Synthesis and Evaluation of a Dimeric RGD Peptide as a Preliminary Study for Radiotheranostics with Radiohalogens

We recently developed 125I- and 211At-labeled monomer RGD peptides using a novel radiolabeling method. Both labeled peptides showed high accumulation in the tumor and exhibited similar biodistribution, demonstrating their usefulness for radiotheranostics. This study applied the labeling method to a dimer RGD peptide with the aim of gaining higher accumulation in tumor tissues based on improved affinity with αvβ3 integrin. We synthesized an iodine-introduced dimer RGD peptide, E[c(RGDfK)] (6), and an 125/131I-labeled dimer RGD peptide, E[c(RGDfK)]{[125/131I]c[RGDf(4-I)K]} ([125/131I]6), and evaluated them as a preliminary step to the synthesis of an 211At-labeled dimer RGD peptide. The affinity of 6 for αvβ3 integrin was higher than that of a monomer RGD peptide. In the biodistribution experiment at 4 h postinjection, the accumulation of [125I]6 (4.12 ± 0.42% ID/g) in the tumor was significantly increased compared with that of 125I-labeled monomer RGD peptide (2.93 ± 0.08% ID/g). Moreover, the accumulation of [125I]6 in the tumor was greatly inhibited by co-injection of an excess RGD peptide. However, a single injection of [131I]6 (11.1 MBq) did not inhibit tumor growth in tumor-bearing mice. We expect that the labeling method for targeted alpha therapy with 211At using a dimer RGD peptide could prove useful in future clinical applications.


Introduction
"Theranostics" is a term that combines "therapeutics" and "diagnostics" and refers to the use of specific agents or techniques that combine diagnosis and targeted therapy [1]. Theranostics has recently gathered considerable attention in oncology as being a safe and effective method for providing personalized medical treatment using a tailored combination of medications to diagnose and subsequently target tumors [2]. In nuclear medicine, cancer is confirmed via molecular imaging with positron emission tomography (PET) or conventional nuclear medicine imaging, including planar or single-photon emission computed tomography (SPECT) imaging, and subsequent therapy involves the delivery of targeted radionuclide therapy with α-particle or β-particle emitter radionuclides. Supposing the diagnostic and therapeutic radiopharmaceuticals show similar biodistribution, absorbed doses in the tumor and each normal tissue during therapy can be calculated based on the quantitative imaging at the time of diagnosis. As the therapeutic effects and side effects of delivered radionuclides are predictable, we can more accurately select the correct therapeutics for each individual patient prior to treatment. Therefore, nuclear medicine is a reasonable method for integrating theranostics. The method of using radioisotopes for theranostics is called "radiotheranostics" [3,4]. medicine is a reasonable method for integrating theranostics. The method of using radioisotopes for theranostics is called "radiotheranostics" [3,4].
In nuclear medicine therapy, targeted alpha therapy (TAT) has gained much attention because of the excellent therapeutic effects derived from the high linear energy transfer (LET) of alpha-particles [5]. Among various alpha emitters, 211 At has become more popular because the half-life (t1/2 = 7.2 h) of 211 At could be sufficient for TAT, and 211 At can be produced from natural bismuth targets via the 209 Bi(α,2n) 211 At nuclear reaction by cyclotron [6]. Numerous promising preclinical studies with 211 At have been reported in recent years [7].
We recently developed a novel 211 At-labeling method for peptides using RGD peptide as a model peptide [8]. On the other hand, the superior biodistribution of tracers, such as higher tumor accumulation, prolonged tumor retention, and lower uptake in normal tissues, would be required for specific therapeutic effects. As one of the strategies to improve the tracer characteristics, it has been reported that the affinity of RGD peptide for αvβ3 integrin is improved by the usage of multivalent peptides, such as dimeric peptides or tetrameric peptides [9].
In this study, we hypothesize that applying the labeling method to a dimeric RGD peptide leads to the development of superior probes for radiotheranostics using a combination of radioiodine and 211 At. Thus, we synthesized and evaluated the 125 Figure 1) as a preliminary step for radiotheranostics with 211 At-labeled dimer RGD peptide. This is because the radiolabeling with radiohalogens such as 125/131 I and 211 At can be achieved using the same tributyltin precursors, and 125 I-and 211 At-labeled probes showed very similar biodistribution patterns in our previous studies [8,10,11]. Moreover, 125 I is commercially available and has a long half-life (t1/2 = 59.4 d), which is appropriate for fundamental research. 131 I is also commercially available and was used for radionuclide therapy to compare the therapeutic effects of 211 At in the future.

Preparation of [ 125/131 I]6
Scheme 1 shows a synthetic scheme of [ 125/131 I]6 and its precursor. After 1 and 2 were synthesized using a general Fmoc solid-phase synthesis method, 3 was synthesized by conjugating Fmoc-Glu-OAll with 1. 4 was synthesized by deprotecting the allyl group of 3, and 5 was synthesized by conjugation of 4 with 2. 6 was synthesized by deprotection of 5. The iodo group in 5 was replaced with a tributylstannyl group via a Pd-catalyzed stannylation reaction to synthesize 7. We then performed 125

Preparation of [ 125/131 I]6
Scheme 1 shows a synthetic scheme of [ 125/131 I]6 and its precursor. After 1 and 2 were synthesized using a general Fmoc solid-phase synthesis method, 3 was synthesized by conjugating Fmoc-Glu-OAll with 1. 4 was synthesized by deprotecting the allyl group of 3, and 5 was synthesized by conjugation of 4 with 2. 6 was synthesized by deprotection of 5. The iodo group in 5 was replaced with a tributylstannyl group via a Pd-catalyzed stannylation reaction to synthesize 7. We then performed 125 6 were verified by comparing their retention times with 6 ( Figure S1). Although the radiochemical purity of [ 131 I]6 was not enough even after HPLC purification, the therapeutic experiments were performed without further purification because the amount of radioactivity was prioritized.
of the protecting groups. The radiochemical yields in the two-  6 were low due to the complicated radiolabeling procedure by two steps. The identities of [ 125/131 I]6 were verified by comparing their retention times with 6 ( Figure S1). Although the radiochemical purity of [ 131 I]6 was not enough even after HPLC purification, the therapeutic experiments were performed without further purification because the amount of radioactivity was prioritized.

αvβ3 Integrin Binding Assay
The affinity of 6 and E[c(RGDfK)]2, a dimer RGD peptide without iodine, for αvβ3 integrin was determined via a competitive binding assay with U-87 MG cells. Representative displacement curves of the assay are shown in Figure 2. Binding of the radioligand [ 125 I]c[RGDy(3-I)V] to αvβ3 integrin was inhibited by 6 and E[c(RGDfK)]2 in a concentration-dependent manner. The half-maximal inhibitory concentration (IC50) values (nM) for 6 and E[c(RGDfK)]2 were 1.2 ± 0.5 and 0.8 ± 0.4 (mean ± SD for three independent experi-

α v β 3 Integrin Binding Assay
The affinity of 6 and E[c(RGDfK)] 2 , a dimer RGD peptide without iodine, for α v β 3 integrin was determined via a competitive binding assay with U-87 MG cells. Representative displacement curves of the assay are shown in Figure 2 indicate that the introduction of an iodine atom into the phenylalanine residue did not significantly impede the affinity of E[c(RGDfK)] 2 for the α v β 3 integrin.
Molecules 2021, 26,6107 ments), respectively. The results indicate that 6 and E[c(RGDfK)]2 possess a specif ity for αvβ3 integrin. Furthermore, these similar values of IC50 of 6 and E[c(RGDfK cate that the introduction of an iodine atom into the phenylalanine residue did no icantly impede the affinity of E[c(RGDfK)]2 for the αvβ3 integrin.

Determination of the Partition Coefficient
An experimental Log P value of [ 125 I]6 was −2.33 ± 0.04. The value was high that of the monomeric radioiodine-labeled RGD peptide [ 125 I]c[RGDf(4-I)K] (−3.04 from our previous study [8]. This result indicates that lipophilicity was increased merization of the RGD peptide.

In Vitro Stability
An in vitro stability experiment of [ 125 I]6 in PBS(−) (pH 7.4) solution was per After incubation at 37 °C for 24 h, 90.8 ± 0.7% (mean ± SD for three samples) of it activity remained intact. Figure 3 and Table S1. [ 125 I]6 was high mulated in the tumor based on the results of the in vitro assay. Specifically, postinjection, the tumor accumulation of [ 125 I]6 (4.12 ± 0.42% ID/g) was significantly than that of [ 125 I]c[RGDf(4-I)K] (2.93 ± 0.08% ID/g) [8]. The accumulation of [ 125 I] liver and intestines also tended to be higher than that of [ 125 I]c[RGDf(4-I)K]. Mea it is known that the radioactive accumulation in the thyroid gland and stomach is a of deiodination of radioiodine labeled probes. In this study, the accumulation of r tivity in the neck containing the thyroid glands and stomach was low (Figure 3), s ing that in vivo deiodination of [ 125 I]6 hardly occurred.

Determination of the Partition Coefficient
An experimental Log p value of [ 125 I]6 was −2.33 ± 0.04. The value was higher than that of the monomeric radioiodine-labeled RGD peptide [ 125 I]c[RGDf(4-I)K] (−3.04 ± 0.46) from our previous study [8]. This result indicates that lipophilicity was increased by dimerization of the RGD peptide.

In Vitro Stability
An in vitro stability experiment of [ 125 I]6 in PBS(−) (pH 7.4) solution was performed. After incubation at 37 • C for 24 h, 90.8 ± 0.7% (mean ± SD for three samples) of its radioactivity remained intact. Figure 3 and Table S1. [ 125 I]6 was highly accumulated in the tumor based on the results of the in vitro assay. Specifically, at 4 h postinjection, the tumor accumulation of [ 125 I]6 (4.12 ± 0.42% ID/g) was significantly higher than that of [ 125 I]c[RGDf(4-I)K] (2.93 ± 0.08% ID/g) [8]. The accumulation of [ 125 I]6 in the liver and intestines also tended to be higher than that of [ 125 I]c[RGDf(4-I)K]. Meanwhile, it is known that the radioactive accumulation in the thyroid gland and stomach is an index of deiodination of radioiodine labeled probes. In this study, the accumulation of radioactivity in the neck containing the thyroid glands and stomach was low (Figure 3), suggesting that in vivo deiodination of [ 125 I]6 hardly occurred.

A comparison of the biodistribution experiments in U-87 MG tumor mice between [ 125 I]6 and [ 125 I]c[RGDf(4-I)K] is shown in
An in vivo blocking study was performed to evaluate whether tumor accumulation was derived from α v β 3 integrin specificity. The effect of c(RGDfK) on tumor uptake of [ 125 I]6 at 1 h postinjection is shown in Figure 4. Co-injection of an excess of c(RGDfK) drastically decreased the tumor uptake of [ 125 I]6, indicating that the tumor accumulation of [ 125 I]6 is caused by its specific binding via α v β 3 integrin. Moreover, it also significantly reduced the accumulation of radioactivity in numerous types of normal tissues (Table S1). It is known that α v β 3 integrin is expressed in the microvessels of normal tissues, such as the liver and lungs [12]. Thus, the result of the blocking study is reasonable for [ 125 I]6 as an α v β 3 integrin-directing agent. An in vivo blocking study was performed to evaluate whether tumor accumulation was derived from αvβ3 integrin specificity. The effect of c(RGDfK) on tumor uptake o [ 125 I]6 at 1 h postinjection is shown in Figure 4. Co-injection of an excess of c(RGDfK) dras tically decreased the tumor uptake of [ 125 I]6, indicating that the tumor accumulation o [ 125 I]6 is caused by its specific binding via αvβ3 integrin. Moreover, it also significantl reduced the accumulation of radioactivity in numerous types of normal tissues (Table S1) It is known that αvβ3 integrin is expressed in the microvessels of normal tissues, such a the liver and lungs [12]. Thus, the result of the blocking study is reasonable for [ 125 I]6 a an αvβ3 integrin-directing agent.  ‡ Expressed as % injected dose. Significance was determined by unpaired Student's t-test (* p < 0.05, ** p < 0.01, *** p < 0.001). † Data from reference [8]. An in vivo blocking study was performed to evaluate whether tumor accumulation was derived from αvβ3 integrin specificity. The effect of c(RGDfK) on tumor uptake of [ 125 I]6 at 1 h postinjection is shown in Figure 4. Co-injection of an excess of c(RGDfK) drastically decreased the tumor uptake of [ 125 I]6, indicating that the tumor accumulation of [ 125 I]6 is caused by its specific binding via αvβ3 integrin. Moreover, it also significantly reduced the accumulation of radioactivity in numerous types of normal tissues (Table S1). It is known that αvβ3 integrin is expressed in the microvessels of normal tissues, such as the liver and lungs [12]. Thus, the result of the blocking study is reasonable for [ 125 I]6 as an αvβ3 integrin-directing agent.

Radionuclide Therapy
The tumor volume and body weight of tumor-bearing mice after treatment in the [ 131 I]6 treatment and control groups are shown in Figure 5 and Figure S2, respectively. There was no significant difference in tumor volume and body weight between the treatment and control groups.

Radionuclide Therapy
The tumor volume and body weight of tumor-bearing mice after treatment in the [ 131 I]6 treatment and control groups are shown in Figures 5 and S2, respectively. There was no significant difference in tumor volume and body weight between the treatment and control groups.

Discussion
We recently reported the simple one-step reaction of an 125 I-and 211 At-labeling method of a monomeric RGD peptide via a Pd-catalyzed stannylation reaction after deprotection [8], whereas, in this study, deprotection after the 125/131 I-labeling reaction, namely the two-step reaction, was performed because the Pd-catalyzed stannylation reaction failed after deprotection. The reason for this failure is not apparent; however, the yield of the stannylation reaction of the monomeric RGD peptide is lower (25%) than that of other stannylation reactions in previous studies (45%-75%) [8,13,14]. We suppose that this difference might be derived from impeding the efficient stannylation reaction by functional groups of the amino acid residues of RGD peptides. As the number of the functional groups in E[c(RGDfK)]{c[RGDf(4-I)K]} (6) is more than that of c[RGDf(4-I)K], it might significantly impede the stannylation reaction. Meanwhile, we expect that the reaction of the 211 At-labeled dimer RGD peptide by the two-step method is possible because the 211 Atlabeled monomer RGD peptide was also synthesized by a similar two-step method [8]. However, the radiochemical yield of the 211 At-labeled dimeric RGD peptide could prove to be too low by this method. Thus, modifying the labeling method might be necessary to improve the complication of the labeling procedure and the radiochemical yields of [ 125/131 I]6 and 211 At-labeled dimer RGD peptides. For this purpose, a one-step radiolabeling method using a radiolabeling precursor without protecting groups would be required. To achieve the precursor synthesis, we will explore the direct stannylation reaction of a nonprotected dimer RGD peptide by improving the metal-catalyzed stannylation reaction or investigating other stannylation reactions, such as photochemical stannylation reactions [15].
The binding affinity of 6 for αvβ3 integrin (IC50: 1.2 ± 0.5 nM) was higher than that of c[RGDf(4-I)K] (IC50: 23.2 ± 17.2 nM) [11]. It was reported that IC50 values of dimeric RGD peptides in the αvβ3 integrin competitive binding assay were an order lower than those of original monomeric RGD peptides [16]. Thus, the IC50 value of dimerizing peptide 6 was consistent with those described in previous studies.
This biodistribution study found that the accumulation of a dimeric RGD peptide, [ 125 I]6, in the tumor tissue was higher than that of a monomeric RGD peptide, [ 125 I]c[RGDf(4-I)K], as reflected in the results of the in vitro αvβ3 integrin binding assay. Therefore, 211 At-labeled dimeric RGD peptide could also accumulate highly in the tumor

Discussion
We recently reported the simple one-step reaction of an 125 I-and 211 At-labeling method of a monomeric RGD peptide via a Pd-catalyzed stannylation reaction after deprotection [8], whereas, in this study, deprotection after the 125/131 I-labeling reaction, namely the two-step reaction, was performed because the Pd-catalyzed stannylation reaction failed after deprotection. The reason for this failure is not apparent; however, the yield of the stannylation reaction of the monomeric RGD peptide is lower (25%) than that of other stannylation reactions in previous studies (45-75%) [8,13,14]. We suppose that this difference might be derived from impeding the efficient stannylation reaction by functional groups of the amino acid residues of RGD peptides. As the number of the functional groups in E[c(RGDfK)]{c[RGDf(4-I)K]} (6) is more than that of c[RGDf(4-I)K], it might significantly impede the stannylation reaction. Meanwhile, we expect that the reaction of the 211 Atlabeled dimer RGD peptide by the two-step method is possible because the 211 At-labeled monomer RGD peptide was also synthesized by a similar two-step method [8]. However, the radiochemical yield of the 211 At-labeled dimeric RGD peptide could prove to be too low by this method. Thus, modifying the labeling method might be necessary to improve the complication of the labeling procedure and the radiochemical yields of [ 125/131 I]6 and 211 Atlabeled dimer RGD peptides. For this purpose, a one-step radiolabeling method using a radiolabeling precursor without protecting groups would be required. To achieve the precursor synthesis, we will explore the direct stannylation reaction of a non-protected dimer RGD peptide by improving the metal-catalyzed stannylation reaction or investigating other stannylation reactions, such as photochemical stannylation reactions [15].
The binding affinity of 6 for α v β 3 integrin (IC 50 : 1.2 ± 0.5 nM) was higher than that of c[RGDf(4-I)K] (IC 50 : 23.2 ± 17.2 nM) [11]. It was reported that IC 50 values of dimeric RGD peptides in the α v β 3 integrin competitive binding assay were an order lower than those of original monomeric RGD peptides [16]. Thus, the IC 50 value of dimerizing peptide 6 was consistent with those described in previous studies.
This biodistribution study found that the accumulation of a dimeric RGD peptide, [ 125 I]6, in the tumor tissue was higher than that of a monomeric RGD peptide, [ 125 I]c[RGDf(4-I)K], as reflected in the results of the in vitro α v β 3 integrin binding assay. Therefore, 211 At-labeled dimeric RGD peptide could also accumulate highly in the tumor because the similar biodistribution of 125 I-and 211 At-labeled dimeric RGD peptides is expected [8]. However, the accumulation of [ 125 I]6 in the liver and intestines was higher than that of [ 125 I]c[RGDf(4-I)K]; we suggest that the reason for this is the increased lipophilicity due to the dimerization of the RGD peptide. Notably, nearly all published studies of radiometallabeled RGD peptides have reported that dimeric RGD peptides showed higher uptakes in the tumor and kidneys than did corresponding monomeric RGD peptides [9,16]. Although Log p values increased by dimerization in those reports, the Log p values were still much lower than those in this study. Thus, the renal excretion of the radiometal-labeled RGD peptides should not be changed. Dijkgraaf et al. proposed that the increased kidney Molecules 2021, 26, 6107 7 of 12 uptake of multimeric RGD peptides was caused by two factors [16]: (1) the expression of β 3 integrins on the endothelial cells of glomeruli vessels [17] and (2) the change in charge brought about by multimerization. Considering the results of this study and these previous reports, [ 125 I]6 could lead to higher accumulation in the kidneys by dimerization; however, the rate of hepatobiliary excretion increased due to increased lipophilicity. Thus, kidney accumulation of [ 125 I]6 did not change significantly compared with that of [ 125 I]c[RGDf(4-I)K]. On the other hand, to further increase the tumor uptake, the introduction of a linker should be effective because it was reported that an appropriate distance between cyclic RGD peptides is important for multivalent effects [18].
In radionuclide therapy, a single administration of [ 131 I]6 (11.1 MBq) did not inhibit tumor growth in tumor-bearing mice ( Figure 5). In a previous study, multiple administrations of the same dose of 90 Y-labeled RGD peptide inhibited tumor growth in tumor-bearing mice; however, a single administration of 90 Y-labeled RGD peptide did not affect tumor growth [19]. As no significant decrease in the body weight was observed in this study ( Figure S2), a higher radiation dosage or multiple administrations may be appropriate and may further inhibit tumor growth. Meanwhile, it has also been reported that conjugation of a 177 Lu-labeled RGD peptide with Evans Blue (EB) as an albumin-binding moiety positively affected its pharmacokinetics to elevate uptake and the residence time in the tumor, and it showed higher tumor growth inhibition than the 177 Lu-labeled RGD peptide without EB [20]. Therefore, increased tumor accumulation of [ 131 I]6 by structural modification such as conjugation with an albumin binder may make it possible to inhibit tumor growth. was synthesized according to a previous report [21]. N,N-Diisopropylethylamine (DIPEA) was purchased from Nacalai Tesque (Kyoto, Japan). 1,3-Diisopropylcarbodiimide (DIPCDI) and 1-hydroxybenzotriazole hydrate (HOBt) were purchased from Kokusan Chemical Co., Ltd. (Tokyo, Japan). 2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethylaminium tetrafluoroborate (TBTU) was purchased from Chem Impex International, Inc. (Wood Dale, IL, USA). Other reagents were of reagent grade and used as received.  (2) was synthesized manually using a standard Fmoc-based solid-phase methodology according to previous reports with slight modifications [8,22]. The crude peptide of 1 and 2 were purified by reversed phase (RP)-HPLC on Cosmosil 5C 18 -AR-II column (20 × 250 mm; Nacalai Tesque) at a flow rate of 12 mL/min with an isocratic mobile phase of 65% methanol in water with 0.1% trifluoroacetic acid (TFA). The solvents were removed by lyophilization to provide 1 (27.1 mg, 30%) and 2 (26.5 mg, 25%) as a white powder.

Synthesis of Reference Compounds and Radiolabeled Compounds
c Compound 4 (7.8 mg, 6.2 µmol) was dissolved in 1 mL of DMF, and then TBTU (3.0 mg, 9.3 µmol) and DIPEA (3.3 µL, 19 µmol) were added to the solution. After stirring at room temperature for 1 h, compound 2 (9.6 mg, 9.3 µmol) was added to the reaction mixture. After 1 h stirring, solvent was removed by rotary evaporation then 1 mL of piperidine was added to the residue. After 1 h stirring at room temperature, purification was performed by RP-HPLC with a Cosmocil 5C 18 -AR-II column (20 × 250 mm) at a flow rate of 12 mL/min with an isocratic mobile phase of 85% methanol in water with 0.1% TFA for 20 min. The solvent was removed by lyophilization to yield 5 (6.9 mg, 54%) as a white powder. E{c
E{c solution (1 mg/mL). After evaporating the solvent using N 2 gas, the residue was treated in 100 µL of a mixture of 95% TFA, 2.5% water, and 2.5% TIS for 90 min at room temperature, and then purified by RP-HPLC with a Cosmosil 5C 18 -AR-II column (4.6 × 150 mm) at a flow rate of 1 mL/min with a gradient mobile phase of 35% methanol in water with 0.1% TFA to 55% methanol in water with 0.1% TFA for 20 min. The radiochemical yield and the radiochemical purities of [ 125 I]6 were 38% and >96%, respectively.

α v β 3 Integrin Binding Assay
Binding affinities of synthesized peptides E[c(RGDfK)] 2 and 6 for α v β 3 integrin were evaluated by competitive inhibition between the peptides and [ 125 I]c[RGDy(3-I)V], which was prepared by Fmoc solid-phase synthesis and following 125 I-labeling with chloramine-T method [23], to α v β 3 integrin according to a previously reported procedure [24]. The peptides' half maximal inhibitory concentration (IC 50 ) values were calculated by curve fitting with nonlinear regression using GraphPad Prism 8.4.3 (GraphPad Software Inc., San Diego, CA, USA). Each data point is the average of four determinations, and IC 50 values were expressed as mean ± standard deviation (SD) from three independent experiments.

Determination of the Partition Coefficient
The partition coefficient of [ 125 I]6 was measured as described previously [25]. The radioactivity in each phase was measured with an auto-well gamma counter. The partition coefficient was determined by calculating the ratio of 1-octanol to the buffer and was expressed as a common logarithm (log P).

In Vitro Stability
To evaluate the stability of [ 125 I]6 in PBS(−), 10 µL of each tracer (37 kBq) was added to 90 µL of PBS(−) (pH 7.4), and the solutions were incubated at 37 • C for 24 h. After incubation, the samples were drawn, and the radioactivity was analyzed by RP-HPLC.

Biodistribution of E[c(RGDfK)]{[ 125 I]c[RGDf(4-I)K]} 2 ([ 125 I]6) in Tumor-Bearing Mice
Experiments with animals were conducted in strict accordance with the Guidelines for the Care and Use of Laboratory Animals of Kanazawa University. The experimental protocols were approved by the Committee on Animal Experimentation of Kanazawa University. The animals were housed with free access to food and water at 23 • C with a 12 h alternating light/dark schedule. We used a U-87 MG cell line because it was reported that the U-87 MG highly expresses α v β 3 integrin [26]. U-87 MG cells were grown and inoculated subcutaneously into 4-week-old female BALB/c nude mice (15-19 g, Japan SLC, Inc., Hamamatsu, Japan) as previously reported [22]. Biodistribution experiments were performed approximately 14 days post-inoculation. A solution of [ 125 I]6 (37 kBq) was intravenously administered to groups of four mice. Mice were sacrificed at 1 and 4 h postinjection. Tissues of interest were removed and weighed. A neck containing thyroid was resected. The radioactivity counts of 125 I was determined with an auto-well gamma counter (ARC-7010, Hitachi, Ltd., Tokyo, Japan) and corrected for background radiation and physical decay during counting. A window from 16 to 71 keV was used for counting 125 I.
To investigate the effect of an excess amount of RGD peptide on biodistribution, U-87 MG tumor-bearing mice were intravenously administered 100 µL of a mixture solution of [ 125 I]6 (37 kBq) with c(RGDfK) peptide (0.2 mg/mouse). Mice (n = 4) were sacrificed at 1 h postinjection, and biodistribution experiments were conducted as described above.

Statistical Evaluation
Animal experiments were compared using unpaired Students' t-test.

Conclusions
In this study, we synthesized an iodine-introduced dimer RGD