111In-Labeled Cystine-Knot Peptides Based on the Agouti-Related Protein for Targeting Tumor Angiogenesis

Agouti-related protein (AgRP) is a 4-kDa cystine-knot peptide of human origin with four disulfide bonds and four solvent-exposed loops. The cell adhesion receptor integrin αvβ3 is an important tumor angiogenesis factor that determines the invasiveness and metastatic ability of many malignant tumors. AgRP mutants have been engineered to bind to integrin αvβ3 with high affinity and specificity using directed evolution. Here, AgRP mutants 7C and 6E were radiolabeled with 111In and evaluated for in vivo targeting of tumor integrin αvβ3 receptors. AgRP peptides were conjugated to the metal chelator 1, 4, 7, 10-tetra-azacyclododecane- N, N′, N″, N‴-tetraacetic acid (DOTA) and radiolabeled with 111In. The stability of the radiopeptides 111In-DOTA-AgRP-7C and 111In-DOTA-AgRP-6E was tested in phosphate-buffered saline (PBS) and mouse serum, respectively. Cell uptake assays of the radiolabeled peptides were performed in U87MG cell lines. Biodistribution studies were performed to evaluate the in vivo performance of the two resulting probes using mice bearing integrin-expressing U87MG xenograft tumors. Both AgRP peptides were easily labeled with 111In in high yield and radiochemical purity (>99%). The two probes exhibited high stability in phosphate-buffered saline and mouse serum. Compared with 111In-DOTA-AgRP-6E, 111In-DOTA-AgRP-7C showed increased U87MG tumor uptake and longer tumor retention (5.74 ± 1.60 and 1.29 ± 0.02%ID/g at 0.5 and 24 h, resp.), which was consistent with measurements of cell uptake. Moreover, the tumor uptake of 111In-DOTA-AgRP-7C was specifically inhibited by coinjection with an excess of the integrin-binding peptidomimetic c(RGDyK). Thus, 111In-DOTA-AgRP-7C is a promising probe for targeting integrin αvβ3 positive tumors in living subjects.


Introduction
Molecular imaging is a rapidly evolving field in biomedical research and provides powerful techniques to noninvasively study a variety of important characteristics of cancers, such as tumor metabolism, proliferation, hypoxia, and receptor expression [1][2][3]. Therefore, novel molecular probes that target these characteristics have been under active and intensive development and investigation. Many platforms including small molecules, peptides, proteins, and nanoparticles have been explored in order to develop molecular probes to image a variety of important disease biomarkers [4,5].
Cystine-knot peptides consist of a stable core motif of at least three disulfide bonds that are interwoven into a "knot" conformation. There is great sequence diversity among cystine-knot family members, as only the disulfidebonded core is conserved; consequently, the loops connecting the cysteine residues are highly tolerant to substitution or incorporation of additional amino acid residues [6][7][8]. Previously, we used a truncated form of the Agoutirelated protein (AgRP * ), a 4-kDa cystine-knot peptide with four disulfide bonds and four solvent-exposed loops, as a molecular scaffold for directed evolution. In this study, highthroughput methods were used to screen a yeast-displayed Table 1: Amino acid sequences of the AgRP * mutants 7C and 6E used in this study. DOTA was site-specific conjugated to the Nterminal amine of these peptides.
AgRP * library to identify several mutants with high affinity and specificity for α v β 3 integrin [8].
The cell adhesion receptor integrin α v β 3 is an important tumor angiogenesis factor that determines the invasiveness and metastatic ability of many malignant tumors. It is upregulated on the tumor cell surface and also overexpressed on activated endothelial cells around the tumor tissues [9][10][11]. Many radiolabeled probes have been developed for imaging integrin expression using positron emission tomography (PET) or single-photon emission computed tomography (SPECT) [12][13][14][15]. In our recent research, the 64 Cu-DOTA conjugated AgRP * mutant 7C ( 64 Cu-DOTA-AgRP-7C) was successfully developed for PET imaging of integrin α v β 3 positive tumors. This study demonstrated the potential of AgRP * -based integrin-binding peptides for use as in vivo molecular imaging applications and highlighted cystineknots as promising molecules for further development as diagnostics or therapeutics against different tumor targets [13].
In the current study, to further characterize the in vivo performance of integrin-binding AgRP * -based peptides and also evaluate their potential use for SPECT imaging applications, two DOTA-conjugated AgRP * mutants (DOTA-AgRP-7C and DOTA-AgRP-6E; Table 1) were synthesized and radiolabeled with the gamma emitter 111 In (t 1/2 = 2.8 days). Tumor uptake, biodistribution, and stability of the resulting probes were evaluated in integrin α v β 3 positive human glioblastoma U87MG tumor xenograft models.     111 In-DOTA-AgRP-7C or 111 In-DOTA-AgRP-6E in 400 μL of mouse serum at 37 • C for 2 h. Afterward, the mixture was re-suspended in 0.5 mL of DMF containing 5 μL of Triton X-100 and centrifuged at 16,000 g for 2 min. The supernatant, which contained >95% of the starting radioactivity, was filtered using a 10 K NanoSep device (Pall Corporation, East Hills, NY, USA). Greater than 99% of the radioactive material passed through this filter. The samples were analyzed by radio-HPLC, and the percentage of intact peptide was determined by quantifying peaks corresponding to the intact peptide and to the degradation products.

In Vitro Cell Uptake Assay.
Cell uptake studies were performed as previously described [16]. Briefly, U87MG cells were seeded at a density of 0.2 × 10 6 in 12well tissue culture plates and allowed to attach overnight. After washing three times with serum-free DMEM, cells were incubated with 111 In-DOTA-AgRP-7C or 111 In-DOTA-AgRP-6E (0.5 μCi/well) with or without the integrin-binding peptidomimetic, c(RGDyK), (2 μg/well) at 37 or 4 • C for 15, 60, and 120 min, respectively. Cells were washed three times with chilled PBS containing 0.2% BSA and dissolved by treatment with 0.1 N NaOH at room temperature for 5 min. The cells' suspensions were collected and the resultant radioactivity was measured using a γ-counter (PerkinElmer 1470, Waltham, MA, USA). Cell uptake was expressed as the percentage of added radioactivity. Experiments were performed twice with triplicate wells.

Biodistribution Studies.
All animal studies were carried out in compliance with federal and local institutional rules for animal experimentation. Approximately, 10 7 U87MG cells were suspended in PBS and subcutaneously implanted in the left shoulders of female athymic nu/nu mice. Tumors were allowed to grow to a size of 0.5 cm (2-3 weeks) before imaging experiments were performed. For biodistribution studies, U87MG tumor-bearing mice (n = 3 for each group) were injected with 111 In-DOTA-AgRP-7C (0.259-0.37 MBq, 7-10 Ci) or 111 In-DOTA-AgRP-6E (0.296-0.444 MBq, 8-12 μCi) via the tail vein and sacrificed at different time points (0.5, 2, 24, and 48 h) after injection (p.i.). Tumor and normal tissues of interest were removed and weighed, and their radioactivity was measured with a γ-counter. Radioactivity uptake was expressed as the percent injected dose per gram of tissue (%ID/g). To test the in vivo α v β 3 integrin targeting specificity of the probe, U87MG tumor-bearing mice (n = 3 for each group) were injected via the tail vein with a mixture of 111 In-DOTA-AgRP-7C (0.259-0.37 MBq, 7-10 μCi) and 330 μg of c(RGDyK). The mice were sacrificed at 2 h p.i. and the biodistribution of the radiolabeled peptide was measured as above.

Statistical Analysis.
Quantitative data were expressed as mean ± SD. Means were compared using the Student ttest. A 95% confidence level was chosen to determine the significance between groups, with P values of less than 0.05 indicating significant differences.

3.1.
Radiolabeling. Due to the high thermal stability of cystine-knot peptides, DOTA-AgRP-6E and DOTA-AgRP-7C were easily labeled with 111 InCl 3 by incubation in NH 4 OAc buffer (pH 5.0-5.5) at 80 • C for 45 min. Radiolabeled peptides were purified by PD-10 columns. The radiochemical yield for both peptides was determined by radio-HPLC to be ∼50% and the radiochemical purity as also determined by radio-HPLC was greater than 99%. The specific activity of the peptides was determined to be approximately 0.25 mCi/nmol.

Peptide Stability.
Both 111 In-DOTA-AgRP-7C and 111 In-DOTA-AgRP-6E were radiochemically stable after 24 h in 0.01 M PBS (pH 7.4), as determined by radio-HPLC analysis (Figures 1(a) and 1(c)). Furthermore, radio-HPLC analysis revealed that over 95% of the probes remained intact after 2 h incubation with mouse serum at 37 • C (Figures 1(b) and  1(d)).   Figure 2. During the first 15 min incubation period at 37 • C, 111 In-DOTA-AgRP-7C exhibited rapid cell accumulation, followed by a steady increase in binding and receptor-mediated uptake throughout the experiment. The cell uptake values of 111 In-DOTA-AgRP-7C after 15, 60 and 120 min at 37 • C were 1.70 ± 0.07%, 2.46 ± 0.29%, and 2.76 ± 0.25%, and at 4 • C were 1.05 ± 0.21%, 1.19 ± 0.27%, and 1.26 ± 0.38%, respectively (Figure 2(a)). Thus, a ∼2 fold greater accumulation of probe was observed in cells incubated at 37 • C compared to those incubated at 4 • C, which is indicative of internalization that occurs at physiological temperature. Moreover, cell surface binding and internalization were significantly inhibited by the addition of a large molar excess of the integrin-binding peptidomimetic c(RGDyK) (P < 0.05) (Figure 2(a)). After 1 h, cell uptake of the probe was inhibited 90% and 86% at 37 • C and 4 • C, respectively, demonstrating that the probe specifically targets cell surface integrin receptors.

Biodistribution Studies. The biodistribution of 111
In-DOTA-AgRP-7C was examined in nude mice bearing U87MG human glioblastoma tumors. The results of these experiments are shown in Table 2. Tumor uptake of the probe was 5.74 ± 1.60, 2.35 ± 0.36, 1.29 ± 0.02, and 0.76 ± 0.17%ID/g at 0.5, 2, 24, and 48 h, respectively, indicating relatively high tumor uptake and moderate tumor retention. The probe was found to have rapid blood clearance, with radioactivity levels of 0.68 ± 0.26 and 0.10 ± 0.02%ID/g remaining in the blood after 0.5 and 2 h p.i., respectively. Moreover, whole-body clearance of radioactivity was equally rapid. Except for the kidneys, accumulation in most organs examined were all lower than 1%ID/g at 2 h p.i. Prominent uptake was observed in the kidneys at early time points (33.93 ± 8.35%ID/g at 0.5 h), with decreasing accumulation from 2 to 48 h p.i. These data clearly indicate a renal excretion route and metabolic processing of the probe by the kidneys. With the rapid clearance of the radiotracer from blood and other normal organs, 111 In-DOTA-AgRP-7C exhibited high tumor-to-normal organ ratios in the blood, muscle, lung, liver, spleen, and pancreas (Table 2 and Figure 3(b)). For example, at 2 h p.i., the tumor-to-blood and tumor-tomuscle ratio of 111 In-DOTA-AgRP-7C was 25.8 and 39.8, respectively. To confirm in vivo integrin binding specificity, co-injection of 111 In-DOTA-AgRP-7C with a large molar excess of c(RGDyK) significantly reduced tumor uptake by ∼77% (0.54 ± 0.07 versus 2.35 ± 0.36%ID/g at 2 h p.i., P < 0.05) ( Table 2). Significant differences were also found for probe uptake in normal tissues such as spleen upon addition of c(RGDyK) blocking peptide.
The biodistribution results of 111 In-DOTA-AgRP-6E are summarized in Table 3. Intermediate tumor uptake of 111 In-DOTA-AgRP-6E was observed at 0.5, 2, and 24 h p.i., with values of 1.76 ± 0.34, 1.21 ± 0.21, and 0.89 ± 0.06%ID/g, respectively. The probe also showed rapid blood clearance, with radioactivity levels of 1.27 ± 0.53 and 0.06 ± 0.05%ID/g remaining in the blood after 0.5 and 2 h. In addition, low normal tissue accumulation was seen with the exception of the kidneys (15.37±3.19%ID/g and 14.68±1.98%ID/g at 0.5 and 2 h, resp.). Finally, because of the rapid clearance of the probe from the blood and other organs, 111 In-DOTA-AgRP-6E also displayed high tumor-to-blood and tumor-to-muscle ratios at 2 and 24 h p.i. (Table 3).

Discussion
The native AgRP is a neuropeptide produced in the human brain that plays an important biological role in increasing appetite and decreasing metabolism and energy expenditure [17][18][19]. The C-terminus of AgRP is a cystine-knot peptide which contains a biologically active loop that naturally binds to melanocortin receptors. A truncated form of the AgRP  cystine knot peptide, AgRP * is emerging as a highly attractive platform for developing novel molecular imaging probes [20,21]. AgRP * is small in size (∼4 kDa) and possesses a rigid structure, yet contains four solvent-accessible loops that could be used for mutagenesis [8]. In addition, peptides based on AgRP * are likely to be nonimmunogenic due to their human origin and high thermal and proteolytic stability; Furthermore, AgRP * is amenable to recombinant and synthetic production, which will allow site-specific incorporation of labels or chemical functionality in future studies.
In our previous studies, the AgRP * -based α v β 3 integrin binder AgRP-7C was discovered and used for PET imaging of tumor angiogenesis [8,13]. Excellent in vivo tumor imaging contrast was achieved which demonstrates the success of using AgRP * -based scaffolds for molecular probe development [13]. This study motivated us to further evaluate engineered AgRP * -based integrin targeting peptides for potential SPECT imaging applications. The high affinity binder DOTA-AgRP-7C (IC 50 ∼20 nM) and moderate affinity binder DOTA-AgRP-6E (IC 50 ∼130 nM) were radiolabeled with 111 In and tested for their ability to target tumors in living subjects.
The cystine-knot motif of AgRP * conferred high stability to the radiolabeled peptides. Compared to the reaction temperature (37 • C) used for 64 Cu radiolabeling of AgRP-7C, 6E was performed at 80 • C. The high reaction temperature did not appear to denature the peptides as demonstrated by the cell uptake and biodistribution studies. Similar to 64 Cu-DOTA-AgRP-7C, both 111 In-DOTA-AgRP-7C and 111 In-DOTA-AgRP-6E were found to be stable in PBS buffer for at least 24 h and in mouse serum for 2 h.
Both 111 In and 64 Cu labeled DOTA-AgRP-7C exhibit high renal uptake, which is likely attributed to (1) the long residence time of radiometabolites produced by lysosomal degradation of the radiolabeled peptides within renal cells; (2) the overall positive charge of the peptides [22][23][24][25]. Several strategies previously reported that the accumulation of radiopeptide in the kidneys can be effectively reduced by coinjection of cationic amino acid such as lysine, or polylysine molecules [16,26]. These methods could potentially be explored to reduce the radiation dose to kidney.
Biodistribution studies reveal that 111 In-DOTA-AgRP-7C has significant higher tumor uptake than that of 111 In-DOTA-AgRP-6E at different time points (0.5, 2, and 24 h) (Figure 3(a)). This could be related to the stronger integrin binding affinity of AgRP-DOTA-7C compared to AgRP-DOTA-6E (IC 50 : 22.6 ± 3.9 nM versus 125.5 ± 16.6 nM, resp.) [27]. Both probes display good retention in tumor and low uptakes in most of normal organs (Tables 2 and 3). Interestingly, the kidney uptake of 111 In-DOTA-AgRP-6E was significantly lower than that of 111 In-DOTA-AgRP-7C (14.68 versus 30.59%ID/g at 2 h p.i.), suggesting that modification of the amino acid sequence of the engineered loop of AgRP * may help to optimize in vivo behavior. Because 111 In-DOTA-AgRP-7C showed higher tumor uptake compared to 111 In-DOTA-AgRP-6E, the in vivo integrin targeting specificity of the probe was further confirmed by a blocking study, using a peptidomimetic (c(RGDyK)) that binds to the same epitope on integrin receptors. Compared to the nonblocking group, the tumor uptake of 111 In-DOTA-AgRP-7C is significantly decreased upon co-injection of c(RGDyK) (2.35±0.36 versus 0.54 ± 0.07%ID/g at 2 h, resp.) Table 2 and Figure 3(b), in agreement with the in vitro cell uptake results.
In a previous study, the tumor uptake of 111 In-DOTAc(RGDfK) was reported as 6.28%ID/g at 1 h p.i. in a SKOV3 xenograft model [28]. In comparison, we show that the tumor uptake of 111 In-DOTA-AgRP-7C is 5.74%ID/g at 0.5 h p.i. in a U87MG xenograft model. Importantly, 111 In-DOTA-AgRP-7C and 111 In-DOTA-AgRP-6E were found to have much lower uptake and faster clearance in liver, lung, spleen and other normal tissues compared to 111 In-DOTA-c(RGDfK). The fast clearance of 111 In-labeled AgRP * mutants resulted in high tumor-to-blood and tumorto-muscle ratios at 24 and 48 h p.i. (Tables 2 and 3, Figure 3), demonstrating the advantages of using an AgRP *based scaffold for imaging probe development [28,29].

Conclusions
In summary, we tested the α v β 3 integrin-targeted cystineknot peptides 111 In-DOTA-AgRP-6E and 111 In-DOTA-AgRP-7C as SPECT imaging agents in mouse tumor models. Compared to 111 In-DOTA-AgRP-6E, 111 In-DOTA-AgRP-7C exhibits higher integrin binding affinity and tumor uptake. Moreover, this probe demonstrates rapid tumor uptake, high tumor-to-normal tissue contrast, and favorable pharmacokinetics. These results suggest that AgRP * mutant 7C has potential for clinical translation as a new SPECT imaging agent for integrin α v β 3 positive tumors. Furthermore, cystine-knot peptides are a promising class of molecular scaffolds, and warrant further development and investigation for imaging applications.