOTA-MAL-Cys-Annexin V was just about 65 min, which was less than that of 18 F-FBEM-Cys-Annexin V. It is important to reduce synthesis time and improve radiochemical yield of 18 F site-specific labeling Annexin V derivatives.
Compared to 18 F-FBEM-Cys-Annexin V, 18 F-AlF-NOTA-MAL-Cys-Annexin V showed similar biological properties, with the exception much higher renal metabolism. The peak value of kidney uptake of 18 F-FBEM-Cys-Annexin V was 11%ID/g at 13 min p.i., which was less than that of 18 F-AlF-NOTA-MAL-Cys-Annexin V (67%ID/g) at about 35 min p.i. The increase of kidney uptake of 18 F-AlF-NOTA-MAL-Cys-Annexin V may due to hydrophilicity of 18 F-AlF-NOTA. Some groups also reported higher kidney uptake of 18 F-AlF-NOTA labeling peptide than that of 18 F-FBEM labeling the same peptide [30].
In vitro and in vivo studies also showed that the radiotracer is a promising apoptosis imaging agent. A Waters high-performance liquid chromatography (HPLC) system with a Waters 2998 photodiode array detector (PDA) and a semi-preparative C18 HPLC column (250×10mm, 5um, Chrom-Matrix Bio-Tech) was used for 18 F-AlF-NOTA-MAL purification. The flow rate is 2 mL/min, and the mobile phase changed from 95% solvent A (0.1% trifluoroacetic acid in water) and 5% solvent B (0.1% trifluoroacetic acid in acetonitrile) (0-2 min) to 35% solvent A and 65% solvent B at 32 min. The UV absorbance was monitored at 218 nm, and the UV spectrum was checked with the PDA detector.
Analyzed HPLC was performed on Waters Breeze system with a TSK-GEL column (swG2000SWXL, 300 × 7.8 mm 5 µm, Tosoh Bioscience Co., Ltd, Shanghai, China). The absorbance was measured on the UV detector at 278 nm. Radioanalysis of the labeled compound was conducted using a Cd (Te) detector. The flow rate was adjusted to 1.0 mL/min and the isocratic mobile phase was 0.05 mol/L phosphate buffer (pH =7.0).
A microPET system (Inveon, Siemens Co. German) and a fluorescence microscope (Olympas X51, Tokyo, Japan) were used. The animal experiments in this study were approved by the Animal Care and Ethnics Committee of Jiangsu Institute of Nuclear Medicine.

Radiosynthesis of 18 F-AlF-NOTA-MAL
A 2 mL centrifuge tube was charged with 3µL of a solution of aluminum chloride (2 nM) in 0.5 M NaOAc (pH=4). Cyclotron target water containing [ 18 F]fluoride (up to 100µL containing up to 3700 MBq) was added, followed by 200µg NOTA-MAL mono TFA, mono hexafluorophosphate salt (Chematech, Dijon, France) in 40µL of 0.5 M sodium acetate buffer (pH=4) and 200µL CH 3 CN. The resulting solution was heated at 90~100°C for 10 min. The reaction mixture was then cooled and injected onto a semi-preparative HPLC column. The radioactivity peak eluting at ~12 min was collected. The total synthesis time for 18 F-AlF-NOTA-MAL was about 25 min and 925±23MBq (n=4) radiochemically pure 18 F-AlF-NOTA-MAL was obtained from 14.8±0.5GBq 18 F-fluoride.

Labeling of Cys-Annexin V with 18 F-AlF-NOTA-MAL
The isolated 18 F-AlF-NOTA-MAL (185-555 MBq) in 100µL was added to a solution of Cys-Annexin V (50~100 µg in 100µL, pH=7.2) PBS, and the mixture was allowed to react at room temperature for 15~30 min (Scheme 1) and loaded onto a NAP-5 column (GE Healthcare, Buckinghamshire, UK). The NAP-5 column was eluted with 250 µL portions of PBS. The most concentrated fraction containing the radiolabeled protein (fraction 3, 150~450MBq) was collected and used for the biological experiments.

Bioactivity study
The bioactivity of 18 F-AlF-NOTA-MAL-Cys-Annexin V was determined by its binding to erythrocytes, according to a previously reported procedure [32,33]. In brief, 18 F-AlF-NOTA-MAL-Cys-Annexin V at 10nmol/L final concentration was added to four tubes containing a final volume of 1 mL of buffer HNKGB (10mM HEPES-Na, pH 7.4, 136mM NaCl, 2.7 mM KCl, 5mM glucose, and 1mg/mL BSA) plus 2.5 mmol/L CaCl 2 . One tube then received 4.2×10 8 erythrocytes with exposed phosphatidylserine in 100µL. The control tube then received an equal volume of buffer, and other two tubes then received 4.2×10 8 erythrocytes with exposed phosphatidylserine and 50-fold of unlabeled Cys-Annexin V and 50-fold of BSA respectively in order to saturate and block any specific binding. Samples were incubated for 15 min at room temperature. After centrifugation at 8,320 g for 3 min, the radioactivity of the supernatant was measured with a packard-multi-prias gamma counter. The binding ratios were determined as follows: Radioactivity bound to erythrocytes (%)=(1-[radioactivity of supernatant in the presence of erythrocytes]/[radioactivity of supernatant in absence of erythrocytes]) ×100. All experiments were performed three times.

Dynamic MicroPET imaging of mice
Four male ICR mice (25±2 g) were anesthetized with 1%-2% isoflurane, positioned prone, immobilized, and were injected via the tail vein with 0.2mL 3.7 MBq (100µCi) 18 F-AlF-NOTA-Cys-Annexin V and imaged dynamically for 3h. The images were reconstructed using a two dimensional ordered-subset expectation maximization (2D OSEM) algorithm without correction for attenuation or scattering. For each scan, regions of interest (ROIs) were drawn over the liver and major organs using vendor software (ASI Pro 5.2.4.0) on decay-corrected wholebody coronal images. The radioactivity concentrations (accumulation) within the liver, heart and kidneys were obtained from mean pixel values within the multiple ROI volume and then converted to megabecquerel per milliliter per minute using the calibration factor determined for the Inveon PET system. These values were then divided by the administered activity to obtain (assuming a tissue density of 1 g/ml) an image-ROI-derived percent injected dose per gram (%ID/g).

MicroPET images of rat model of apoptosis
Twelve male SD rats (258 ± 3g) were treated IV with 10 mg/kg cycloheximide to induce liver apoptosis and then were divided into three groups as treated group, blocking group and BSA-group. Other four male SD rats (259 ± 2g ) were treated IV with saline as the control group. 3 h after the treatment, rats of treated and control group were anesthetized with 1%-2% isoflurane and were injected via the tail vein with 0.2 mL (7.4 MBq, 200 µCi) 18 F-AlF-NOTA-MAL-Cys-Annexin V. Rats of blocking and BSA group were coinjected with 7.4 MBq 18 F-AlF-NOTA-MAL-Cys-Annexin V and blocking dose (5mg/ kg body weight) of unlabeled Cys-Annexin V or bovine serum albumin (BSA), respectively. Ten-minute static scans were acquired at 1h after injection with a MicroPET (Inveon, Siemens), respectively. Immediately after MicroPET imaging, the livers were dissected. Then, using the livers, formalin-fixed paraffin-embedded specimens were prepared for Terminal deoxynucleotidyl transferasemediated nick end labeling (TUNEL) staining.

TUNEL staining
Because our imaging studies were designed to determine the uptake and biodistribution of 18 F-AlF-NOTA-MAL-Cys-Annexin V after chemically induced apoptosis, it was important to confirm apoptosis in the livers of treated rats by independent methods that provide quantitative results. A marker of apoptosis was scored by performing a TUNEL assay that measures DNA fragmentation, a characteristic feature of apoptosis. Terminal deoxynucleotide transferase adds labeled nucleotides to the 3′ termini at double-stranded breaks in the fragmented DNA. TUNEL assays were performed according to the manufacturer's instructions, using the fluorescein-conjugated colorimetric TUNEL apoptosis assay kit (Beyotime Institute of Biotechnology, Shanghai, China). Briefly, slices were freed of paraffin through xylene and graded EtOH washes and then incubated with proteinase K (Beyotime Institute of Biotechnology) (2 mg/mL in 10 mmol/L Tris, pH 8.0).
After proteinase digestion, the slides were equilibrated in pH 7.4 buffer, the terminal deoxynucleotide transferase enzyme and Biotin-dUTP labeling mix (Beyotime Institute of Biotechnology) were added, and the slides were incubated at 37 °C for 1 h in a humid chamber. The number of TUNEL-positive cells was counted on 10 randomly selected ×100 fields for each section by use of a Olympus fluorescence microscope.

Statistical analysis
Quantitative data are expressed as mean ± SD. Means were compared using one-way analysis of variance (ANOVA) and Student's t test. P values <0.05 were considered statistically significant.

CONCLUSIONS
Cys-Annexin V was successfully labeled with 18 F via conjugated with 18 F-AlF-NOTA-Mal, which is relative ease to radiochemical synthesis compared to 18 F-FBEM. 18 F-AlF-NOTA-MAL-Cys-Annexin V showed promising characterizations in vitro and in vivo for apoptosis imaging.

Author contributions
Huixin Yu and Zichun Hua conceived and designed the experiments; Chunxiong Lu, Quanfu Jiang and Cheng Tan performed the experiments; Chunxiong Lu and Huixin Yu analyzed the data; Zichun Hua and Minjin Hu contributed Cys-Annexin V; Chunxiong Lu wrote the paper.