Introduction
Cancer is the leading cause of mortality in economically developed countries, and the second leading cause of mortality in developing countries (1). The global burden of cancer continues to significantly increase due to the aging and growth of the global population, and increasing adoption of cancer-causing behaviors, particularly smoking, within economically developing countries (2). Early diagnosis of cancer is key; however; the majority of malignancies are diagnosed and treated at the advanced stages, which leads to poor overall survival. Researchers and clinicians have increased their focus towards the specific molecular markers targeting tumorigenesis and metastasis.
It is well established that solid tumor growth is correlated with angiogenesis (3). The formation of new blood vessels, which supply metabolites to aid the survival and metastasis of tumor cells, is achieved through angiogenesis, a multi-step process that relies on the tumor-driven production of proangiogenic factors. In contrast to normal blood vessels, tumor vasculature exhibits an abnormal, incomplete endothelial lining that is characterized by large inter-endothelial junctions and an increased number of fenestrations (4).
In our previous study (5), we modified the structure of arginine-arginine-leucine (RRL) to tyrosine-RRL (tRRL; Tyr-Cys-Gly-Gly-Arg-Arg-Leu-Gly-Gly-Cys), and successfully radioiodinated it by the chloramine-T (CH-T) method. Simple biodistribution and single-photon emission computed tomography (SPECT) imaging results indicated that 131I-tRRL was specifically concentrated in the tumor tissue in human PC3 prostate tumor-bearing xenografts (5). Additionally, by further experiments, Lu et al confirmed the cytotoxicity of 131I-tRRL and revealed that tRRL adhered to tumor cells adjacent to tumor-derived endothelial cells, by an in vitro binding experiment and cellular uptake tests (6).
In the present study, we further explored the characteristics of 131I-tRRL and estimated the radiation dosimetry of 131I-RRL for humans using mice data.
Materials and methods
Synthesis of the peptide
The sequence of tRRL (Tyr-Cys-Gly-Gly-Arg-Arg-Leu-Gly-Gly-Cys-NH2) was designed and synthesized as described in a previous study (7). The cysteine at the C-terminal of the tRRL peptide was amidated to protect the peptide from the biocatalysts.
Radiosynthesis
Radioiodination of tRRL was performed using the CH-T method (8). The tRRL (50 μg) was dissolved in 0.5 M phosphate buffer (81 μl, pH 7.4), and 131INa (10 μl, 74 MBq) was added. Subsequently, fresh CH-T (9 μl, 10 μg/μl) was added to the mixture. The mixture was gently agitated for 1 min at room temperature, then quenched by the addition of sodium metabisulfite (45 μl, 1 μg/μl).
For routine quality control of labeling, the labeling efficiency and radiochemical purity of radiolabeled tRRL probes were calculated by paper chromatography on Xinhua filter paper (Hangzhou Xinhua Paper Industry Co., Ltd, China) with acetone and water as the mobile phase.
B16 cell culture
B16 mouse melanoma cell lines were a gift from the Department of Pathology, Peking University First Hospital (Beijing, China), and were maintained in the media recommended by the department. B16 cells were grown in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) and 100 mg/ml penicillin-streptomycin (Gibco, Carlsbad, CA, USA). Cells were cultivated under the standard conditions of a temperature of 37°C and a humidified atmosphere of 5% CO2. Cells were experimented with and harvested by trypsin treatment (0.25% trysin/0.02% ethylenediaminestetraacetic acid; 3 min, 37°C) between passages 4 and 12. The cell growth status was monitored by inverted microscopy with a phase contrast microscope (CKX31 inverted microscope, Olympus, Tokyo, Japan).
Biodistribution
Animal experiments were approved by the Peking University Animal Studies Committee, according to the Guidelines for the Care and Use of Research Animals (Peking University, Beijing, China; approval ID, J201138). Forty male Kunming mice (weight, 20±3 g; age, 4–6 weeks)obtained from the Department of Laboratory Animal Science, Peking University First Hospital were used in this study. The mice were inoculated with 1×107 B16 cells in the right upper limbs, and the tumors were allowed to grow to a diameter of 0.8–1.0 cm. The mice were maintained using a standard diet, bedding and environment, with free access to food and drinking water. The mice received 400 mg sodium perchlorate over 3 days for thyroid blockage.
The 40 Kunming mice were randomly divided into 8 groups of 5 mice each. The purified and isolated 131I-tRRL was diluted with 0.5 M phosphate buffer (100 μl, pH 7.4). 131I-tRRL was then injected into each mouse via the lateral tail vein. All injections were tolerated well. The animals were sacrificed by cervical dislocation at 15 and 30 min, and 1, 3, 6, 24, 48 and 72 h, following injection of the radiolabeled derivatives.
Subsequently, the mice were dissected and the tissues of interest (the blood, heart, liver, spleen, lungs, kidneys, stomach, small intestine, bladder, bone, skeletal muscle and tumor tissue) were weighed, and their radioactivities were measured using a γ-well counter along with a standard of the injection. Radioactivity results were recorded as the percentage of injected radioactivity per gram of tissue (%ID/g) corrected for background and decay.
<sup>131</sup>I-tRRL SPECT imaging in the B16 xenograft model
Four Kunming mice with B16 xenografts were used in the SPECT imaging protocol. Whole-body imaging was performed at 1, 3, 6, 24, 48 and 72 h following injection of 131I-tRRL at the Department of Nuclear Medicine, Peking University First Hospital, using SPECT imaging (SPR SPECT; GE Healthcare, Inc., Milwaukee, WI, USA) equipped with a high-energy general purpose collimator (HEGP). Whole body static images (200,000 counts) were obtained with a zoom factor of 2.0, and were digitally stored in a 256×256 matrix.
Radiation-absorbed dose calculations
The calculated mean percentages of injected activity per gram (%ID/g) for the mice organs were extrapolated to uptake in the organs of a 70-kg adult, using the standard medical internal radiation dose (MIRD) scheme along with the following formula (9,10):
[
A
i
(
t
)
/
A
0
]
h
u
m
a
n
=
[
A
i
(
t
)
/
A
0
]
m
i
c
e
×
[
m
t
o
t
a
l
b
o
d
y
/
m
i
]
m
i
c
e
×
[
m
i
/
m
t
o
t
a
l
b
o
d
y
]
h
u
m
a
n
where [Ai(t)/A0] is the %ID/g of source organ i administered at time t; At is the radioactivity of source organ i administered at time t; A0 is the initial radioactivity administered at time 0; mtotal body is the whole body mass and mi is the mass of source organ i.
The extrapolated values (%ID/g) in the human organs at 15 and 30 min, and 1, 3, 6, 24, 48 and 72 h, were plotted as time-activity curves (TACs). The curves were then fitted with exponential models and integrated to obtain the number of disintegrations in the source organs, using OriginPro 5.0 software (OriginLab Corporation, Northampton, MA, USA). Residence times (τ) were computed and normalized to the injected activities by calculating the area under the TACs of each organ, as follows:
τ
i
=
∫
0
∞
A
i
(
t
)
A
0
d
t
Where τi is the residence time of source organ i; Ai(t) is the radioactivity of source organ i administered at time t; and A0 is the initial radioactivity administered at time 0. Then, using these residence times, organ-absorbed doses and effective doses were estimated with the MIRDOSE 3.1 software package (11).
Statistical analysis
All statistical analyses were performed using SPSS software (version 17.0; SPSS Inc., Chicago, IL, USA). Biodistribution data were expressed as the mean ± standard deviation, and a one-way ANOVA analysis was performed. OriginPro 5.0 and MIRDOSE 3.1 software were used in the radiation dose estimations. P<0.05 was considered to indicate a statistically significant difference.
Results
Radiosynthesis
The radiolabeling efficiency of 131I-tRRL was 70±2.91% (n=5). Radiochemical purities of >95% were obtained following purification.
Biodistribution test
Biodistribution data are shown in Table I. At different time phases following injection, 131I-tRRL primarily accumulated in the kidneys and the bladder. The blood data revealed that 131I-tRRL was characterized by rapid blood clearance, with 6.29%ID/g remaining 15 min following administration, and 2.41%ID/g remaining after 3 h. The specific uptake of 131I-tRRL in the tumor increased after 15 min, and remained at a relatively high level until 6 h following administration. Furthermore, at 24, 48 and 72 h, the tumor tissue demonstrated a higher concentration of the probe compared with the other organs (P<0.05; Fig. 1).
As a result, the ratio of tumor-to-non-tumor (T/NT) tissue accumulation following the administration of 131I-tRRL was significantly higher, particularly between 6 and 24 h. The ratio of tumor-to-skeletal muscle (T/SM) tissue exceeded 4.75 at the 24-h time point, and the ratio of tumor-to-blood (T/B) tissue peaked at 3.36 at 72 h (Fig. 2).
<sup>131</sup>I-tRRL SPECT imaging in the B16 xenograft model
In mice-bearing B16 xenografts, the tumors were imaged 3 h following the administration of 131I-tRRL. Uptake of the 131I-tRRL molecular probe gradually increased over time, and a significant increase (P<0.05) in the uptake in the tumor tissues, calculated using the ROI (region of interest) technique, appeared from 24 h following intravenous injection (Fig. 3). As a result of thyroid blockage, uptake in the thyroid glands was not evident at any time point. As predicted from the biodistribution studies, the renal route of elimination of the probes also led to substantial radioactivity accumulation in the abdomen.
Radiation-absorbed dose calculations
The extrapolated values (%) in the human organs at different time points were plotted as TACs (Fig. 4).
The radiation-absorbed doses were estimated using the MIRDOSE 3.1 software and the mean values obtained are shown in Table II, which lists the individual organ radiation-absorbed doses and the individual effective dose (ED) results of the estimated radiation doses for humans as mGy/MBq and mSv/MBq administered with 131I-tRRL, with the data derived from the mouse data. The ED for the adult male dosimetric model was estimated to be 0.0293 mSv/MBq. Higher absorbed doses were estimated for the stomach (0.102 mGy/MBq), the small intestines (0.0699 mGy/MBq), the kidneys (0.061 mGy/MBq) and the liver (0.055 mGy/MBq). The unit density sphere model was applied to a 2-g tumor, and the estimated absorbed dose was 0.015 mGy/MBq.
Discussion
Tumor angiogenesis is important in the growth and metastasis of malignant tumors. An increasing number of studies are focusing on radiolabeled molecular probes, which target newly formed tumor vessels (12–16).
Our previous in vivo study in nude mice demonstrated that 131I-tRRL had potential for mapping tumor angiogenesis (5). Radiolabeled molecules are important in evaluating tumor characteristics, such as aggressiveness and angiogenesis, and in identifying the effectiveness of cancer treatments, such as chemotherapy and radiotherapy (17).
The tripeptide sequence RRL was previously considered to be a tumor endothelial cell-specific binding sequence by an in vitro bacterial peptide display library panned against tumor cells derived from SCC-VII murine squamous cell carcinomas. The fluorescent RRL studies demonstrated that the peptide preferentially adhered to the tumor vasculature in vivo, and the target was not only tumor-specific but also endothelial in location (18).
In our previous study, we modified the structure of RRL by adding tyrosine to its amino terminal, and termed this tRRL. We then radiolabeled the tRRL with 131I using the CH-T method. The biodistribution data of 131I-tRRL and the in vivo imaging demonstrated prospective application in BALB/c nude mice-bearing PC3 human prostate carcinoma xenografts (5). Further study confirmed the non-cytoxicity of tRRL, although 131I-RRL caused significant cytotoxicity in HepG2 cells. An in vitro binding experiment using fluorescein isothiocyanate (FITC) demonstrated improved adherence between tRRL and different types of tumor cells (6). Lu et al also hypothesized that VEGFR-2 is not the sole biding ligand for tRRL targeting tumor angiogenic endothelium, and that radioiodinated tRRL may be a non-invasive method for functional molecular imaging of tumor angiogenesis (19).
In the present study, we further explored the characteristics of 131I-tRRL and the safety of the new radiotracer. The radiolabeling efficiency and radiochemical purities of 131I-tRRL were similar to what had been previously found (5). Within 1 h following administration of the molecular probe, the probe mainly accumulated in the blood, the kidneys and the bladder. We considered the higher concentration in the kidneys and the bladder to reflect the excretion pathway of 131I-tRRL. The blood data revealed that 131I-tRRL was characterized by rapid blood clearance, with 6.29%ID/g remaining 15 min following administration and 2.41%ID/g remaining after 3 h (i.e., >60% of the probe was metabolized to other organs and tissues). The specific uptake of 131I-tRRL in the tumor tissue increased after 15 min and remained at a relatively high level until 6 h following administration. Furthermore, from 24 to 72 h following administration of the probe, a higher accumulation of the probe was observed in the tumor tissue compared with the other organs (P<0.05), except for the bladder. The hogher ratio of T/NT tissue is also a marker for evaluating a potential oncologic radiotracer. The ratio of T/B and T/SM tissue revealed a good ratio of T/NT tissue (Fig. 2). The highest T/B tissue ratio was 3.36 (at 72 h), and the highest T/SM tissue ratio was 4.76 (at 24 h), following injection of 131I-tRRL.
Arginine-glycine-aspartate (RGD) peptide is a widely studied angiogenesis-specific targeting peptide. The rapid blood clearance and high tumor uptake of 131I-tRRL was similar to the results on cyclic RGD radiolabeled with 99mTc, which is integrin αvβ3-specific and widely used in integrin expression imaging. Zhang et al studied the radiosynthesis and biodistribution of the 131I-RGD dimer in melanoma xenografts. The results demonstrated that the ratios of T/SM and T/B were 4.42 and 2.27 at 24 h, respectively, which were similar to the results of the present study (20).
The results of the SPECT imaging using 131I-tRRL paralleled the biodistribution results. In vivo scintigraphic imaging revealed a higher tumor uptake in the mice-bearing B16 xenografts. From the biodistribution data and the SPECT images, the most suitable acquisition time for tumor imaging is 24 h following administration of 131I-tRRL. The images were clearer than Yu et al and Lu et al reported (5,19).
Continual imaging of the xenografts of mice-bearing B16 xenografts revealed that 131I-tRRL accumulated in the organs of the renal system, the stomach and the small intestine, which partially contributed to its clearance from the circulatory system. This partly explains the reason for the higher radiation doses estimated for the stomach, small intestine and liver.
The safety of a radioligand is determined by its toxicity and radiation. Lu et al confirmed the non-cytotoxicity of tRRL by in vitro experiments. The present study evaluated the absorbed radiation dose of 131I-tRRL in humans (6).
In nuclear medicine, the most commonly used method for the calculation of internal radionuclide dose estimates was developed by the MIRD Committee and most recently summarized in the MIRD Primer (21). The mouse biodistribution data was translated into patient dose estimations, and the organ values were extrapolated from this to a 70 kg adult using the standard MIRD scheme and S tables.
The ED for the adult male dosimetric model was estimated to be 0.0293 mSv/MBq, which is lower than that of 131I-MIP-1145 used in the treatment of stage III and IV melanoma (22). Higher absorbed doses were estimated for the stomach (0.102 mGy/MBq), the small intestines (0.0699 mGy/MBq), the kidneys (0.0611 mGy/MBq) and the liver (0.055 mGy/MBq). Beer et al studied the dosimetry of an RGD probe (18F-Galacto-RGD) clinically studied in humans (23). It was found that the effective radiation dose of the 18F-labeled RGD tracer was similar to that of 18F-FDG, and that the tracer could be safely and effectively used for imaging integrin αvβ3 in humans. The effective dose of 131I-tRRL (2.93 mGy/MBq) was higher than that of 18F-Galacto-RGD (18.7 μGy/MBq). The different dosimetry was most likely to be due to the different isotopic characteristics.
In conclusion, current radiation dosimetry estimates for 131I-tRRL from non-human primates have demonstrated the safety of 131I-tRRL and its potential for use in humans. The results of the present study highlight the potential of 131I-tRRL as a ligand for the SPECT imaging of tumor angiogenesis. Administration of 131I-tRRL led to a reasonable radiation dose burden and was safe for human use.