Improved Metabolic Stability for ¹⁸F PET Probes Rapidly Constructed via Tetrazine trans-Cyclooctene Ligation

Abstract

The fast kinetics and bioorthogonal nature of the tetrazine trans-cyclooctene (TCO) ligation makes it a unique tool for PET probe construction. In this study, we report the development of an ¹⁸F-labeling system based on a CF₃-substituted diphenyl-s-tetrazine derivative with the aim of maintaining high reactivity while increasing in vivo stability. c(RGDyK) was tagged by a CF₃-substituted diphenyl-s-tetrazine derivative via EDC-mediated coupling. The resulting tetrazine-RGD conjugate was combined with a ¹⁹F-labeled TCO derivative to give HPLC standards. The analogous ¹⁸F-labeled TCO derivative was combined with the diphenyl-s-tetrazine-RGD at μM concentration. The resulting tracer was subjected to in vivo metabolic stability assessment, and microPET studies in murine U87MG xenograft models. The diphenyl-s-tetrazine-RGD combines with an ¹⁸F-labeled TCO in high yields (>97% decay-corrected on the basis of TCO) using only 4 equiv of tetrazine-RGD relative to the ¹⁸F-labeled TCO (concentration calculated based on product’s specific activity). The radiochemical purity of the ¹⁸F-RGD peptides was >95% and the specific activity was 111 GBq/μmol. Noninvasive microPET experiments demonstrated that ¹⁸F-RGD had integrin-specific tumor uptake in subcutaneous U87MG glioma. In vivo metabolic stability of ¹⁸F-RGD in blood, urine and major organs showed two major peaks: one corresponded to the Diels-Alder conjugate and the other was identified as the aromatized analog. A CF₃-substituted diphenyl-s-tetrazine displays excellent speed and efficiency in ¹⁸F-PET probe construction, providing nearly quantitative ¹⁸F labeling within minutes at low micromolar concentrations. The resulting conjugates display improved in vivo metabolic stability relative to our previously described system.

INTRODUCTION

Positron emission tomography (PET) is a powerful and highly sensitive imaging technology with the capacity to observe metabolic processes and track radiolabeled biomolecules in vivo.¹ Of the various positron emitting radionuclides, ¹⁸F finds most extensive use due to its clinically attractive half-life (t_1/2 = 110 min) and high positron efficiency (β⁺ = 99%). To date, clinical applications of PET have largely involved small molecule probes such as ¹⁸F-2-deoxy-2-fluoroglucose.^{2, 3} Currently, there is great interest in the development of peptidic and protein-based probes for ¹⁸F PET imaging, with a correlated need to develop methods for bioligand probe construction. New methods for probe construction must operate efficiently within the constraints of ¹⁸F-labeling chemistry, which include the limited nucleophilicity and short half-life of fluoride and the need to efficiently conjugate molecules at low concentrations relevant to radiochemical experimentation.

A variety of ¹⁸F-labeled synthons have been developed and successfully applied to a host of peptides and proteins for ¹⁸F-PET probe construction. The utility of many ¹⁸F PET probes is hindered by multistep probe syntheses where ¹⁸F is carried through multiple chemical intermediates—a major limitation given the technically demanding nature of ¹⁸F radiochemistry. For the conjugation of ¹⁸F to the biological ligand, peptides or proteins are often used in large excess in order to obtain reasonable yields for ¹⁸F attachment. Additionally, ¹⁸F-tagged proteins or large-peptides are often inseparable from their unlabeled precursors which can compromise the signal through competitive inhibition and result in low specific activity. Furthermore, sacrificing milligram quantities for labeling reactions is impractical for peptides or proteins that are not available in large quantity. Greatly needed are efficient and robust methods for labeling proteins and peptides by ¹⁸F at low concentrations.

In 2008, we described methods for synthesizing trans-cyclooctene derivatives⁴ and applying them in fast bioorthogonal reactions with tetrazines.⁵ With strained TCO derivatives, rate constants of k₂ > 10⁶ M⁻¹s⁻¹ have been measured.^6–9 Contemporaneous with the initial study of TCO, several groups described reactions of tetrazines with derivatives of norbornene¹⁰ or the Reppe anhydride,¹¹ with a measured rate constant of k₂ = 1.9 M⁻¹s⁻¹ at 20 °C in PBS for norbornene conjugation. Recently, cyclopropenes,^{12, 13} cyclooctynes,^{6, 14, 15} and terminal alkenes¹⁶ have also been used as dienophiles for tetrazine ligation. While each of these dienophiles offers complementary advantages, TCO derivatives display the fastest rate constants.

The tetrazine-TCO ligation has become broadly used for research in nuclear medicine including applications in pretargeted imaging,^17–19 and studies have been directed toward optimizing and improving the pharmacokinetics and pharmacodynamics for systems based on dipyridyl-s-tetrazine and monoaryltetrazines.^{7, 17–25} In 2010, Robillard first showed that the tetrazine ligation method could be applied in a pre-targeted antibody using single photon emission computed tomography (SPECT).²³ More recently, Robillard has described factors that contribute to the in vivo stability of TCO’s toward isomerization,⁷ and clearing agents have been developed that improve tumor-to-blood (125 fold) ratios.²² Weissleder has shown that polymer modified tetrazines can be used for in vivo bioorthogonal labeling and PET imaging using an ¹⁸F-labeled TCO derivative.²⁰ More recently, Weissleder and Lewis reported a pretargeting approach for PET imaging based on this method and demonstrated dramatically reduced nontargeted organ uptake.²⁴ Recently, the reaction of a ¹¹C-labeled tetrazine with a TCO derivative was described,²⁶ and Kuntner and Mikula described the development of a ¹⁸F-labeled tetrazine with favorable pharmacokinetic properties.²⁵

In 2010, we developed a radiolabeling method for bioconjugation based on the Diels-Alder reaction between dipyridyl-s-tetrazines and an ¹⁸F-labeled trans-cyclooctene.²⁷ As shown in Figure 1, ¹⁸F-labeled TCO 2 could be obtained in high radiochemical yield (71%) by combining nosylate 1 with ¹⁸F-fluoride (100 mCi). ¹⁸F-2 is an effective reagent for creating ¹⁸F-labeled probes within seconds at low micromolar concentrations, and we have used this reagent to make cyclic RGD (cRGD) and VEGF protein conjugates for cancer imaging^{28, 29} and exendin-4 conjugates for applications in insulinoma imaging and diabetes monitoring.³⁰ Notably, these conjugates were synthesized without using a large excess of the peptidic labeling precursor (Figure 1b). Avoiding excess labeling precursor is critically important for proteins and large polypeptides such as exendin-4 (MW 4775), where the labeled and unlabeled peptide are not readily separable, and unlabeled peptide can significantly decrease signal due to competitive inhibition.³⁰ Weissleder and Lewis have also used ¹⁸F-2 in a number of applications including a recent demonstration of pretargeted imaging.²⁴

Figure 1.

(a) Synthesis of ¹⁸F-TCO 2. (b) Rapid ¹⁸F-labeling of a 4.8 kDa peptide takes place rapidly at low concentration using only a 2-fold excess of the peptide precursor. (c) Probe 4b has been used to image U87MG tumors in mice. (d) A new probe that can also be constructed rapidly and used for U87MG tumor imaging in mice with the added benefit of improved metabolic stability.

An important factor that has not received significant attention is the metabolic stability of tetrazine-TCO conjugates. Dipyridyl-s-tetrazine-based probes with good pharmacokinetic properties have been constructed through conjugation to large peptides (exendin-4)³⁰ or by including PEG-spacers in antibody-pretargeting studies.^{7, 22, 23} We also described that ¹⁸F-trans-cyclooctene 2 undergoes very rapid conjugation with a dipyridyl-s-tetrazine-cRGD construct, and the resulting isomeric conjugates 4b can be used to image tumors in mice (Fig. 1c). Here, a relatively high level of organ uptake in the liver and kidneys was observed, presumably due to the hydrophobic nature of the probe. Because of the high residence time in the liver, we expected that this system would provide a good platform to test and improve the metabolic stability of tetrazine-based probes. In our study on 4b, an attempt to re-isolate radioactive 4b from the major organs, urine and blood of a mouse was not unsuccessful, and only hydrophilic degradation products were observed by radio-HPLC analysis. We hypothesized that the imines of 4b, flanked with electron withdrawing pyridines, may be susceptible to nucleophilic attack and thereby provide a possible handle for degradation. We also hypothesized that a conjugate (Fig 1d) with less electron withdrawing aromatic groups would be more stable. Here, we describe a 3,6-diphenyl-s-tetrazine derivative that displays fast conjugation rates toward ¹⁸F-2 and gives conjugates with improved metabolic stability in an in vivo mouse study. ¹⁸F-labeling yields are discussed and the metabolic stability of the ¹⁸F-2 tagged cRGD conjugate is described. The PET probe was evaluated for integrin α_vβ₃ imaging in U87MG tumor-bearing mice by microPET.

RESULTS

Chemistry

The CF₃-substituted diphenyl-s-tetrazine 6³⁵ was prepared by an improved two-step procedure and elaborated by EDC coupling to cRGD derivative 7 (Fig. 2a) To gauge the reactivity of derivatives of diaryltetrazine 6, stopped flow kinetic analysis was used to measure the rate of the Diels-Alder reaction between tetrazine derivative 8 and equatorially substituted trans-cyclooctene derivative 9, the precursor to ¹⁸F-2 (Fig. 2). The undecaethylene glycol sidechain of 8 was added to enhance water solubility. In water/MeOH (6:4, v/v) at 25 °C, a second order rate constant k₂ = 1000 ± 100 M⁻¹s⁻¹ was measured.

Figure 2.

(a) Synthesis of a cRGD-diphenyl-s-tetrazine conjugate. (b) The rate of the conjugation of 8 with 9 was determined by stopped-flow kinetic analysis. (c) The exponential plot of the reaction of 8 (25 μM) and 9 (1.0 mM) in 40:60 MeOH:water was monitored at 292 nm. Data was recorded for 20 s at 298 K, with triplicate runs on three independent samples at three different concentrations (27 runs total). (d) The average of three observed rates k′ vs concentration of 9 for the reaction between 8 and 9. Under these pseudo-first order conditions the second order rate constant (k₂) was determined by nonlinear regression to be 1000 +/− 100 M⁻¹s⁻¹.

The cycloaddition reaction of tetrazine-cRGD 7 and ¹⁹F-2 provided ¹⁸F-cRGD conjugates 10, which were used as radiolabeling standards and for the integrin receptor binding assay. Consistent with the high reactivity of 8, the pink color of tetrazine-cRGD 7 disappeared immediately upon mixing with ¹⁹F-2. The identity of the ¹⁹F-cRGD conjugates 10 were confirmed by LC-MS. As expected based from prior observation and a model study,^{5, 9, 36} both aromatized conjugates 10b were formed in addition to dihydropyridazine conjugates 10a (Fig. 3a,b). Collectively, we refer to the mixture of aromatic and dihydropyridazine conjugates as 10.

Figure 3.

(a) Conjugation of a cyclic-RGD-tetrazine 7 with F-TCO 2 gives conjugates 10a, which slowly oxidize to aromatic isomers 10b under ambient conditions in aqueous solution. (b) LC/MS analysis of the Diels-Alder conjugate from ¹⁹F–2 acquired after the sample had been allowed to stand overnight shows a mixture of aromatized and more slowly eluting dihydroaromatic products. Chromatographic resolution was higher and retention times longer in this LC/MS run than in radio-HPLC analyses (Fig 4, 5). (c) Cell-binding assay of c(RGDyK) and ¹⁹F-Diels-Alder conjugates 10 using U87MG cells (integrin α_vβ₃-positive human glioblastoma). The cell-binding affinity of the peptides was determined by performing competitive displacement studies with ¹²⁵I-echistatin (n = 3).

Radiochemistry

¹⁸F-labeled trans-cyclooctene (¹⁸F-2) was produced using the protocol developed in our laboratories,²⁷ and utilized in radiolabeling experiments with tetrazine-cRGD derivative 7 (Fig. 4). With only a 4-fold excess of 7 (4 μM) relative to ¹⁸F-2 (1 μM, calculated based on the specific activity of ¹⁸F-10a), a 97% radiochemical yield of ¹⁸F-10a was obtained (Fig. 4b). The specific activity of ¹⁸F-10 was determined to be 3.0 ± 1.0 Ci/μmol after purification by comparing the UV absorption with standard titration curve.

Figure 4.

¹⁸F labeling by ¹⁸F-2 (1 μM) with differing concentrations of 7. The product ¹⁸F-10a is a mixture of regioisomers. (a) HPLC standard of ¹⁸F-2. (b) 97% radiochemical yield with 4 equiv of 2. The specific activity of ¹⁸F-10 was determined to be 3.0 ± 1.0 Ci/μmol after purification by comparing the UV absorption with standard titration curve.

Cell Integrin Receptor-Binding Assay

Receptor-binding affinity studies of ¹⁹F-10 and unmodified c(RGDyK) toward α_vβ₃ integrin were performed using α_vβ₃ integrin–positive U87MG cells. Binding on the cell membrane allows cross-linking and integrin receptor multimerization, through which multivalent binding and clustering of receptor is studied in the natural context of the integrin. We compared the receptor-binding affinity of ¹⁹F-10 with that of unlabeled c(RGDyK) by performing competitive displacement studies with ¹²⁵I-echistatin (Fig. 3c). Both ¹⁹F-10 and unmodified c(RGDyK) peptides inhibited the binding of ¹²⁵I-echistatin to α_vβ₃ integrin–positive U87MG cells. The IC₅₀ value for ¹⁹F-10 and c(RGDyK) was 39.8 ± 4.5 and 19.6 ± 3.2 nmol/L, respectively. Thus, the fluoride incorporation via tetrazine ligation into the cRGD peptide had minimal effect on binding affinity to the α_vβ₃ receptors.

In vivo Metabolism of ¹⁸F-10

The metabolic stability of ¹⁸F-10 was determined in mouse blood, urine and in liver and kidney homogenates at 2 h after tracer injection. The extraction efficiency of all organs was between 56% and 98%. The lowest extraction efficiency was found for the kidney homogenates and the highest extraction efficiency was from blood sample. The intact probes were 75%, 51%, 57%, and 62% for blood, kidney, liver, and urine samples respectively (Fig. 5). The major metabolites correlate well with the aromatized product. These results showed that the new probe displayed significantly higher stability than previous dipyridyl analogs.^{27, 29}

Figure 5.

Metabolic stability of ¹⁸F-10 in mouse blood and urine samples and in liver, and kidney homogenates at 1 h after injection. Fractions were collected every minute and radioactivity measured by γ-counter. The radio-HPLC profile of ¹⁸F-10 standard is also shown. In each of the metabolic extracts, two peaks were observed. One peak corresponded to dihydropyridizine isomers ¹⁸F-10a, the other peak corresponded to isomers of pyridizine ¹⁸F-10b.

microPET Studies

The localization of ¹⁸F-10 in human U87MG tumor-bearing nude mice (n = 5) was performed by multiple time-point static microPET scans. Figure 6a shows microPET images of a female mouse at different times after injection of 7.4 MBq (200 μCi) of ¹⁸F-10. All microPET images were decay corrected. The tumor was clearly visualized with good contrast. Fig. 6b shows the microPET images with a blocking dose of unlabeled c(RGDyK) peptide coinjection. The tumor uptake of the radio probe was clearly reduced. The microPET imaging study demonstrated high and specific binding of ¹⁸F-10 to human U87MG tumors. Quantification of activity accumulation in the tumor and major organs (Fig. 6c) was determined by biodistribution studies conducted 2 h post injection.

Figure 6.

microPET images of athymic nude mice bearing U87MG tumor at 0.5, 1, and 2 h after injection of ¹⁸F-10 (a) without or (b) with a blocking dose of c(RGDyK) peptide (10 mg/kg body weight) (n = 5). Tumors are indicated by arrows. (c) bio-distribution study of ¹⁸F-10 in nude mice bearing U87MG tumor at 2 h p.i..

DISCUSSION

Although dipyridyl-s-tetrazine conjugates with ¹⁸F-2 can be created rapidly and efficiently, in prior studies on RGD-based imaging²⁹ we observed that these conjugates have only moderate metabolic stability in vivo. We hypothesized that a conjugate with less electron withdrawing phenyl groups would be more stable, and in line with prior observations may spontaneously oxidize to give aromatic pyridazine products that are also highly stable.^{5, 9} To test this hypothesis, we prepared the diphenyl-s-tetrazine derivatives 7 and 8 from precursor 6. Compound 8 reacts with trans-cyclooctene derivative 9 in water/MeOH (6:4, v/v) at 25 °C with k₂ = 1000 ± 100 M⁻¹ s⁻¹. When compared under similar conditions, the rate of reactivity for the CF₃-substititued tetrazine 8 falls within an order of reactivity of the faster dipyridyl-s-tetrazine derivatives 3.^{6, 37}

Encouraged by the efficient reactivity of 8 with 9, we reacted the cRGD-diphenyl-s-tetrazine derivative 7 with ¹⁹F-2 to provide conjugate 10a, a mixture of isomers (Fig. 3a). In line with observations from model compounds,³⁶ we found that 10a partially oxidized spontaneously in solution to provide aromatic 10b. Shown in Fig. 3b is the HPLC–MS trace of the Diels-Alder conjugate from 7 (10 μM) and ¹⁹F–2 (10 μM) analyzed after standing overnight in aqueous solution. As expected,^{5, 9} in addition to peaks from the dihydroaromatic Diels-Alder adducts 10a (m/z 1176), we also observed the aromatized pyridizine adducts 10b (m/z 1178). As shown in Fig. 3c, the receptor-binding affinity of 10 was compared to that of unlabeled c(RGDyK) by performing competitive displacement studies with ¹²⁵I-echistatin. The ¹⁹F-cRGD conjugate 10 was comparable to the unlabeled cyclic RGD peptide in the ability to inhibit the binding of ¹²⁵I-echistatin to α_vβ₃ integrin–positive U87MG cells.

To study the stability of the ¹⁸F-labeled Diels-Alder conjugates, an in vivo metabolic study was carried out by injecting ¹⁸F-10 into an athymic nude mouse that was sacrificed 2 h post injection. The organ uptake by the kidneys and liver for ¹⁸F-10 (Fig. 6) is similar to what was observed with dipyridyl-s-tetrazine-RGD construct 4b (Fig. 1c). As shown previously, the organ uptake was greatly reduced when the more hydrophilic probe 4a based on the exendin-4 ligand (Fig. 1b) was used to image a GLP-1R positive tumor in mice.³⁰ Similarly, we anticipate that the pharmacokinetic/pharmacodynamic properties of probes related to 4a will readily be improved with protein-based probes, or with peptide-based probes where hydrophilic spacer molecules are employed. For the present study to investigate the stability of the tetrazine-TCO conjugate in vivo, the organ uptake of ¹⁸F-10 was considered advantageous as it allowed study of probe that had been retained in these organs. Thus, major tissues were collected and homogenized, and the activity was extracted and analyzed by HPLC (Fig. 5). Fractions were collected each minute and radioactivity measured with the γ-counter. The average fraction of intact tracer was significantly improved relative to the first generation system 4b (Fig. 1c), where only degradation products were observed by HPLC in similar attempts to recover radioactivity from blood, urine and organs of the animal. For ¹⁸F-10, a hydrophilic byproduct was not observed by HPLC analysis, and the probe was detected with high fidelity in extracts from the kidneys, liver, blood and urine. To ensure that there was not a hydrophilic byproduct in the homogenates, we also analyzed the aqueous phase from the blood sample. The HPLC profile was very similar to that from the organic phase. In each of the metabolic extracts, two peaks were observed. Upon comparison of the HPLC data (Fig. 5) and LC/MS (Fig. 3b) data with cold conjugates it was concluded that one peak corresponded to the dihydropyridizine isomers of ¹⁸F-10a, and the other peak to the aromatized isomers ¹⁸F-10b. One limitation of using ¹⁸F-2 is the high number of isomeric conjugates that are formed upon conjugation with unsymmetrical tetrazines, which may present an issue for clinical translation. Efforts to ameliorate this issue by using higher symmetry cyclooctene derivatives are in progress.

¹⁸F-10 exhibited good metabolic stability in vivo, and injection of ¹⁸F-10 into a U87MG mouse model resulted in an effective method for α_vβ₃ imaging. The integrin α_vβ₃ receptor specificity was confirmed by blocking experiments, in which unlabeled cRGD was administered prior to the injection of the ¹⁸F-10 (Fig. 6). Thus, this labeling system has improved product stability, and no defluorination of ¹⁸F-10 was observed as no visible bone uptake was observed in any of the microPET scans. We also performed PBS stability study on newly synthesized ¹⁸F-10. Around 28% of product got aromatized at 2 h post incubation (Fig S2).

A major advantage of the tetrazine ligation lies in the ability to enable fast reactivity at low micromolar concentrations within minutes and without an excess of either reactant. After demonstrating that the tetrazine 7 is robust and that conjugates with ¹⁸F-2 have good stability in vivo, we explored the lower limit of concentration for the ¹⁸F labeling reaction. As benchmarks, the decay-corrected labeling yield was 35–45% when N-succinimidyl-4-¹⁸F-fluorobenzoate was combined with an RGD derivative at 0.11 mM,³⁸ and 70% when an RGD derivative at 1.8 mM was labeled with ¹⁸F by Cu-catalyzed azide/alkyne cycloaddition.³⁹ By contrast, the tetrazine–TCO ligation reaction is nearly quantitative at a concentration that is more dilute by more than 3 orders of magnitude (Fig. 4). Thus, a 97% radiochemical yield of ¹⁸F-10 was obtained in 5 minutes when ¹⁸F-2 (4 μCi/μL, 1 μM) was combined with only 4 equiv. of 7 (0.4 μg, 0.4nmol, 4 μM). We believe that this combination of fast reactivity to yield metabolically stable conjugates should continue to enable applications in ¹⁸F-based labeling and imaging.

CONCLUSION

A CF₃-substituted 3,6-diphenyl-s-tetrazine derivative displays fast conjugation rates toward ¹⁸F-2, providing nearly quantitative ¹⁸F labeling was observed within minutes at low micromolar concentrations. This bioorthogonal ligation reaction was used to construct an ¹⁸F-cRGD conjugate, which was evaluated for integrin α_vβ₃ imaging in U87MG tumor-bearing mice by microPET. The conjugate was further shown to display improved metabolic stability in an in vivo mouse study.

ASSOCIATED CONTENT

Experimental procedures, spectral data for all new compounds, kinetic plots, and HPLC traces. This material is available free of charge via the Internet at http://pubs.acs.org.

MATERIALS AND METHODS

All commercially available chemical reagents were used without further purification. The syringe filter and polyethersulfone membranes (pore size, 0.22 μm; diameter, 13 mm) were obtained from Nalge Nunc International (Rochester, NY). ¹²⁵I-Echistatin was purchased from PerkinElmer (Piscataway, NJ). c(RGDyK) was obtained from Peptides International (Louisville, KY). All HPLC conditions are gradient. HPLC methods, NMR spectra and mass spectrometry details are listed in supplementary data. MicroPET scans were performed on a microPET R4 rodent model scanner (Siemens Medical Solutions USA, Inc., Knoxville, TN), or a GE eXplore Vista.

Chemistry

Detailed synthetic procedures and characterization details are provided as supporting material.

Stopped-flow kinetic analysis

The second order rate constant was measured under pseudo-first order conditions using an excess of TCO 9 by following the exponential decay of absorbance due to the tetrazine chromophore of 8 at 292 nm using an SX 18MV-R stopped-flow spectrophotometer (Applied Photophysics Ltd.). Thus, equal volumes of solutions of TCO 9 (0.50 mM, 1.0 mM or 2.0 mM in 60:40 water : methanol) and tetrazine 8 (0.050 mM in 60:40 water : methanol) were mixed in the stopped flow device. The final concentration of 8 was 25 μM, and 9 was 0.25 mM, 0.50 mM or 1.0 mM. At each concentration, kinetic data was repeated nine times (triplicate runs on three independent samples) at 298 K. Thus, 27 rate measurements were made. The rate constant was determined by nonlinear regression analysis using Prism (GraphPad Software, Inc.). The mean second order rate constant under these conditions was measured to be 1000 +/− 100 M⁻¹s⁻¹.

Radiochemistry

The ¹⁸F-labeled TCO (¹⁸F-2, Figure 1) was synthesized as reported.²⁷ A solution of 0.4 mCi (14.8 MBq) ¹⁸F-2 in ethanol was added to different concentrations of tetrazine-cRGD 7 in DMSO (total volume 100 μL). After vigorous vortexing for 1 min at room temperature, the reaction was quenched with 1 mL 0.1% TFA in water and loaded onto a C-18 HPLC column to determine the labeling yield. With a loading of 7 at 4μM, probe 10 was obtained with 97% labeling yield (Fig. 4). For small animal study, the HPLC fraction containing ¹⁸F-10 was collected and the HPLC eluent was removed using a rotary evaporator. ¹⁸F-10 was reconstituted in 1 mL PBS and passed through a 0.22 μm syringe filter for animal injection.

Cell Integrin Receptor-Binding Assay

In vitro integrin-binding affinities and specificities of tetrazine-cRGD peptides were assessed via displacement cell binding assays using ¹²⁵I-echistatin as the integrin-specific radioligand. Experiments were performed on the human glioblastoma U87MG cell line by modification of a method previously described.³¹

Animal Models

Animal procedures were performed according to a protocol approved by the UNC Institutional Animal Care and Use Committee. Human brain cancer carcinoma xenografts were induced by subcutaneous injection of 10⁷ U87MG cells into the right front leg of female athymic nude mice. Three weeks after inoculation of the tumor cells, when the tumor reached 0.4–0.6 cm in diameter, the mice were used for microPET experiments.

Metabolic Stability

The metabolic stability of ¹⁸F-10 was evaluated in an athymic nude mouse bearing a U87MG tumor according a reported procedure.³² Detailed procedures are included as supplementary material.

microPET Studies

PET of tumor-bearing mice was performed on an eXplore Vista microPET/CT rodent model scanner using a reported method.^{33, 34} In brief, the mice were injected with 7.4 MBq of ¹⁸F-RGD conjugate 10 with or without a blocking dose of unlabeled RGD peptide via the tail vein and then anesthetized with 2% isoflurane and placed near the center of the FOV of the microPET where the highest image resolution and sensitivity are obtained.