Noninvasive Imaging of Islet Transplants with ¹¹¹In-Exendin-3 SPECT/CT

Radiolabeling

[Lys⁴⁰(DTPA)]exendin-3 (Peptides Specialty Laboratories) (Fig. 1) (referred to as exendin-3 in the remainder of the text) was radiolabeled as described previously (14). Briefly, 150 MBq of ¹¹¹InCl₃ were added to 1 μg of exendin-3 dissolved in 0.1 M 2-(N-morpholino)ethanesulfonic acid (MES), pH 5.5 (Sigma Aldrich), and incubated for 20 min at room temperature. After incubation, ethylenediaminetetraacetic acid (Sigma Aldrich) and polysorbate 80 (Sigma Aldrich) were added to a final concentration of 5 mM and 0.1%, respectively. The radiochemical purity of ¹¹¹In-exendin-3 was determined by instant thin-layer chromatography–silica gel (Agilent Technologies), using 0.1 M NH₄Ac (Sigma Aldrich) in 0.1 M ethylenediaminetetraacetic acid, pH 5.5. The reaction mixture was purified by solid-phase extraction using a hydrophilic–lipophilic balance cartridge (30 mg, Oasis; Waters) to remove unlabeled ¹¹¹In. The cartridge was activated with 1 mL of ethanol, washed with 2 mL of water, and conditioned with 1 mL of 0.1 M MES, pH 5.5. Subsequently, the labeling mixture was loaded onto the cartridge and washed with 1 mL of 0.1 M MES and 2 mL of water. ¹¹¹In-exendin-3 was eluted from the cartridge with 200 μL of 100% ethanol. For injection, the purified labeling solution was diluted in 0.5% phosphate-buffered saline/bovine serum albumin (w/v) to obtain a final ethanol concentration of less than 10%.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Chemical structure of [Lys⁴⁰(DTPA)]exendin-3.

Animals

Six- to 8-wk-old female WAG/Rij rats were purchased from Charles River Laboratories. WAG/Rij rats were used both as islet donor and as recipient. The animal experiments were approved by the animal welfare committee of the Radboud University Nijmegen.

Islet Isolation

Pancreatic islets were isolated from female WAG/Rij rats by collagenase digestion. The rats were euthanized by CO₂/O₂ suffocation, the abdominal cavity was opened, and a canula was placed in the common bile duct. The pancreas was perfused with 8 mL of RPMI medium R0883 (Gibco/BRL Thermo Fisher Scientific) containing collagenase type V (1 mg/mL; Sigma Aldrich). Subsequently, the pancreas was dissected and kept in ice-cold RPMI containing collagenase until digestion. The pancreata were digested for 11.5 min at 37°C. Digestion was stopped by addition of complete RPMI medium (RPMI medium supplemented with 10% fetal calf serum [HyClone; Celbio], 2 mM l-glutamine [Sigma Aldrich], and penicillin and streptomycin [100 U/mL; Sigma Aldrich]). The digested tissue was washed twice in complete RPMI medium (2 min at 210g) and passed through a 500-μm mesh. Subsequently, the islets were purified on a discontinuous Ficoll gradient (densities of 1.108, 1.096, and 1.037 g/mL; Cellgro [Mediatech Inc.]) by centrifugation at 625g for 16 min (without brake). The islets were collected from the interface of the second and third layers. Finally, the remaining Ficoll was removed by washing twice with complete RPMI medium (2 min at 210g). The purified islets were cultured overnight at 37°C with 5% CO₂ in complete RPMI medium. Both islet viability and islet purity exceeded 90% after the isolation procedure.

In Vitro Function Assay

Islet function after isolation was assessed by a static glucose incubation assay. After isolation and overnight recovery, islets were collected, counted, and resuspended in complete RPMI. Islets were seeded in a 24-well plate (Corning Inc.) at 30 islets per well (n = 5) and preincubated for 1 h (37°C, 5% CO₂) in 600 μL of Krebs buffer (115 mM NaCl, 24 mM NaHCO₃, 5 mM KCl, 2.2 mM CaCl₂, 1 mM MgCl₂, 20 mM C₈H₁₈N₂O₄S, and 0.2% bovine serum albumin [Sigma Aldrich]) supplemented with 1.67 mM d-glucose (Sigma Aldrich). After 1 h, the incubation buffer was replaced by 600 μL of fresh Krebs buffer supplemented with 1.67 mM d-glucose. Subsequently, islets were incubated in Krebs buffer supplemented with 16.7 mM d-glucose and 1.67 mM d-glucose, 1 h for both conditions. After each incubation step, the supernatant was collected and samples were stored at −20°C until insulin determination by an insulin enzyme-linked immunosorbent assay (Mercodia AB), performed according to the manufacturer’s protocol.

In Vitro Binding Assay

After overnight recovery, islets were collected, counted, and resuspended in complete RPMI. After transwell saturation, using binding buffer (RPMI containing 0.5% bovine serum albumin [v/w]), islets were transferred to 24-well transwell plates (50, 100, 200, or 400 islets per transwell, n = 3 per condition) (Corning Inc.) and incubated in 500 μL of Krebs buffer for 30 min at 37°C. Subsequently, islets were washed with 500 μL of binding buffer and approximately 2 kBq of ¹¹¹In-exendin-3 were added, followed by incubation for 4 h at 37°C. To determine whether the binding was GLP-1R–mediated, 1 μg of unlabeled exendin-3 was added together with the radiolabeled exendin-3 in separate wells (n = 3 per condition). After incubation, islets were washed 5 times using binding buffer and islet-associated radioactivity was measured in a well-type γ-counter (Wallac 1480 Wizard; Perkin Elmer).

Transplantation

After overnight recovery, isolated islets were collected, spun down at 15g for 2 min, and resuspended in RPMI without fetal calf serum. Recipient rats were anesthetized with isoflurane (induction, 4%–5%; maintenance, 1.5%–2% in 50% O₂ and 50% air) (Abbott Laboratories), and the right hind leg was shaved and disinfected with povidone–iodine (Meda Pharma B.V.) before islet inoculation. Islets were aspirated into a catheter line and infused manually in the calf muscle at low speed (injected volume, 100 ± 10 μL) using a syringe (Hamilton Co.). Rats were transplanted with either 400 (n = 6) or 800 (n = 6) islets.

SPECT Acquisition and Biodistribution

Four weeks after transplantation, rats were injected with 15 ± 0.4 MBq of ¹¹¹In-exendin-3 via the tail vein (peptide dose, 0.1 μg in 200 μL of phosphate-buffered saline, 0.5% bovine serum albumin). One hour after injection of the radiolabeled exendin-3, rats were anesthetized with isoflurane (induction, 4%–5%; maintenance, 1.5%–2% in O₂ and air) and SPECT/CT was acquired using a dedicated small-animal SPECT/CT scanner (U-SPECT-II; MILabs). SPECT images were acquired with a 1.0-mm pinhole general-purpose rat and mouse collimator, 36 bed positions, and a total acquisition time of 50 min. CT images were acquired for anatomic information (65 kV, 615 μA, 1 bed position, 160-μm spatial resolution).

After the SPECT/CT acquisition (2 h after injection), rats were euthanized by CO₂/O₂ suffocation and the muscle containing the transplant, as well as other relevant tissues, were dissected, weighed, and measured in a γ-counter (Wallac 1480 Wizard). Muscles were fixed in 4% formalin (w/v).

SPECT Reconstruction and Quantitative SPECT Analysis

The SPECT images were reconstructed using the U-SPECT software (U-SPECT-Rec; MILabs) with the following settings: ordered-subsets expectation maximization, 1 iteration, 16 subsets, and a voxel size of 0.75 mm³.

Uptake of ¹¹¹In-exendin-3 in the transplant was quantified using Inveon Reconstruction Workplace (Siemens Healthcare). A volume of interest was placed over the transplant. Total voxel intensity in the islet transplant was corrected for the background by subtracting the background signal in the contralateral control muscle using an identical volume of interest and the injected activity.

Immunohistochemical Analysis

The formalin-fixed muscle containing the transplant was dehydrated and embedded in paraffin. Four-micrometer sections were cut and stained with hematoxylin and eosin for localization of the transplant. Consecutive sections were stained for the presence of insulin, as described previously (15), and for the presence of GLP-1R.

The GLP-1R staining was performed as follows: antigen retrieval was performed in 10 mM sodium citrate, pH 6.0, for 10 min at 99°C. Subsequently, sections were incubated for 10 min with 3% H₂O₂ in phosphate-buffered saline at room temperature in the dark, to block endogenous peroxidase activity. Nonspecific binding was blocked by incubation for 30 min with 5% swine serum. The primary anti-GLP-1R antibody (ab39072; Abcam) was diluted in phosphate-buffered saline, 1% bovine serum albumin (1:1,000). Primary antibody incubation for 90 min was followed by incubation with swine-antirabbit peroxidase (1:100) (p0271; DAKO). Finally, Bright-DAB (BS04500; Immunologic BV) was used to visualize peroxidase activity. Microautoradiography to visualize tracer accumulation was performed on 4-μm sections as described by Brom et al. (15).

Sections stained for insulin and GLP-1R were analyzed using a DM5000 microscope (Leica), and color images were obtained with a camera (Evolution MP [Leica], using AxioVision software, version 4.4 [Zeiss]). To determine transplant volume, the insulin-positive area of the transplant per section was multiplied by the intersection distance. Three animals were excluded from the analysis because no transplant was found immunohistochemically.

Statistical Analysis

All mean values are expressed as mean ± SD. Statistical analysis was performed using the unpaired 2-tailed t test, and Pearson correlation coefficients were calculated using Prism (version 5.03; GraphPad Software, Inc.). The level of significance was set at a P value of less than 0.05.