In Vitro and In Vivo Characterization of Three 68Ga- and 111In-Labeled Peptides for Cholecystokinin Receptor Imaging

Cholecystokinin (CCK) receptors are overexpressed in several human tumor types, such as medullary thyroid carcinomas and small cell lung cancers. Several ligands for the CCK2 receptor (CCK2R) have been developed for radionuclide targeting of these tumors. In this study, we evaluated whether radiolabeled DOTA-sCCK8 and its stabilized derivative, DOTA-sCCK8[Phe2(p-CH2SO3H), Nle3,6], are suitable for imaging of CCK2R-positive tumors, using DOTA-MG0 as a reference. In vivo targeting of CCK2R-positive tumors with DOTA-sCCK8, DOTA-sCCK8[Phe2(p-CH2SO3H), Nle3,6], and DOTA-MG0, labeled with 111In or 68Ga, was evaluated in BALB/c nude mice with a subcutaneous A431-CCK2R tumor. Biodistribution studies and single-photon emission computed tomography (SPECT) and positron emission tomography (PET) were performed at 1 hour postinjection. All peptides specifically accreted in the CCK2R-expressing tumors. Both 111In-DOTA-sCCK8 and 111In-DOTA-sCCK8[Phe2(p-CH2SO3H), Nle3,6] showed good tumor retention (4.65% ID/g and 5.44% ID/g, respectively, at 4 hours postinjection). On PET/computed tomographic (CT) and SPECT/CT scans, subcutaneous A431-CCK2R tumors were clearly visualized with low uptake of sCCK8 peptides in the intestines. Whereas radiolabeled DOTA-MG0 showed high kidney uptake (70% ID/g), the sCCK8 peptides showed low uptake in the kidneys. Sulfated CCK8 analogues combined high tumor uptake with low retention in the kidney and are therefore promising tracers for imaging of CCK2R-positive tumors.

T HE DEVELOPMENT OF RADIOPHARMACEUTI- CALS for diagnostic and therapeutic purposes based on regulatory peptides is of particular interest in oncology as many receptors are overexpressed in different types of human cancers, for example, somatostatin receptors, gastrin-releasing peptide receptors, and gastrin receptors.In this study, we focused on the cholecystokinin receptor (CCKR) as a target for peptide receptor radionuclide imaging and therapy.The CCK2 receptor (CCK2R) is abundantly expressed in most medullary thyroid carcinomas and small cell lung cancers, as well as in stromal ovarian cancers and astrocytomas. 1n earlier studies, a gastrin analogue (minigastrin 5 MG0) was explored as a therapeutic agent for radionuclide targeting of CCK2R-positive tumors.MG0 is a peptide that binds selectively to the CCK2R and has a high tumor uptake in CCK2R-positive tumors.However, MG0 also displays high kidney uptake, which may hamper visualization of tumors in the vicinity of the kidneys but is especially a problem when using this peptide for therapeutic applications. 1,2As an alternative to MG0, sCCK8 peptides display affinity for both the CCK1 receptor (CCK1R) and the CCK2R, whereas uptake in the kidneys is relatively low.A disadvantage of sCCK8 compared to MG0 is that sCCK8 contains an easily hydrolyzable sulfated tyrosine (Tyr) residue.Apart from that, the sCCK8 sequence contains two methionines (Met) that are prone to oxidation during radiolabeling and potentially in vivo. 3,4In our previous studies, we showed that the stability of sCCK8 can be increased by replacing the Tyr sulfate moiety by phenylalanine sulfonate (Phe(p-CH 2 SO 3 H) and that the Met residues can be replaced by norleucine (Nle) with minimal impact on the receptor affinity. 5n this study, we compared the in vitro and in vivo properties of DOTA-sCCK8, DOTA-sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ], and DOTA-MG0.The peptides were labeled with either 111 In or 68 Ga and evaluated in terms of 50% inhibitory concentration (IC 50 ) and biodistribution.Targeting of CCK2R-positive tumors with the radiolabeled peptides was visualized by micro-single-photon emission computed tomography (SPECT)/computed tomography (CT) ( 111 In-labeled peptides) and micro-positron emission tomography (PET)/CT ( 68 Ga-labeled peptides).

Cell Culture
In these studies, A431 cells transfected with CCK2R and A431 cells transfected with a control transfectant were used.Cells were constructed as described previously. 6Cells were maintained in Dulbecco's Modified Eagle's Medium (DMEM) with 4.5 g/L D-glucose (Gibco, Invitrogen, Breda, the Netherlands) supplemented with 10% fetal calf serum and 250 mg/mL G418 (Geneticin) in a humidified 5% CO 2 atmosphere at 37uC.The cells were harvested by trypsinization with trypsin/EDTA.

IC 50 Determination
The apparent IC 50 of the peptides labeled with 115 In or 69 Ga for binding the CCK2R was determined on A431-CCK2R cells.Cells were grown to confluency in six-well plates.Cells were washed with binding buffer (DMEM supplemented with 0.5% w/v bovine serum albumin [BSA]) and incubated at room temperature for 10 minutes in binding buffer.Subsequently, peptide labeled with 115 In or 69 Ga was added at final concentrations ranging from 0.1 to 1,000 nM.Then a trace amount of 111 In-DOTA-sCCK8 (40,000 cpm/well) was added.After incubation at room temperature for 2 hours, binding buffer was removed, cells were washed twice with binding buffer, and cell-associated radioactivity was determined.The apparent IC 50 was defined as the peptide concentration at which 50% of binding without competitor was reached.IC 50 values were calculated using GraphPad Prism software version 4.00 for Windows (GraphPad Software, San Diego, CA).

Biodistribution Studies of 111 In-Labeled Peptides
Tumor targeting of 111 In-labeled DOTA peptides was investigated in female athymic BALB/c nude mice.Subcutaneous tumors were induced by inoculation with CCK2R-expressing A431 cells.Mice were inoculated with 2 3 10 6 A431-CCK2R cells in the right flank and 2 3 10 6 A431-mock cells in the left flank.When tumors had reached a weight of approximately 0.2 g (as measured by caliper measurements), mice were divided into groups (n 5 5/group), and 185 kBq of 111 In-DOTA-peptide (0.05 mg, 0.03 nmol) was injected intravenously.Specificity was studied in groups that received a 1,000-fold molar excess of unlabeled peptide.Groups of mice were killed 1 hour and 4 hours postinjection, a blood sample was drawn, and organs of interest were dissected, weighed, and counted in a gamma-counter along with a standard of the injected activity to allow calculation of the injected dose per gram of tissue (%ID/g).Animal experiments were approved by the local animal welfare committee and carried out according to national regulations.

Biodistribution Studies of 68 Ga-Labeled Peptides
Tumor targeting of 68 Ga-labeled DOTA peptides was investigated as described above for the 111 In-labeled peptides.In this study, 2 MBq of 68 Ga-DOTA-peptide (0.05 mg, 0.03 nmol) was injected intravenously after the tumors had reached a weight of 0.

IC 50 Values
Table 2 shows the apparent IC 50 values of the labeled peptides as determined in competitive binding assays.DOTA-sCCK8, labeled with 115 In or 69 Ga, displayed the highest affinity (0.77 and 0.83 nM, respectively), whereas the synthetic peptide sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ] showed the lowest affinity (16.72 and 16.54 nM, respectively).No significant differences in IC 50 values were found between 69 Ga-labeled peptide and 115 In-labeled peptides.

Biodistribution Studies of Radiolabeled DOTA Peptides
The in vivo characteristics of the radiolabeled peptides were studied in mice with A431-CCK2R and A431-mock tumors.The biodistribution of the DOTA peptides, labeled with 68 Ga and 111 In, for in vivo targeting of CCK2R-expressing tumors was determined at 1 hour ( 68 Ga and 111 In) and 4 hours ( 111 In) postinjection.The biodistribution profile for the 111 In-labeled peptides is summarized in Figure 1.
In-DOTA-MG0 showed a higher uptake in the tumor (12.5 6 2.88% ID/g at 1 hour postinjection) than both sCCK8 peptides.For both sCCK8 peptides, specific uptake was also found in the pancreas and the stomach, presumably due to expression of the CCK1 receptor in these organs in rodents.
In Figure 2, the biodistribution profile of the 68 Galabeled peptides is summarized.

SPECT and PET
Subcutaneous A431-CCK2R tumors were clearly visualized on SPECT images 1 hour after injection of 111 In-labeled DOTA peptides (Figure 3).For anatomic correlation, the SPECT images were fused with CT images.The A431mock tumor did not show uptake with any of the peptides and thus was not visualized.In-DOTA-sCCK8 showed some uptake in the abdomen, presumably due to accretion of the peptide in the pancreas.With 111 In-DOTA-sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ], the stomach was also visualized.Tumor uptake of the synthetic peptide was somewhat lower than that of   3,6 ], and (C) DOTA-MG0.Radiotracer uptake is clearly visible in A431-CCK2R tumors (arrow), whereas no uptake is observed in the mock-transfected A431 tumors (arrowhead).CCK2R to mock tumor ratios were 12.2, 7.9, and 27.5 for DOTA-sCCK8, DOTA-sCCK8 [Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ], and DOTA-MG0, respectively.
the other two peptides.In line with the biodistribution results, pronounced uptake of 111 In-DOTA-MG0 was observed in the kidneys.
PET also allowed for clear visualization of the A431-CCK2R tumors 1 hour after injection of 68 Ga-labeled DOTA peptides (Figure 4).Besides tumor uptake, both sCCK8 and sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ] showed intestinal uptake, which was most pronounced for sCCK8.Excretion of 68 Ga-DOTA-sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ] was observed in the bladder.As was seen in the biodistribution studies and on the PET images, 68 Ga-DOTA-MG0 showed high uptake in the kidneys.

Discussion
In earlier studies, it was found that various gastrin analogues (such as MG0) show good tumor targeting, accompanied by high kidney uptake. 7,8In contrast, sCCK8 peptides have shown low kidney uptake and better tumor to kidney ratios. 5,9In this study, we have characterized the in vitro and in vivo properties of CCKR binding peptides, MG0 and two sCCK8 peptides.DOTA-MG0 was used as a reference peptide because this peptide has been studied most extensively. 2,3,7,10,11To this end, the peptides were labeled with 111 In (gamma emitter) and 68 Ga (positron emitter) and evaluated for detection of CCK2R-positive tumors with SPECT and PET, respectively.
Competitive binding assays on A431-CCK2R cells revealed that all evaluated compounds have affinities for the CCK2R in the low nanomolar range.Labeled DOTA-sCCK8 had the highest affinity, whereas the synthetic peptide DOTA-sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ] showed a significantly lower affinity.However, labeled with 68 Ga, the tumor uptake of both compounds was similar.The biodistribution studies with the 111 In-labeled peptides showed that 4 hours postinjection, tumor uptake was maintained for DOTA-sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ], whereas tumor uptake of the DOTA-sCCK8 and DOTA-MG0 decreased over time.This improved tumor retention of DOTA-sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ] may be due to the enhanced resistance against oxidation of this peptide compared to DOTA-MG0 and DOTA-sCCK8.
PET is advantageous over conventional scintigraphy and SPECT because of its high sensitivity in combination with improved spatial resolution.However, SPECT tracers are less expensive than PET tracers because different gamma-emitting nuclides can be used, which are mostly longer-lived and more easy to obtain than PET tracers.
On the PET images in this study, tumors were visualized with higher contrast than on the SPECT images.Both SPECT and PET images showed uptake of radiolabeled sCCK8 peptides in the intestines, presumably due to the accretion of tracer in the pancreas.
Earlier research showed that 68 Ga-DOTA-MG0 is a suitable PET tracer for the detection of CCK2R-positive tumors. 10However, the high kidney uptake might hamper the visualization of tumors in the vicinity of the kidneys.In this study, high kidney uptake is confirmed for DOTA-MG0 by PET, by SPECT, and in the biodistribution studies.This may be due to tubular reabsorption of the peptide via scavenger receptors. 7,11Be ´he ´and colleagues showed that kidney uptake of radiolabeled DOTA-MG0 could be reduced by coadministration of polyglutamic acids. 7Recently, it was found that replacement of the penta-L-glutamate sequence by five D-glutamic acid or Dglutamine residues significantly lowered the kidney uptake, without affecting the tumor uptake. 12,13In addition, a peptide with only 3 N-terminal D-Gln also displayed a strongly reduced kideny retention, but its in vitro serum stability was less than that of the other analogues. 13The sCCK8 peptides also accumulated specifically in the pancreas.This specific uptake can be explained by the fact that in rodents, the pancreas abundantly expresses CCK1R and CCK2R, whereas in human pancreatic tissue, CCK2R is not expressed.Therefore, in a clinical setting, pancreatic uptake of radiolabeled sCCK8 analogues is not expected. 5,9Given that both sCCK8 peptides show low kidney uptake and good tumor retention, they might be promising tracers for imaging of CCK2R-positive tumors.

Conclusion
All three peptides showed CCK2R affinity in the low nanomolar range and good tumor uptake in CCK2Rpositive tumors.Both 111 In-labeled and 68 Ga-labeled DOTA peptides showed high uptake in subcutaneous A431-CCK2R tumors.Subcutaneous tumors were clearly visualized with microPET after injection of 68 Ga-labeled DOTA peptides.In this study, DOTA-sCCK8 combines a high affinity with high tumor uptake and low kidney uptake.The synthetic peptide DOTA-sCCK8[Phe 2 (p-CH 2 SO 3 H), Nle 3,6 ] showed the best tumor retention, but the uptake is lower.Radiolabeled DOTA-MG0 showed the highest uptake in the tumor, but its high kidney uptake is disadvantageous.Our findings may serve as a stimulus to perform more research on the potential of sCCK8 analogues in visualizing CCK2R-positive tumors.

Figure 2 .
Figure 2. Biodistribution of 68 Galabeled peptides in BALB/c nude mice bearing CCK2R-expressing tumors.Values are expressed as a percentage of the injected dose per gram of tissue (n 5 5 mice/group).Blocking was performed by coinjection of a 1,000fold molar excess of unlabeled DOTA-sCCK8 (n 5 3 mice/group).Mice were dissected 1 hour postinjection.
resolution 160 mm, 40 kV, 612 mA) for anatomic reference.At 4 hours postinjection, the mice were euthanized and scanned, and after the scan was recorded, the uptake of111In-labeled peptide in dissected tissue was determined as described for the biodistribution study.Scans were reconstructed with software from MILabs, which uses an ordered-subset expectation maximization algorithm, with a voxel size of 0.375 mm.Scans were analyzed Inveon Research Workplace software version 3.0 (Preclinical Solutions, Siemens Healthcare Molecular Imaging, Knoxville, TN).
4 hours SPECT images were acquired with a U-SPECT-II/ CT scanner (MILabs, Utrecht, the Netherlands).Mice were scanned in the prone position under general anesthesia (isoflurane and O 2 ) for 30 to 60 minutes using a 1.0 mm diameter pinhole rat collimator tube, followed by a CT scan (spatial When the tumors had a diameter of 2 to 5 mm, the mice were injected intravenously with 4.2 to 5.4 MBq of 68 Ga-DOTA peptide (50-58 MBq/nmol).Mice were euthanized by CO 2 /O 2 suffocation and were scanned in the prone position.PET images were acquired for 30 minutes, followed by a CT scan (spatial resolution 113 mm, 80 kV, 500 mA).Scans were reconstructed using Inveon Acquisition Workplace software version 1.5 using an ordered-set expectation maximization three-dimensional maximum a posteriori algorithm with the following parameters: matrix, 256 3 256 3 159; pixel size, 0.43 3 0.43 3 0.8 mm 3 ; and b value of 1.5 with uniform variance.