Development of 99mTc-N4-NIM for Molecular Imaging of Tumor Hypoxia

The nitro group of 2-nitroimidazole (NIM) enters the tumor cells and is bioreductively activated and fixed in the hypoxia cells. 1,4,8,11-tetraazacyclotetradecane (N4) has shown to be a stable chelator for 99mTc. The present study was aimed to develop 99mTc-cyclam-2-nitroimidazole (99mTc-N4-NIM) for tumor hypoxia imaging. N4-NIM precursor was synthesized by reacting N4-oxalate and 1,3-dibromopropane-NIM, yielded 14% (total synthesis). Cell uptake of 99mTc-N4-NIM and 99mTc-N4 was obtained in 13762 rat mammary tumor cells and mesothelioma cells in 6-well plates. Tissue distribution of 99mTc-N4-NIM was evaluated in breast-tumor-bearing rats at 0.5–4 hrs. Tumor oxygen tension was measured using an oxygen probe. Planar imaging was performed in the tumor-bearing rat and rabbit models. Radiochemical purity of 99mTc-N4-NIM was >96% by HPLC. Cell uptake of 99mTc-N4-NIM was higher than 99mTc-N4 in both cell lines. Biodistribution of 99mTc-N4-NIM showed increased tumor-to-blood and tumor-to-muscle count density ratios as a function of time. Oxygen tension in tumor tissue was 6–10 mmHg compared to 40–50 mmHg in normal muscle tissue. Planar imaging studies confirmed that the tumors could be visualized clearly with 99mTc-N4-NIM in animal models. Efficient synthesis of N4-NIM was achieved. 99mTc-N4-NIM is a novel hypoxic probe and may be useful in evaluating cancer therapy.


Introduction
Recent studies demonstrated that the hypoxic environment induces more malignant neoplastic phenotypes [1]. Disruption of oxygen delivery to tumors could diminish apoptotic potential and increase the chemotherapy/radiation resistance, while an improvement in oxygen delivery to tumors increases tumor sensitivity to radiation and chemotherapy [2][3][4]. Due to its significant prognostic and therapeutic implication, efforts have been made to invent efficient noninvasive methods to assess the presence and extent of tumor hypoxia because information on the hypoxic fraction in tumors could aid to reveal the mechanisms of aggressive behavior. The success of the endeavor to noninvasively detect the hypoxic fraction of tumor by nuclear molecular imaging such as single-photon emission computed tomography (SPECT) allows physicians to select patients for additional or alternative treatment regimens that may circumvent or overcome the ominous impact of tumor hypoxia and improve disease control [5].
The nitro group of nitroimidazole is enzymatically reduced by ribonucleoside reductase within viable hypoxic cells. This mechanism is well understood through numerous in vitro and in vivo studies from the past three decades [6]. In aerobic cells the reduced nitroimidazole is immediately reoxidized and washed out rapidly. On the contrary, in cells with low oxygen concentration the reoxidation is slowed, which allows further reductive reactions to take place. This leads to the formation of reactive products that could covalently bind to cell components and thus diffuse more slowly out of the tissue in an oxygen-dependent manner [7]. 18 F-Fluoromisonidazole ( 18 F-FMISO) and 18 F-fluoroerythronitroimidazole ( 18 F-FETNIM), 2-nitroimidazole analogues, have been used with PET to evaluate tumor hypoxia, but the chemistries are dependent on a cyclotron to produce 18 F in addition to slow serum clearances [8][9][10].
In an attempt to identify efficient and clinically userfriendly chelator-based hypoxia tracers, our team has developed a new class of 99m Tc hypoxia SPECT tracers based on the nitroimidazole backbone. The nitrogen, oxygen, and sulfur combination has shown to be stable chelators for radiometals [11][12][13][14][15][16][17][18]. The chelator used in this paper is 1,4,8,11tetraazabicyclohexadecane (N4) which has shown to be a stable chelator for 99m Tc [11,12]. In addition, N4-chemistry could be easily modified and selectively reacted with halogenated homing compounds. In this paper, the potential use of 99m Tc-N4-NIM as tumor hypoxic imaging agent was evaluated. The advantage of 99m Tc-N4-nitroimidazole over 18 F-FMISO and 18 F-FETNIM is its easier chemistry and its water solubility that may lead to the rapid clearance of unbounded tracers after intravenous injection.

Chemicals and Analysis.
All chemicals of analytical grade and solvents of high performance liquid chromatography (HPLC) grade were obtained from Sigma-Aldrich (St. Louis, MO, USA). Nuclear magnetic resonance (NMR) was performed on a Bruker 300-MHz spectrometer (Bruker BioSpin Corporation, Billerica, MA, USA), and mass spectra were performed on a Waters Q-TOF Ultima mass spectrometer (Waters, Milford, MA, USA) at the chemistry core facility at The University of Texas MD Anderson Cancer Center (Houston, TX, USA). Sodium pertechnetate (Na 99m TcO 4 ) was obtained from a 99 Mo/ 99m Tc generator (Covidien, Mansfield, MA, USA).

Sodium Salt of 2-Nitroimidazole (Compound 1).
The synthetic scheme of N4-NIM is shown in Figure 1. One molar equivalent of NaOH 1 M (0.3864 g, 9.66 mmol, 9.66 mL) was added to 2-nitroimidazole (1.09 g, 9.66 mmol) and warmed 30 minutes at 50 • C to dissolve it. If the compound was not dissolved in the solution then NaOH was added drop by drop until the solid was dissolved completely and we continue heating for 15 minutes more. Water was removed on rotary evaporator. Crude compound was dissolved in minimum quantity of water, filtered, and lyophilized. Yield: 5.31 g, (90%).

Bromopropane Nitroimidazole (Compound 2).
In twoneck flask one molar equivalent of Sodium salt of 2-nitroimidazole (2.00 g, 14.7 mmol) in solid form was added in 40 mL of acetonitrile (anhydrous) to dissolve it, then 18crown-6 (3.88 g, 14.7 mmol) was added to this mixture. In nitrogen atmosphere 12.5 molar equivalent of 1,3-dibromopropane (18.65 mL, 183.76 mmol, 37.09 g) was added to the reaction mixture. The reaction mixture was then refluxed in a nitrogen environment at 50 • C overnight. The reaction was filtered to purify the precipitate of NaBr using the 0.22 μM filter paper. Soon after the solvent was evaporated, the crude compound was purified by column chromatography with a CHCl 3 : CH 3  259 mmol) was placed into the two-neck flask and dissolved in 25 mL of DMF (anhydrous). While heating the solution at 50 • C, bromonitroimidazole (0.5329 g, 2.259 mmol) dissolved in 2.0 mL of DMF (anhydrous) was added to the reaction mixture. After adding K 2 CO 3 (anhydrous)(0.7805 g; 5.6475 mmol) in 2.5 molar equivalent solid to the reaction mixture, it was refluxed at 70 • C overnight. Crude compound was filtered using 0.22 μM filter paper and checked by TLC in chloroform: methanol in an 8 : 2 solvent system. Solvent was then evaporated and crude compound was purified by column chromatography in (8 : 2) of chloroform and methanol. Yield: 6.96 g, (75.86%). Ms (m/z) 408.30 [M] + .

Radiolabeling of N4-NIM with 99m
Tc. Radiosynthesis of 99m Tc-N4-NIM was achieved by adding a required amount of 99m Tc-pertechnetate into a kit containing the lyophilized residue of N4-NIM (1 mg) and SnCl 2 (100 μg). Final pH of the preparation was 5.5-7.4. Radiochemical purity was determined by TLC and HPLC. Radio-TLC (Waterman No. 1) was obtained by eluting 99m Tc-N4-NIM with acetone and saline, respectively. Radio-HPLC (Waters) was obtained by eluting 99m Tc-N4-NIM on a C-18 reverse phase column After incubation, cells were washed with ice-cold phosphatebuffered solution twice and detached using a treatment of 0.5 mL of trypsin for 5 minutes. Cells were then collected and the radioactivity of the cells was measured in triplicate with a gamma counter. Radioactivity was expressed as mean ± standard deviation percent of cellular uptake (%Uptake).  Studies were performed 14 to 17 days after implantation when tumors reached approximately 1 cm diameter. The rats were divided into 3 groups and each rat was injected intravenously with 25 ± 0.5 μCi of 99m Tc-N4-NIM and 99m Tc-N4. Each group was examined at 1 of 3 time points (0.5, 2, or 4 hours after injection). At each time point, the rats were killed and the selected tissues were excised, weighed, and measured for radioactivity by gamma counter. For each sample, radioactivity was expressed as mean percentage of the injected dose per gram of tissue wet weight (%ID/g). Counts from a 1/10 diluted sample of the original injection were used as a reference.
Tumor/nontarget tissue count density ratios were calculated from the corresponding %ID/g values. Olinda software (version 1.1; Vanderbilt University, Nashville, TN, USA) was used to determine dosimetry, based upon preclinical source organ residence time estimates as followed: rat organ distribution data was processed using in-house software to determine residence times (τ) based on AUC. The data was then converted to human residence time estimates using the correction factor for each organ, and subsequent τ values were entered into Olinda software to generate human dose estimates. were performed using the Eppendorf computerized histographic system. Twenty to twenty-five pO 2 measurements along each of two to three linear tracks were performed at 0.4 mm intervals on each tumor (40-75 measurements total). Tumor pO 2 measurements were made on three tumorbearing rats. Using an on-line computer system, the pO 2 measurements of each track were expressed as absolute values relative to the location of the measuring point along the track and as the relative frequencies within a pO 2 histogram between 0 and 100 mmHg with a class width of 2.5 mm.

Planar Scintigraphic Imaging in Tumor-Bearing Models.
Two animal models were created. For Rabbit model, the VX-2 tumor mass was inoculated (im) to the thigh region of male New Zealand white rabbits (2 kg

Chemistry and Radiochemistry.
The synthetic scheme is shown in Figure 1. The total synthesis yield of precursor N4-NIM via our 5-step procedure was 14%. The structure and purity of N4-NIM were confirmed by 1 H-and 13 (Figure 2). HPLC analysis showed the retention time for 99m Tc-N4-NIM was 3.663 min (UV-254 nm), 3.650 min (UV-210 nm), and 4.200 min (NaI detector). Because 99m Tc-N4-NIM is a kit product and labeled without any further purification, its radiochemical yield was assumed to be identical to its radio-chemical purity.

In Vitro Cellular Uptake Assays.
The cellular uptake kinetics of 99m Tc-N4-NIM and 99m Tc-N4 in rat mammary tumor cells and rat mesothelioma cells are shown in Figure 3.
The uptake for 99m Tc-N4-NIM increased dramatically up to 240 minutes, but this was not the case for the 99m Tc-N4, suggesting that 99m Tc-N4-NIM can enter tumor cells specifically and accumulate rapidly.

Biodistribution and Radiation Dosimetry Estimates of 99m
Tc-N4-NIM. Tissue distribution of 99m Tc-N4 and 99m Tc-N4-NIM is shown in Tables 1 and 2. Biodistribution studies showed that tumor/blood and tumor/muscle count density ratios at 0.5-4 hr gradually increased for 99m Tc-N4-NIM (Table 2). No significant uptake in thyroid and stomach suggests in vivo stability of 99m Tc-N4-NIM. The optimal tumor imaging time for 99m Tc-N4-NIM is at 2 hr postadministration of 99m Tc-N4-NIM. Although tumor/blood and tumor/muscle count density ratios at 0.5-4 hr gradually increased for 99m Tc-N4, yet there was almost no tumor uptake (Table 1). Based upon preclinical studies, dosimetry of 99m Tc-N4-NIM was estimated from MIRDose. It is safe to use 99m Tc-N4-NIM in human because the whole body, liver, and effective dose equivalent for the proposed single dose at 20 mCi of 99m Tc-N4-NIM were less than the limits for 3 rem annual and 5 rem total dose equivalent, and other organs of single dose at 5 rem annual and total dose equivalent at 15 rem if the subject did not have any other radiation exposure (Table 3).

Discussion
The development of new tumor hypoxia agents is clinically desirable for detecting primary and metastatic lesions as well as predicting radioresponsiveness and time to recurrence [10,19]. None of the contemporary imaging modalities accurately measures hypoxia since the diagnosis of tumor hypoxia requires a pathologic examination. It is often difficult to predict the outcome of a therapy for hypoxic tumor without knowing at least the baseline of hypoxia in each tumor treated. Although the Eppendorf polarographic oxygen microelectrode can measure the oxygen tension in a tumor, this technique is invasive and needs a skillful operator. Additionally, this technique can only be used on accessible tumors (e.g., head and neck, cervical) and multiple readings are needed. Therefore, an accurate and easy method of measuring tumor hypoxia will be useful for patient selection. However, tumor-to-normal tissue uptake ratios vary and depend upon the radiopharmaceuticals used. Therefore, it would be rational to correlate tumor-to-normal tissue uptake ratios with the gold standard Eppendorf electrode measures of hypoxia when new radiopharmaceuticals are introduced to clinical practice. In biodistribution, 99m Tc-N4-NIM in tumor tissue was decreased as same as 99m Tc-N4 (Tables 1 and 2). This decreased uptake might be due to slower uptake 99m Tc-N4 and 99m Tc-N4-NIM as a function of increased renal excretion. However, the tumor uptake as well as tumor/blood and tumor/muscle ratios in 99m Tc-N4-NIM were higher than that in 99m Tc-N4 group.
Hypoxia-Inducible Factor (HIF)-1α/β heterodimer is a master transcription factor for several genes involved in angiogenesis, glycolysis, pH balance, and metastasis. These HIF-1 target genes help tumors to overcome forthcoming metabolic obstacles as they grow. Under normoxic condition, the HIF-1α subunit is hydroxylated by its specific prolyl-4 hydroxylase 2, given the acronym PHD2, thus stabilizing it under normoxic conditions [20]. In vitro cellular uptake assay, the uptake of 99m Tc-N4-NIM in tumor cells increases up to 240 min. However, this assay was performed under normoxic condition. The increased uptake of 99m Tc-N4-NIM in tumor cells under normoxic condition might be due to stabilized HIF-1α. NaTcO 4 was reduced to +5 Tc [O] and bound to the three nitrogens of cyclam. The charge of 99m Tc-N4-NIM is neutral. Cell uptake of 99m Tc-N4-NIM was via passive diffusion. In our animal model, tumor oxygen tension was determined to be 3.2 to 6.0 mmHg, whereas normal muscle tissue had 30 to 40 mmHg. Although another factor such as anemia may have influenced the level of tumor hypoxia, there was no attempt in identifying this factor. In biodistribution, 99m Tc-N4-NIM in tumor tissue was decreased as same as 99m Tc-N4 (Tables 1 and 2). This decreased uptake might be due to slower uptake of 99m Tc-N4 and 99m Tc-N4-NIM as a function of increased renal excretion. However, the tumor uptake as well as tumor/blood and tumor/muscle ratios in 99m Tc-N4-NIM was higher than those in 99m Tc-N4 group.
Due to better imaging characteristics and lower cost, attempts are made to replace the 123 I-, 131 I-, 67 Ga-, and 111 Inlabeled compounds with corresponding 99m Tc-labeled compounds when possible. Our radiochemistry data indicated N4-NIM could be labeled with 99m Tc very easily and efficiently at room temperature with high radiochemical purity.  Figure 4: Planar scintigraphy of 99m Tc-N4 and 99m Tc-N4-NIM (400 μCi/rat, iv, acquired 500,000 count) showed higher tumor-to-muscle count density ratio in 99m Tc-N4-NIM compared to that of 99m Tc-N4. The numbers are tumor-to-muscle count density ratios (counts/pixel) at 60-120 min.  Figure 5: Planar scintigraphy of 99m Tc-N4-NIM (15 mCi/rabbit, iv, acquired 500,000 count) showed higher tumor-to-muscle count density ratio (counts/pixel) at 60 and 120 min.
In vivo tissue distribution studies showed that radiation dosimetry of blood-forming organs was within radiation dose limits. Our planar imaging studies indicate that 99m Tc-N4-NIM is feasible to assess tumor hypoxia.
In summary, N4-NIM kits could be labeled with 99m Tc easily and efficiently, with high radiochemical purity and cost-effectiveness. In vitro cellular uptake and scintigraphic imaging studies demonstrated the pharmacokinetic distribution and feasibility of using 99m Tc-N4-NIM for tumor hypoxia imaging.