Radioiodination of Rhenium Cyclized α-Melanocyte-Stimulating Hormone Resulting in Enhanced Radioactivity Localization and Retention in Melanoma

Radiohalogen-labeled biomolecules such as monoclonal antibodies (MAbs) and peptides have attracted intensive interest for the development of diagnostic and therapeutic radiopharmaceuticals for targeting tumors (1, 2, 3, 4, 5, 6). The clearance of radiohalogenated MAbs and peptides from normal tissues is generally faster than that for radiometal labeled MAbs, and a variety of radiohalogens are available for potential imaging and therapeutic applications, giving radiohalogenated proteins an important role in radiopharmaceutical development for cancer diagnosis and treatment (7, 8). However, rapid degradation in vivo is often observed for internalized radiohalogenated biomolecules such as ¹³¹I-labeled peptides and MAbs radiolabeled with conventional chloramine-T or Iodogen methods. Dehalogenation and proteolysis generally decrease the residence time of radiohalogenated proteins or peptides in the target tumor.

Several halogen labeling approaches have been developed to decrease dehalogenation and increase tumor cell retention. Conjugation labeling agents such as N-succinimidyl 3-iodobenzoate (SIB) or N-succinimidyl 4-iodobenzoate (PIB) have been reported to be inert to dehalogenation; thus, MAbs labeled with N-succinimidyl 3-iodobenzoate showed higher accumulation in a human tumor model and enhanced therapeutic efficacy (9, 10). Additional methods to increase radiohalogen retention in the tumor cell have involved the use of positively charged molecules and nonmetabolizable adducts. For example, N-succinimidyl 5-iodo-3-pyridinecarboxylate-labeled Mabs were shown to resist cellular washout due to the trapping of positively charged catabolites [pyridinium cations (11, 12)]. The use of N-succinimidyl 5-iodo-3-pyridinecarboxylate for radioiodination (12), the polycationic peptide [D-KRYRR (13)], and nonmetabolizable carbohydrate-tyramine adducts (14, 15, 16, 17) has enhanced cellular retention of radioactivity; however, low radiolabeling yields and MAb aggregation have been problems associated with the carbohydrate-tyramine adducts (14, 18, 19, 20).

The presence of α-melanocyte-stimulating hormone (α-MSH) receptors on multiple murine and human melanoma cell lines has led to the development of radiolabeled peptides for potential melanoma detection and treatment. α-MSH receptors display nanomolar to subnanomolar affinities for α-MSH peptides and are rapidly internalized on ligand binding (21, 22). High receptor affinity and specificity make α-MSH peptide analogs attractive vehicles for delivering radionuclides to melanoma cells (23, 24, 25). Efforts to develop radiohalogenated α-MSH analogs for melanoma targeting have been disappointing for the most part. Radiolabeling of the gold standard α-MSH analog, [Nle⁴,d-Phe⁷]α-MSH (NDP), directly with ¹²⁵I at the Tyr² residue resulted in very slow clearance (low tumor:blood ratios) and in vivo deiodination, making it unsuitable for melanoma radioimaging or radiotherapy (26). NDP labeled with N-succinimidyl-3-iodobenzoate (¹²⁵I-SIB) at Lys¹¹ residues to give ¹²⁵I-3- or 4-iodobenzoate (IBA)-NDP (conjugate labeling approach) resulted in increased affinity (lower K_D; 10 versus 140 pm) and significantly lower thyroid and stomach uptake than ¹²⁵I-Tyr²-NDP in normal mice (7). Likewise, succinimidyl-4-fluorobenzoate (¹⁸F-SFB) labeled NDP exhibited rapid clearance in normal mice; however, no in vivo data are reported in tumor-bearing mice (27).

A novel family of α-MSH analogs has been developed that incorporates the transition metals rhenium or technetium directly into the peptide’s structure to generate the cyclic α-MSH analog ReO[Cys^3,4,10,d-Phe⁷]α-MSH_3–13 [ReCCMSH (23, 24)]. ^99mTc/¹⁸⁸Re cyclized [Cys^3,4,10,d-Phe⁷]-α-MSH_3–13 (Tc/Re-CCMSH) showed high tumor uptake, prolonged tumor retention, and high stability in both B16/F1 murine and TXM-13 human melanoma-bearing mouse models (23, 24). Substitution of Arg¹¹ for Lys¹¹ in ReCCMSH resulted in the analog, ReCCMSH(Arg¹¹), which showed greater tumor uptake and lower kidney accumulation (28, 29). Therefore, ReCCMSH(Arg¹¹) was used as a structural motif for radiohalogenation.

In this report, radioiodinated ReCCMSH(Arg¹¹) α-MSH analogs were designed and synthesized to demonstrate the potential for incorporating radiohalogens with improved tumor uptake and retention as agents for melanoma detection (¹⁸F and ¹²³I) and therapy (¹³¹I and ²¹¹At). A commercially available reagent, ¹²⁵I-PIB, was selected for radioiodination of ReCCMSH(Arg¹¹) because of its strong resistance to dehalogenation on incorporation into biomolecules (2). A Lys or d-Lys residue was added at position 2 in ReCCMSH(Arg¹¹) for incorporation of the radiohalogen at the Lys ε-amino group. The d-amino acid residue (d-Lys) should further improve the in vivo stability of the peptide. The syntheses, radioiodination chemistry, in vitro cell binding studies, and in vivo biodistribution studies in tumor-bearing mice are reported.

All 9-fluorenylmethyloxycarbonyl amino acids and resins were purchased from Nova-Biochem Co. and Advanced Chemtech (Louisville, KY). All other chemicals were obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO). [¹²⁵I]PIB was purchased from DuPont New England Nuclear LifeScience (Boston, MA; 2200 Ci/mmol). B16/F1 murine melanoma cells were obtained from American Type Culture Collection. An Applied Biosystems (Foster City, CA) Synergy432A desktop solid-phase peptide synthesizer and an ISCO Inc. (Lincoln, NE) high-performance liquid chromatography (HPLC) system were used for peptide synthesis and purification.

NDP, Ac-Lys-CCMSH(Arg¹¹), and Ac-d-Lys-CCMSH(Arg¹¹) were synthesized by conventional solid-phase peptide synthesis methods using 9-fluorenylmethyloxycarbonyl/O-benzotriazolyl-tetramethyluronium hexafluorophosphate chemistry. Peptide identities were confirmed by electrospray ionization mass spectrometry (ESI-MS). Ac-Lys-ReCCMSH(Arg¹¹) and Ac-d-Lys-ReCCMSH(Arg¹¹) were prepared by cyclization with rhenium coordination using a transchelation reaction with rhenium-glucoheptonate, as described previously (24), and they were lyophilized and identified by ESI-MS after purification. Ac-SYS(Nle)EHdFRWGKPV-NH₂ (NDP), Ac-KCCEHdFRWCRPV-NH₂ [Ac-Lys-CCMSH(Arg¹¹)], Ac-K-ReO[CCEHdFRWC]RPV-NH₂ [Ac-Lys-ReCCMSH(Arg¹¹)], Ac-_DKCCEHdFRWCRPV-NH₂ [Ac-d-Lys-CCMSH(Arg¹¹)], and Ac-_DK-ReO[CCEHdFRWC]RPV-NH₂ [Ac-d-Lys-ReCCMSH(Arg¹¹)] were synthesized and characterized, with the ESI-MS results for these peptides in agreement with the calculated values.

The peptides NDP, Ac-Lys-ReCCMSH(Arg¹¹), and Ac-d-Lys-ReCCMSH(Arg¹¹) were dissolved in dimethyl formamide to a final concentration of 5 μg/μl solution. 4-Iodobenzoic acid (13.1 μl; 7.2 μg/μl), diiospropylethylamine (0.25 μl), and O-benzotriazolyl-tetramethyluronium hexafluorophosphate (13.4 μl; 10.8 μg/μl) were mixed together in a small tube and allowed to sit at room temperature for 10 min. The peptide (29 μl; 5 μg/μl) was then added, and the resulting solution was incubated at room temperature for 20 min. The resulting product, the iodinated peptide, was isolated by reversed phase HPLC (RP-HPLC) using a gradient system of 0.1% trifluoroacetic acid in acetonitrile (solvent B) and 0.1% trifluoroacetic acid in water (solvent A). The gradient system was optimized for each peptide analog, ranging from an initial solvent composition of 18% to 28% solvent B (depending on the peptide) going to 90% solvent B by 23 min and returning to the initial solvent composition by 32 min. Under the optimized solvent system for each peptide, retention times ranged from 13 to 23 min. After lyophilization, the iodinated peptides were identified by ESI-MS. Ac-SYS(Nle)EHdFRWGK(IBA)PV-NH₂ [NDP(Lys¹¹-IBA)], Ac-K(IBA)-ReO[CCEHdFRWC]RPV-NH₂ [Ac-Lys(IBA)-ReCCMSH(Arg¹¹)], and Ac-_DK(IBA)-ReO[CCEHdFRWC]RPV-NH₂ [Ac-d-Lys(IBA)-ReCCMSH(Arg¹¹)] were synthesized and characterized, with the ESI-MS results consistent with the calculated formulations.

The peptides NDP, Ac-Lys-ReCCMSH(Arg¹¹), and Ac-d-Lys-ReCCMSH(Arg¹¹) [5 μg/μl in dimethyl formamide] and 3 μl of N-diisopropylethylamine were added to 5 μl of [¹²⁵I]PIB (approximately 100 μCi). The reaction mixtures were incubated for 20 min at room temperature. After incubation, the radioiodinated peptides were purified by RP-HPLC, lyophilized, and stored at −20°C until further use. The stability of the radiolabeled complexes was determined in 0.01 m PBS (pH 7.4)/0.1% BSA.

B16/F1 murine melanoma cells were cultured in RPMI 1640 containing NaHCO₃ (2 g/liter), which was supplemented with 10% heat-inactivated FCS, 2 mm l-glutamine, and 48 mg of gentamicin. The cells were expanded in 75-cm² tissue culture flasks and kept in a humidified atmosphere of 5% CO₂ at 37°C, with the medium changed every other day. A confluent monolayer was detached with 0.02% EDTA in Ca²⁺- and Mg²⁺-free 0.01 m PBS (pH 7.4) and dissociated into a single cell suspension for further cell culture.

Receptor binding affinity, internalization, and cellular retention assays for the ¹²⁵I-labeled peptides were performed using B16/F1 murine melanoma cells lines. Cells were prepared by seeding at a density of 0.2 million cells/well in 24-well tissue culture plates and allowed to attach overnight. After washing once with the binding media (Modified Eagle’s Medium with 25 mm HEPES, 0.2% BSA, and 0.3 mm 1,10-phenanthroline), the cells were incubated at 25°C for 2 h with 100,000–500,000 dpm of the radiolabeled complex in 0.5 ml of binding media. The nonspecific binding was determined by coincubation with nonradiolabeled NDP at a final concentration of 10 μm. The cells were rinsed twice with 0.01 m PBS (pH 7.4)/0.2% BSA and lysed in 0.5 ml of 1 m NaOH for 5 min, and the radioactivity was measured. The cell binding capacity was calculated as the percentage of radioactivity bound to the cells divided by the total radioactivity added. All cell binding experiments were carried out in triplicate.

The apparent equilibrium dissociation constants (K_D) of the peptides were determined using competitive binding experiments with radioiodinated peptides over a 10⁻¹⁴ to 10⁻⁶ (m) concentration range of NDP, using the method described previously (30). B16/F1 cells were prepared as described above in 24-well tissue culture plates and incubated at 25°C for 3 h with approximately 100,000–500,000 dpm of ¹²⁵I-labeled α-MSH analogs in 0.5 ml of binding media containing NDP at different concentrations. The radioactivity in the cells and in the media was collected separately and measured. The data were processed with the RADLIG software program, and the K_D values of the radioiodinated complexes were calculated using the Origin 6.1 software program and the Cheng-Prusoff equation.

Internalization and cellular retention of the three ¹²⁵I-labeled peptides were examined in B16/F1 cells prepared as described above in 24-well tissue plates. Approximately 50,000 dpm of ¹²⁵I-labeled peptide in 0.5 ml of binding media were added to each well containing cells and incubated at 25°C for 10 min to 4 h. Internalization of the radiolabeled peptide was determined by washing the cells with acidic buffer [40 mm sodium acetate (pH 4.5) containing 0.9% NaCl and 0.2% BSA] to remove the membrane-bound radiocomplex and then measuring the remaining internalized radioactivity. The cellular retention properties of the membrane-bound and internalized ¹²⁵I-labeled peptides were determined by incubating B16/F1 cells in binding media at 37°C with the radiolabeled complexes for 2 h. The binding media were then removed, the cells were washed twice with cold 0.01 m PBS (pH 7.4)/0.2% BSA, 0.5 ml of culture media was added, and radioactivity release into the media was monitored at various time points over a 4-h incubation period.

All animal studies were carried out in compliance with federal and local institutional rules for the conduct of animal experimentation. C57 BL/6 female mice (7–8 weeks old) were inoculated s.c. in the right flank with 1 × 10⁶ cultured B16/F1 murine melanoma cells. Nine to ten days after inoculation, when the tumors had grown to a weight of ∼500 mg, the mice received injection with 2 μCi of ¹²⁵I-labeled peptide through the tail vein; the mice were housed separately, and their urine and feces were collected. Because the mice were not housed in metabolic cages, the urine in the bladder of the mouse plus all urine in the cage was pooled for each time point. Precautions were taken to carefully separate the urine and feces, and absorbent paper was placed in each cage to soak up the urine excreted by the mouse. The mice were sacrificed at different time points from 30 min to 72 h postinjection (p.i.). Tumors and normal tissues of interest were removed and weighed, and their radioactivity was measured using a gamma counter. The radioactivity uptake in the tumor and normal tissues was expressed as a percentage of the injected radioactive dose per gram of tissue (% ID/g) or as a percentage of the injected dose (% ID).

Statistical analysis was performed using the Student’s t test for unpaired data. A 95% confidence level was chosen to determine the significance between compounds, with P < 0.05 being significantly different.

Urine samples were obtained during the animal experiments and pooled together for each time point. They were diluted with 500 μl of 0.9% NaCl and filtered through Centricon-10 filters to isolate the low molecular weight (<10,000) catabolites. The filtrate and filter were counted on a gamma counter. The filtrates were analyzed by RP-HPLC under the identical conditions used for analyzing the original radiolabeled compound. Fractions were collected in 30-s intervals and counted on a gamma counter.

Three α-MSH analogs [the rhenium cyclized CCMSH analogs Ac-Lys-ReCCMSH(Arg¹¹) and Ac-d-Lys-ReCCMSH(Arg¹¹) (Fig. 1) and NDP] were radioiodinated with ¹²⁵I-PIB for biological investigation. All peptides were synthesized using conventional solid-phase peptide synthesis methods, followed by a rhenium cyclization reaction. The nonradioactive iodinated peptides were prepared with 4-iodobenzoic acid to allow for characterization of the radioiodinated analogs of these peptides. All of the peptides were purified by RP-HPLC and characterized by ESI-MS. The radioiodinated products, which have retention times similar to the nonradioactive iodinated peptides under the same HPLC gradients, were used for biological evaluation. The effects of rhenium cyclization on the biological properties of the radiolabeled peptide were compared using the radioiodinated linear peptide NDP and the two rhenium cyclized peptides. The difference between radioiodinated d-Lys and l-Lys tumor uptake and retention was assessed.

Because the amount of receptor present on the tumor cells is limited, the radiopharmaceutical (agonist) should be of high specific activity. The conjugation of ¹²⁵I-PIB with the peptides in dimethyl formamide and N-diisopropylethylamine was usually higher than 80%. The radioiodinated NDP, Ac-Lys-ReCCMSH(Arg¹¹), and Ac-d-Lys-ReCCMSH(Arg¹¹) were easily separated from their nonradiolabeled counterparts by HPLC, with the difference in retention times of radiolabeled and unlabeled peptide being greater than 10 min, making the specific activity of the three radioiodinated peptides nearly the same as the specific activity of the starting radioiodination agent, ¹²⁵I-PIB (2200 Ci/mmol).

The radiochemical stability of the ¹²⁵I-PIB-labeled analogs was evaluated in PBS (pH 7.4). Over a 24-h period of incubation at 25°C in PBS, only the radiolabeled peptide peak was observed by RP-HPLC. The lyophilized radiolabeled complex could be stored at −20°C for 2 weeks without any observed degradation.

Fig. 2 shows the cell binding and internalization of ¹²⁵I-IBA-NDP, Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹), and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) in B16/F1 cells at 37°C over a 4-h incubation period. The cell binding observed for ¹²⁵I-IBA-NDP at 1 h of incubation was 12%, and this decreased to 3.5% at 4 h. However, the receptor binding for Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) steadily increased over the 4-h incubation time to 4.2% and 5.1%, respectively. The three radiolabeled analogs showed rapid internalization with approximately 80% internalization of the receptor-bound radiolabeled complexes within 5 min of incubation. The internalized versus total binding percentages did not change significantly with the incubation time.

Cellular retention of the radiolabeled analogs as a function of time is shown in Fig. 3. After a 2-h incubation with B16/F1 cells, the cells were returned to the cell culture medium and incubated at 37°C for a 4-h period. The receptor-bound radioiodinated linear α-MSH analog NDP showed radioactivity release from the cells into the medium very quickly, with only 59% remaining bound to the receptors at 10 min and decreasing to 11% at 4 h. Although 69% and 34% of the cell-associated radioactivity for Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) remained in the cells at 10 min and 4 h, respectively, Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) exhibited the slowest receptor-bound radioactivity wash out rate of the analogs, 82% and 43% of the radioactivity associated with the cells at 10 min and 4 h, respectively. These in vitro data clearly demonstrate that rhenium cyclization significantly enhances peptide trapping and retention in the cells, as does d-amino acid incorporation. The combination of these two effects resulted in a 2.9-fold increase in the retention of radioactivity for Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) relative to ¹²⁵I-IBA-NDP at 4 h.

In vitro receptor binding of radioiodinated α-MSH analogs was performed using the murine B16/F1 cell line. The specific/nonspecific binding ratios for ¹²⁵I-IBA-NDP, Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹), and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) were 14.8, 7.9, and 5.0, respectively. The apparent equilibrium dissociation constants were determined using nonradioactive NDP as a competitor and calculated using Origin 6.1 software. The K_D values for ¹²⁵I-IBA-NDP, Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹), and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) were 1.02 ± 0.48 × 10⁻¹¹, 1.41 ± 0.14 × 10⁻¹¹, and 2.08 ± 0.04 × 10⁻¹¹ m, respectively. The K_D value for ¹²⁵I-IBA-NDP determined is similar to the reported value of 1.0 ± 0.5 × 10⁻¹¹ m (7).

The in vivo biodistribution of radioiodinated NDP, Ac-Lys-ReCCMSH(Arg¹¹), and Ac-d-Lys-ReCCMSH(Arg¹¹) was examined in a B16/F1 melanoma tumor-bearing mouse model. Biodistribution data are presented in Table 1. The tumor uptake values for ¹²⁵I-IBA-NDP were significantly lower than that of Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) for all time points investigated (0.05 < P < 0.01 at 0.5 h and P < 0.01 at 2, 4, and 24 h). The highest tumor uptake observed for Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) was 17.69 ± 4.13% ID/g at 2 h p.i., compared with 5.34 ± 1.35% ID/g for radioiodinated NDP at 0.5 h p.i. (Table 1). The tumor uptake observed for the l-Lys analog was lower than that observed for the d-Lys analog at most time points, although the difference was not statistically significant (P > 0.05).

The radioiodinated NDP quickly washed out of the tumor, with only 0.50 ± 0.20% ID/g remaining at 4 h. The radioiodinated rhenium cyclized peptides showed significantly higher tumor retention, with 4.92 ± 0.48 and 7.18 ± 2.14% ID/g remaining for Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹), respectively, at 24 h p.i. These values are approximately 19 and 28 times greater than the tumor retention observed for ¹²⁵I-IBA-NDP.

The thyroid (<1% ID) and stomach (<1.61% ID) accumulations at all time points were very low (Table 1), suggesting that these ¹²⁵I-PIB-labeled peptides are stable to deiodination.

¹²⁵I-IBA-NDP showed lower uptake in the blood and other blood-rich organs (e.g., heart, lung, spleen, and liver) than did Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) at the earlier time points [0.5, 2, and 4 h (P < 0.01)], although there were no significant differences between these two compounds in most of the normal tissues at 24 h. The Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) showed significantly higher blood uptake than Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) at most time points (P < 0.01) and also had higher radioactivity accumulation in most other normal tissues, but with no significance.

Although the three radiolabeled peptides were eliminated primarily through the kidneys into the urine, the clearance for the radioiodinated linear peptide NDP was much more rapid than that for the radioiodinated rhenium cyclized peptides. The urine activity was 88.03 ± 3.86% ID for ¹²⁵I-IBA-NDP at 4 h p.i., whereas it was only 56.01 ± 4.74% ID and 50.19 ± 2.28% ID were observed for Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) and Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹), respectively. The two ReCCMSH peptides exhibited higher gastrointestinal (GI) clearance, with feces radioactivity of 12.18 ± 1.57% ID/g and 16.71 ± 3.30% ID/g at 24 h p.i., respectively, for Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) versus 3.18 ± 2.27% ID/g for ¹²⁵I-IBA-NDP. The reason for the increased GI excretion for Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) is not clear; however, it is not due to the Re cyclization of the peptide. ^99mTc/¹⁸⁸Re-CCMSH exhibits predominantly urinary clearance, and ¹¹¹In-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA)-ReCCMSH shows lower GI uptake and clearance than the non-rhenium cyclized ¹¹¹In-DOTA-CCMSH (31).

The radiolabeled rhenium cyclized peptides exhibited higher radioactivity uptake in normal tissues (e.g., kidneys) than did radioiodinated NDP; however, the rhenium-incorporated peptides showed high tumor uptake and retention, resulting in higher tumor:kidneys ratios than observed for NDP at all time points investigated (Table 1). Additionally, the d-Lys analog exhibited higher tumor uptake and faster blood clearance than the l-Lys analog. Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) showed the highest tumor:blood and tumor:muscle ratios at most time points, with ratios of 34.3 and 1046.7 observed at 24 h p.i. (Table 1).

To better understand the differences in the biological patterns of the iodinated rhenium cyclized peptides and the linear peptide NDP, catabolites from both the in vitro cell culture studies and in vivo from urine were analyzed by RP-HPLC.

The supernatants from in vitro cell culture assays at the 4 h time point were collected, separated according to molecular weight using Centricon-10 filters (M_r 10,000 cutoff), and counted. The major difference observed between the catabolites of the radioiodinated linear peptide and the rhenium cyclized peptides generated from in vitro incubation with B16/F1 melanoma cells was that the majority of the activity for ¹²⁵I-IBA-NDP was in the low molecular weight (LMW) form (67% versus 24% for the Re cyclized peptides), whereas 76% of the radioactivity for both radioiodinated rhenium cyclized peptides was in the high molecular weight-associated form (versus 33% for ¹²⁵I-IBA-NDP).

Urine samples collected from animal experiments were pooled, separated on Centricon-10 filters into high molecular weight (>10,000) and LMW (<10,000) catabolites, and then counted. The results are presented in Table 2. The majority of the catabolites for these radioiodinated peptides are present in the LMW forms in the urine, and there were no significant differences in the percentage of LMW species for three complexes at various time points (Table 2). The LMW species were analyzed to determine the nature of the catabolites excreted from mice receiving ¹²⁵I-IBA-NDP, Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹), and Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) (Fig. 4). The identities of the peaks were determined by comparison with nonradioactive standards of the expected catabolites, 6-[N-(3-iodobenzamido)]-2-aminocaproic acid (IBA-Lys), 2-[N-(3-iodobenzamido)]acetic acid (IBA-Gly), and 3-iodobenzoic acid (IBA; Refs. 27 and 32), analyzed under the same HPLC conditions. The hydrophilicity order of these potential catabolites is iodide > IBA-Gly > IBA-Lys > IBA (27, 32), which is their order of expected elution under the gradient system used in this study.

The radiochromatogram of ¹²⁵I-IBA-NDP indicated that all of the radiolabeled complex decomposed rapidly to small radioiodinated conjugates such as [¹²⁵I]IBA-Lys, [¹²⁵I]IBA-Gly, and some unknown species. However, a small amount of Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) peptide (approximate RT = 18.5 min) remained in the urine. The amount of small radioiodinated conjugates increased gradually as a function of time. On the other hand, for Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹), the majority of radioactivity remained in its original form for up to 24 h p.i. In addition, catabolite analysis of both iodinated l-Lys and d-Lys containing rhenium cyclized peptides revealed one large unknown peak eluting at a later time (RT = 25 min). These results clearly demonstrate that the order of in vivo stability of these three peptides is Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) > Ac-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) > ¹²⁵I-IBA-NDP.

The presence of α-MSH receptors on melanoma cells and the fact that the receptors undergo rapid internalization on ligand binding have led scientists to explore the possibility of using peptide ligands as carriers to deliver radionuclides for detection and treatment of melanoma (21). Because radioiodinated MAbs or peptides usually exhibit faster clearance rates from normal tissues compared with radiometal-labeled ones, and a variety of radiohalogens are available for possible imaging and therapeutic purposes, halogenated proteins continue to play an important role in radiopharmaceutical development for cancer diagnosis and treatment (7, 8). However, rapid release of radioactivity from tumors after internalization of radiohalogenated peptides remains a major concern to be resolved before therapeutic trials (14, 33).

In this study, a peptide consisting of three segments was designed and evaluated, namely, Ac-[¹²⁵I-PIB]-d-Lys-ReCCMSH(Arg¹¹). ReCCMSH(Arg¹¹) is the receptor targeting component that resists proteolytic degradation, d-Lys serves as a residue for label conjugation and inhibits peptide degradation, and the PIB label reduces dehalogenation in vivo. This combination resulted in high tumor uptake and retention for the radiohalogenated peptide.

The in vitro receptor binding and in vivo experiments demonstrated that attachment of an iodobenzoate group at either Lys¹ in rhenium cyclized CCMSH or Lys¹¹ in NDP did not alter their bioactivity. Low K_D and high tumor uptake values for the radioiodinated peptides were observed. The biodistribution of NDP in B16/F1 murine melanoma-bearing C57 BL/6 mice labeled with ¹²⁵I using chloramine T was reported in our previous work (26). From the biodistribution data, the conjugation labeling approach showed several advantages over direct labeling, including significantly lower stomach and thyroid uptake than was observed using the chloramine-T method at all time points investigated, indicating a decrease in the in vivo deiodination of the radiolabeled complex. The iodination site generated by the ¹²⁵I-PIB labeling method is structurally dissimilar to iodotyrosine, a compound that is susceptible to multiple endogenous deiodinases. The stability to deiodination of conjugation labeling agents such as PIB and SIB has been demonstrated repeatedly (7, 32).

NDP labeled with ¹²⁵I-PIB demonstrated faster clearance from normal tissues and the body than peptides directly labeled with ¹²⁵I. At 4 h p.i., 88.03% ID of the radioactivity was in the urine for ¹²⁵I-IBA-NDP, whereas only 42.67% ID was present in the urine for directly labeled ¹²⁵I-NDP. ¹²⁵I-NDP showed significantly higher radioactivity accumulation in the blood and blood-rich organs, such as the heart, lung, spleen, and liver, and higher muscle uptake than ¹²⁵I-IBA-NDP. In addition, our data are comparable with the biodistribution data for ¹²⁵I-IBA-NDP and ¹²⁵I-NDP reported in normal mice (7).

The ¹²⁵I-PIB labeling approach appears to increase tumor uptake and retention compared with direct labeling. ¹²⁵I-IBA-NDP exhibited a higher tumor:blood ratio at both 30 min and 4 h, demonstrating improved tumor uptake and retention for the conjugation labeling method. The reason for this may be that Lys¹¹ is closer to the binding sequence His-d-Phe-Arg-Trp than Tyr². When proteolysis occurs, ¹²⁵I on Lys¹¹ may remain associated with a peptide fragment retaining its receptor binding capability, therefore increasing the tumor uptake.

The ReCCMSH motif plays a major role in improving the tumor uptake and retention of the radioiodinated complexes. Both ¹²⁵I-labeled ReCCMSH analogs exhibited significantly higher tumor uptake and retention than the ¹²⁵I-labeled linear peptide NDP, most likely a result of high in vivo stability, as demonstrated by the urine catabolite analysis of the three radiolabeled complexes. The use of d-Lys coupling rather than l-Lys with ReCCMSH increased the in vivo stability of the iodinated complex, thus contributing to higher tumor retention for the d-Lys analog.

The ReCCMSH motif also resulted in iodinated complexes with significantly higher nonspecific cell binding than iodinated NDP. The high nonspecific binding properties of Ac-d- or l-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) might be responsible for their high blood activity, as well as the high molecular weight catabolites observed in both cell culture supernatant and urine samples. These complexes appear to be more lipophilic than ¹²⁵I-IBA-NDP based on HPLC retention times, and this would result in greater hepatobiliary excretion, which is observed (Table 1). The more rapid clearance rate from blood and other normal tissues for Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) compared with the l-Lys analogs, however, is not understood at this point.

The long-term goal of our research is to develop therapeutic radiopharmaceuticals for melanoma treatment based on radiolabeled α-MSH analogs. Because α-MSH analogs labeled with either radiohalogens (i.e., ¹³¹I and ²¹¹At) or radiometals (⁹⁰Y, ¹⁷⁷Lu, and ²¹²Bi) are potential candidates for melanoma targeting, it is important to compare the distributions of radiohalogenated (¹²⁵I-labeled) and radiometalated (¹¹¹In-labeled) α-MSH analogs. A comparison of the biodistributions of ¹²⁵I-IBA-NDP and ¹¹¹In-DOTA-NDP and of Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) and ¹¹¹In-DOTA-ReCCMSH(Arg¹¹) is listed in Tables 3 and 4, respectively.

Table 3 compares the biodistribution data for the two radiolabeled linear NDP analogs, ¹²⁵I-IBA-NDP and ¹¹¹In-DOTA-NDP, and it is clear that their biodistribution patterns are different. The tumor uptake for the ¹¹¹In-labeled complex was approximately 10 times that of the ¹²⁵I-labeled peptide at 24 h p.i.; however, it cleared at a slower rate from normal tissues. The uptake for the dose-limiting kidneys was significantly higher at 24 h p.i. for ¹¹¹In-DOTA-NDP than for ¹²⁵I-IBA-NDP (10.2% ID/g versus 0.23% ID/g). These observations can be attributed to the different metabolic patterns of radioiodinated versus radiometal chelated peptides. The catabolites from the iodinated peptide (i.e., IBA, IBA-Lys, and IBA-Gly) are most likely expelled from the cells and excreted from the body quickly. However, the free radiometal and/or the radiometal chelate complex remained in the cells after metabolism, thus resulting in an apparent higher radioactivity accumulation for those organs (34, 35, 36, 37, 38). The clearance routes were also different, with ¹¹¹In-DOTA-NDP clearing predominantly via the kidneys/urine, and ¹²⁵I-IBA-NDP clearing through both the urinary and GI tracts.

The rhenium incorporation markedly changed the biodistribution of radioiodinated and radiometal labeled α-MSH peptide analogs, as seen in Table 4. Both Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) and ¹¹¹In-DOTA-ReCCMSH(Arg¹¹) showed good tumor uptake and retention, although the iodinated analog was slower to reach its maximal uptake. The clearance rate of the ¹¹¹In-labeled analog was faster than that for the ¹²⁵I-labeled analog from most normal tissues, except the kidneys. The ¹¹¹In-labeled peptide cleared primarily through the urinary system, whereas the ¹²⁵I-labeled analog showed both urinary and GI clearance. The kidney clearance of the ¹²⁵I-labeled analog was much better than that seen for the ¹¹¹In-labeled analog, although its initial uptake was higher. Similar tumor uptake and significantly lower kidney uptake for the iodinated peptide compared with the ¹¹¹In-labeled peptide suggest that Ac-d-Lys-ReCCMSH(Arg¹¹) labeled with therapeutic halogens such as ¹³¹I and ²¹¹At might have some advantages for melanoma therapy over DOTA-ReCCMSH(Arg¹¹) labeled with α- and β-emitting radiometals.

In conclusion, Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) exhibits high tumor uptake and retention, comparable with ¹¹¹In-DOTA-ReCCMSH(Arg¹¹). It is also rapidly cleared from normal tissues. The favorable biodistribution characteristics of this molecule make it a superb candidate for conjugation with radionuclides used for treatment of melanoma. Ac-d-Lys(¹²⁵I-IBA)-ReCCMSH(Arg¹¹) was developed as an iodinated α-MSH analog, which exhibits high tumor uptake and retention and rapid normal tissue clearance. To the best of our knowledge, this is the first iodinated peptide developed with such long tumor retention. Melanoma radiotherapy experiments will be performed based on this molecule.

We express our gratitude to Drs. Timothy J. Hoffman, Wynn Volkert, and Susan L. Deutscher for helpful discussions and assistance and Dr. Nellie K. Owen and Donna Whitener for technical support.