OtherBASIC SCIENCE INVESTIGATIONS

Internalization of sst2, sst3, and sst5 Receptors: Effects of Somatostatin Agonists and Antagonists

Renzo Cescato, Stefan Schulz, Beatrice Waser, Véronique Eltschinger, Jean E. Rivier, Hans-Jürgen Wester, Michael Culler, Mihaela Ginj, Qisheng Liu, Agnes Schonbrunn and Jean Claude Reubi
Journal of Nuclear Medicine March 2006, 47 (3) 502-511;

Abstract

The uptake of radiolabeled somatostatin analogs by tumor cells through receptor-mediated internalization is a critical process for the in vivo targeting of tumoral somatostatin receptors. In the present study, the somatostatin receptor internalization induced by a variety of somatostatin analogs was measured with new immunocytochemical methods that allow characterization of trafficking of the somatostatin receptor subtype 2 (sst2), somatostatin receptor subtype 3 (sst3), and somatostatin receptor subtype 5 (sst5) in vitro at the protein level. Methods: Human embryonic kidney 293 (HEK293) cells expressing the sst2, sst3, or the sst5 were used in a morphologic immunocytochemical internalization assay using specific sst2, sst3 and sst5 antibodies to qualitatively and quantitatively determine the capability of somatostatin agonists or antagonists to induce somatostatin receptor internalization. In addition, the internalization properties of a selection of these agonists have been compared and quantified in sst2-expressing CHO-K1 cells using an ELISA. Results: Agonists with a high sst2-binding affinity were able to induce sst2 internalization in the HEK293 and CHO-K1 cell lines. New sst2 agonists, such as Y-DOTA-TATE, Y-DOTA-NOC, Lu-DOTA-BOC-ATE (where DOTA is 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid; TATE is [Tyr3, Thr8]-octreotide; NOC is [1-NaI3]-octreotide; and BOC-ATE is [BzThi3, Thr8]-octreotide), iodinated sugar-containing octreotide analogs, or BIM-23244 were considerably more potent in internalizing sst2 than was DTPA-octreotide (where DTPA is diethylenetriaminepentaacetic acid). Similarly, compounds with high sst3 affinity such as KE108 were able to induce sst3 internalization. In sst2- or sst3-expressing cell lines, agonist-induced receptor internalization was efficiently abolished by sst2- or sst3-selective antagonists, respectively. Antagonists alone had no effect on sst2 or sst3 internalization. We also showed that somatostatin-28 and somatostatin-14 can induce sst5 internalization. Unexpectedly, however, potent sst5 agonists such as KE108, BIM-23244, and L-817,818 were not able to induce sst5 internalization under the same conditions. Conclusion: Using sensitive and reproducible immunocytochemical methods, the ability of various somatostatin analogs to induce sst2, sst3, and sst5 internalization has been qualitatively and quantitatively determined. Whereas all agonists triggered sst2 and sst3 internalization, sst5 internalization was induced by natural somatostatin peptides but not by synthetic high-affinity sst5 agonists. Such assays will be of considerable help for the future characterization of ligands foreseen for nuclear medicine applications.

Interest in somatostatin and somatostatin analogs is increasing, largely because of the success of in vivo targeting of somatostatin receptors in tumors (1). In this clinical application, not only is binding of the radiolabeled somatostatin analogs to the receptor important but also internalization of the receptor–ligand complex for successful in vivo targeting of tumoral peptide receptors using radiopeptides (1,2). Therefore, during the course of optimal development of new radiopeptide analogs for in vivo receptor targeting, peptides need to be tested not only for receptor binding and biodistribution but also for their receptor internalization properties.

Most of the internalization studies performed with radiopeptides, including radiolabeled somatostatin analogs, have been done with methods that measure internalization of the radioligand but not of the receptor itself (3–8). Although such methods give a good indication of the internalization capability of a given radioligand, they are not always easy to interpret because of the extremely complex mechanisms of intracellular receptor trafficking and intracellular processing of the internalized radioligand (9); furthermore, in these radioligand internalization studies, receptor internalization can be quantitated only at subsaturating ligand concentrations, rather than at a large range of agonist concentrations and receptor occupancies. Finally, the role of agonists versus antagonists in the internalization process could not be thoroughly investigated in the previous studies: Although there is a consensus that antagonists generally do not trigger the internalization of G-protein–coupled receptors (10), examples exist of peptide receptor antagonists that do stimulate internalization, such as cholecystokinin-, 5-HT2A-, endothelin-, and neuropeptide Y–analogs (10–13). In the somatostatin receptor field, a recent report indicated that somatostatin receptor agonists, but not somatostatin receptor antagonists, are able to internalize the somatostatin receptor subtype 2 (sst2) (14). Many studies describing new radiopeptides for in vivo targeting do not give experimental evidence of whether these radioligands are agonists or antagonists.

Somatostatin action is mediated by 5 somatostatin receptors (15). However, not all will equally internalize on agonist binding (15,16). sst2, somatostatin receptor subtype 3 (sst3), and somatostatin receptor subtype 5 (sst5) are internalized to a much higher extent than is somatostatin receptor subtype 1 (sst1) or somatostatin receptor subtype 4 (sst4) (15,16). Up to now, clinically relevant radioligands were predominantly tested for internalization on sst2 model systems (3–7,14) because of their predominant sst2-binding affinity. Recently, however, increasing numbers of reports have been published on the development of somatostatin analogs with distinct affinity profiles for sst2, sst3, and sst5 such as DOTA-NOC (where DOTA is 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid and NOC is [1-NaI3]-octreotide) (4), BIM-23244 (17), or even analogs with a pansomatostatin profile such as KE108 (18). Thus, a thorough investigation of the internalization properties of these analogs is required at each of the somatostatin receptor subtypes.

One aim of the present study was to evaluate a variety of somatostatin analogs, either in clinical use or in development, for their capability to induce somatostatin receptor internalization, using morphologic or nonmorphologic immunocytochemical assays that are able to identify the receptor protein of the 3 somatostatin receptor subtypes: sst2, sst3, and sst5. The principle of the morphologic immunocytochemical assay is, first, to allow the living cells to interact for a given time with the nonradioactive agonists or antagonists to be tested. The cells are then fixed and made permeable for immunostaining of the somatostatin receptor of interest using specific somatostatin receptor antibodies. Localization of the receptor (cell surface, intracellular) can then easily be detected morphologically using an immunofluorescence microscope. In addition, a nonmorphologic ELISA has been used allowing the precise quantitation of the agonist-induced internalization process. These 2 assays therefore directly measure the internalization of the receptor itself rather than the bound ligand detected in radioligand internalization studies. Tested compounds included representatives of clinically used drugs, such as [Tyr3]-octreotide (TOC), lanreotide, and vapreotide (19–23); of pansomatostatins such as KE108 (18); or of analogs selective for sst2 (L-779,976) (24), for sst5 (L-817,818) (24), or for both NOC-ATE ([1-NaI3, Thr8]-octreotide) (25) and BIM-23244 (17). A number of chelated analogs of the first generation (DTPA-octreotide [where DTPA is diethylenetriaminepentaacetic acid] or DOTA-lanreotide) (26) or second generation (Y-DOTA-NOC, Y-DOTA-TATE, Lu-DOTA-BOC-ATE, or Lu-DOTA-NOC-ATE) (where TATE is [Tyr3, Thr8]-octreotide and BOC-ATE is [BzThi3, Thr8]-octreotide) (4,26) have also been tested, as well as several iodinated, sugar-containing octreotide analogs (27). For comparison, established sst2 or sst3 antagonists have been used (28,29). This study focused on sst2 internalization, because sst2 is the most important somatostatin receptor from a clinical point of view (1) and because most of the clinically available somatostatin analogs have a strong sst2 affinity (15,26). However, sst3 and sst5 internalization has also been investigated because many of the newly developed compounds have affinities for somatostatin receptors other than the sst2 subtype.

MATERIALS AND METHODS

Reagents

All reagents were of the best grade available and were purchased from common suppliers. The R2-88 antibody to the sst2A was generated as previously described and has been extensively characterized (30,31). The sst3-specific antibody (SS-850) and the corresponding C-terminal antigen peptide (S-851) were purchased from Gramsch Laboratories. The sst5-specific antibody (6005) and the corresponding antigen peptide (amino acids 12−20 of the human sst5) were provided by Dr. Stefan Schulz. The secondary antibody Alexa Fluor 488 goat antirabbit IgG (H+L) was from Molecular Probes, Inc. The rabbit polyclonal hemagglutinin epitope antibodies were purchased from Covance or from Sigma-Aldrich. The horseradish peroxidase substrate kit and the goat antirabbit IgG (H+L)-horseradish peroxidase conjugate were purchased from Bio-Rad Laboratories, Inc.

Peptides

Peptides were obtained as follows: somatostatin-14, somatostatin-28, KE108 (18), Coy-14 (BIM-23A760) (28), and sst3-ODN-8 (29) were synthesized at the Salk Institute and were provided by Dr. Jean Rivier; [Tyr3]-octreotide (TOC) (19) was from Novartis Inc.; vapreotide (RC160) (20) was from Calbiochem, somatostatin-28 was from Bachem; lanreotide (BIM-23014) (21–23) and BIM-23244 (17) were provided by Biomeasure Inc.; NOC-ATE (25), Y-DOTA-lanreotide (26), Y-DOTA-NOC (4), Lu-DOTA-BOC-ATE, and Lu-DOTA-NOC-ATE were provided by Dr. Helmut R. Mäcke; L-779,976 and L-817,818 (24) were from Merck Pharmaceuticals; DTPA-octreotide (MP2321) (26) and DOTA-lanreotide (MP2353) (26) were from Mallinckrodt; Y-DOTA-TOC (26), Y-DOTA-TATE (26), I-Gluc-TOC (27), I-Gluc-TATE (27), I-Gluc-S-TATE (27), and I-Gal-S-TATE (27) (where Gluc is glucose, S is mercaptopropionyl spacer, and Gal is galactose) were provided by Dr. Hans-Jürgen Wester. All peptides were dissolved in 10 mmol/L acetic acid except L-817,818, which was dissolved in 1:3 DMSO:H2O.

Cell Lines

The HEK293 cell lines expressing either the T7-epitope–tagged human sst2A (HEK-sst2), the human sst3 (HEK-sst3), or the human sst5 (HEK-sst5) were provided by Dr. Stefan Schulz and were cultured at 37°C and 5% CO2 in Dulbecco's modified eagle medium containing 10% fetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin, and 500 μg/mL G418. The clonal CHO-K1 cell line expressing the hemagglutinin-epitope–tagged rat sst2A receptor (CHO-sst2) was generated by transfection of CHO-K1 cells as previously described (14) and grown at 37°C and 5% CO2 in Ham's F12 medium containing 10% fetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin, and 250 μg/mL G418. All culture reagents were from Gibco BRL.

Binding-Affinity Measurements

The sst2, sst3, and sst5 binding affinity of the various compounds was measured as described previously using in vitro receptor autoradiography with 20-μm-thick sections from membrane pellets of the respective transfected cells (26).

Immunofluorescence Microscopy

Immunofluorescence microscopy–based internalization assay for sst2, sst3, and sst5 was performed as previously described by Liu et al. (14) for sst2. The internalization assay was extensively validated through a whole series of experiments in HEK-sst2 cells showing that sst2 is membrane bound in the absence of agonist; that there is a time, temperature, and agonist concentration dependency for sst2 internalization; and that sst2 is internalized via clathrin-coated vesicles and localizes to mannose 6-phosphate receptor–positive intracellular compartments, most likely the trans-Golgi network (TGN)/late endosome. Moreover, the internalization property of the HEK-sst2 (HEK293 cells expressing the T7-epitope tagged human sst2A) was found to be comparable to that of HEK293 cells expressing the wild-type sst2A in the internalization assay.

HEK-sst2, HEK-sst3, and HEK-sst5 cells were grown on poly-d-lysine (10 μg/mL) (Sigma-Aldrich)–coated 35-mm 4-well plates (Cellstar; Greiner Bio-One GmbH). Cells were treated either with the various agonists listed in Table 1 at concentrations ranging from 1 to 10,000 nmol/L, or with the agonists in the presence of an excess of antagonist (Coy-14, sst3-ODN-8; 50–100 times the concentration of the agonist), or with antagonist alone for 30 min at 37°C in growth medium and then rinsed twice with 100 mmol/L phosphate buffer containing 0.15 mol/L sucrose (PS). After the cells were fixed and made permeable for 7 min with cold methanol (−20°C), they were rinsed twice with PS, and nonspecific binding sites were blocked by incubating the cells in PS containing 0.1% bovine serum albumin for 60 min at room temperature. The cells were subsequently incubated for 60 min at room temperature with either the sst2A-specific primary antibody (R2-88) diluted 1:1,000 in PS, the sst3-specific primary antibody (SS-850) diluted 1:1,000 in PS, or the sst5-specific primary antibody (6005) diluted 1:2,000 in PS. After the antibody incubation, the cells were washed 3 times for 5 min with PS containing 0.1% bovine serum albumin and then incubated for 60 min at room temperature in the dark with the secondary antibody, Alexa Fluor 488 goat antirabbit IgG (H+L), diluted 1:600 in PS. Subsequently, the cells were washed 3 times for 5 min each with PS containing 0.1% bovine serum albumin, embedded with 1:1 PS:glycerol, and covered with a glass slip. No immunostaining was observed in HEK-sst2 cells with R2-88 (1:1,000) preabsorbed with antigen peptide, consistent with the known specificity of this receptor antibody (30,31). Similarly, no immunostaining was observed in HEK-sst3 cells with the sst3-specific antibody SS-850 preabsorbed with the corresponding antigen peptide S-851 or in HEK-sst5 cells with the sst5-specific antibody 6005 preabsorbed with the corresponding antigen peptide. The cells were imaged using a Leica DM RB immunofluorescence microscope and an Olympus DP10 camera.

View this table:
TABLE 1

sst2, sst3, and sst5 Binding Affinities and Internalization of Somatostatin and Somatostatin Analogs

Quantitation of Internalized Somatostatin Receptors by Immunofluorescence Microscopy

The internalization assay for sst2 and sst3 was performed as described previously by Liu et al. (14). Single cells were then analyzed for the amount of internalized somatostatin receptors after agonist stimulation. The immunofluorescence intensity of the labeled cells was densitometrically determined using a Zeiss Axioskop microscope equipped with a Roper CoolSNAP cf monochrome camera. The relative optical density (ROD) of the total area of the cell and the ROD of the area of the internalized somatostatin receptors were determined, and the percentage of internalized somatostatin receptors after agonist stimulation was then calculated according to the following equation:Math

The data were analyzed using the MCID Basic 7.0 program (Imaging Research Inc.). For each tested agonist concentration, 10–12 cells were analyzed and the mean value was used for the graph. The GraphPad Prism program, version 3.0, was used to create the graphs.

Quantitation of Internalized Somatostatin Receptors by ELISA

Receptor internalization was also quantitatively assessed in another cell line, CHO-K1 cells expressing the hemagglutinin-epitope tagged rat sst2A receptor (CHO-sst2), using an ELISA as described in detail previously (14).

RESULTS

Table 1 lists the somatostatin analogs tested for their ability to induce somatostatin receptor internalization using immunocytochemical detection methods. Apart from the 2 natural peptides, somatostatin-14 and somatostatin-28, they can be divided into 4 groups: a series of well-established peptide and nonpeptide analogs of current preclinical or clinical interest, which are nonchelated and, when tested previously, reported to be agonists; various peptide analogs linked to a chelator, for use in nuclear medicine, expected to be agonists but not systematically tested for agonism; several iodinated, sugar-containing analogs; and established somatostatin receptor antagonists. All compounds in Table 1 have a high affinity for one or more somatostatin receptors. The binding affinity data presented in Table 1 were determined using receptor autoradiography with membrane pellets (26,29); the results generally agree with previously published data generated with different methods (17,24,28).

All compounds in Table 1 were analyzed by immunofluorescence microscopy for internalization of sst2, sst3, and sst5 using a concentration ranging from 1 to 1,000 nmol/L, or even up to 10,000 nmol/L when the analogs were not of the highest affinity (e.g., chelated analogs at sst5). A compound was considered active when it induced somatostatin receptor internalization at an agonist concentration of at least 100 nmol/L. The results in Table 1 show that all agonists with a high affinity for sst2 induce internalization of sst2. The compounds that do not induce internalization of sst2 are the sst5-selective L-817,818 and the sst2 antagonist Coy-14. Figure 1 illustrates the sst2 internalization triggered by various analogs. Compared with the control (no peptide added), for which the sst2 is localized exclusively to the cell surface, each of the tested compounds can efficiently induce sst2 internalization, detectable as prominent punctate perinuclear staining as shown in Figure 1. This intracellular, perinuclear sst2 staining was shown previously (14) to colocalize with the TGN/late endosome marker protein mannose 6–phosphate receptor. Figure 1 also shows that agonist-induced sst2 internalization can be abolished by the sst2-specific antagonist Coy-14. Table 1 summarizes all the cases in which Coy-14 was used to antagonize sst2 internalization after agonist stimulation. These results indicate that sst2 internalization can be triggered only by somatostatin agonists, but not by the antagonist Coy-14. Figure 2A shows the potency of second-generation somatostatin analogs such as TOC, Y-DOTA-TATE, or I-Gal-S-TATE to elicit sst2 internalization, as compared with DTPA-octreotide. Abolition of TOC-induced sst2 internalization by Coy-14 is illustrated as well. Figure 2B illustrates the dose-response experiment and clearly shows that TOC is almost 2 orders of magnitude more potent in stimulating sst2 internalization than is DTPA-octreotide. In addition and for comparison, Table 1 also shows the internalization properties of the clinically relevant somatostatin analogs in another cell line, CHO-sst2, as measured by ELISA. Potencies (median effective concentration, EC50) for stimulation of endocytosis varied by more than 280 times between DTPA-octreotide and the best of the tested compounds, the sst2/sst5-selective BIM-23244. Among the chelated analogs, the most efficient internalization was found for Y-DOTA-TOC, Y-DOTA-TATE, Y-DOTA-NOC, and Lu-DOTA-BOC-ATE. Moreover, all tested iodinated sugar-containing octreotide analogs showed a highly efficient sst2 internalization as well.

FIGURE 1. 
FIGURE 1. 

sst2 internalization is induced by various agonists and is abolished by antagonist Coy-14. HEK-sst2 cells were treated for 30 min either with vehicle (no peptide [A]) or with various agonists at concentrations inducing submaximal internalization effect (100 nmol/L [B and C] and 10 nmol/L [D–F]). (H–L) Cells treated with same agonists at same concentrations as in B–F but in presence of excess of specific sst2-antagonist Coy-14 (5 [H], 10 [I], and 1 [J–L] μmol/L). Effect of antagonist alone (10 μmol/L [G]) is also shown. After incubation with peptides, cells were processed for immunocytochemistry. Clear punctate perinuclear staining is detectable for all tested agonists. This punctate staining is efficiently abolished by excess of antagonist Coy-14. Antagonist alone has no effect on internalization.

FIGURE 2. 
FIGURE 2. 

Dose response of agonist-induced sst2 internalization. HEK-sst2 cells were treated with 1 nmol/L, 10 nmol/L, 100 nmol/L, 1 μmol/L, or 10 μmol/L of various agonists. Also included are cells treated with 100 nmol/L TOC plus 5 μmol/L Coy-14 or with 1 μmol/L TOC plus 50 μmol/L Coy-14. After 30 min of incubation with peptides, cells were processed for immunocytochemistry, and internalized sst2 was quantified. (A) Representative experiment with dose-response curves for various agonists inducing sst2 internalization. TOC, Y-DOTA-TATE, and I-Gal-S-TATE are considerably more potent in inducing sst2 internalization than is DTPA-octreotide. TOC-induced internalization is efficiently abolished by excess of Coy-14. (B) Representative immunofluorescence images of agonist-induced internalized sst2 using DTPA-octreotide and TOC. TOC is almost 2 orders of magnitude more potent in inducing sst2 internalization than is DTPA-octreotide.

Table 1 further shows that analogs with high sst3 affinity are able to trigger sst3 internalization. In addition to somatostatin-14 and somatostatin-28, this effect is observed for KE108, Y-DOTA-NOC, Lu-DOTA-NOC-ATE, and Lu-DOTA-BOC-ATE. Other compounds with comparatively low sst3 affinity were unable to induce sst3 internalization at a dose of 100 nmol/L. Figure 3 illustrates the effects of somatostatin-28, Y-DOTA-NOC, and KE108 that can be antagonized by the sst3-selective antagonist sst3-ODN-8. Figure 4 shows a dose-response curve with somatostatin-28 and KE108 and also shows that the somatostatin-28–induced internalization of sst3 is abolished by the sst3-specific antagonist sst3-ODN-8.

FIGURE 3. 
FIGURE 3. 

sst3 internalization is induced by various agonists and is efficiently abolished by sst3-specific antagonist sst3-ODN-8. HEK-sst3 cells were treated either with vehicle (no peptide [A]) or with various agonists at concentrations inducing submaximal internalization effect (100 nmol/L [B–D]). Cells treated with same agonists at same concentrations as in B–D but in presence of excess of specific sst3-antagonist sst3-ODN-8 (5 μmol/L [F–H]). Effect of antagonist alone (50 μmol/L [E]) is also illustrated. After incubation with peptides, cells were processed for immunocytochemistry. Clear perinuclear staining is detectable for all agonists tested. This staining is efficiently abolished by 50-fold excess of antagonist sst3-ODN-8. Antagonist alone has no effect on internalization.

FIGURE 4. 
FIGURE 4. 

Dose response of agonist-induced sst3 internalization. HEK-sst3 cells were treated with 1 nmol/L, 10 nmol/L, 100 nmol/L, 1 μmol/L, or 10 μmol/L of somatostatin-28 or KE108. Also included are cells treated with 100 nmol/L somatostatin-28 plus 5 μmol/L sst3-ODN-8 or with 1 μmol/L somatostatin-28 plus 50 μmol/L sst3-ODN-8. After incubation with peptides, cells were processed for immunocytochemistry, and internalized sst3 was quantified. This experiment shows that somatostatin-28 and KE108 are of similar potency in inducing sst3 internalization. Somatostatin-28–induced internalization effect is efficiently abolished by excess of sst3-ODN-8

Finally, Table 1 and Figure 5 show that both somatostatin-14 and somatostatin-28 trigger sst5 internalization. The sst5 internalization is less pronounced than is observed with sst2 or sst3, partly because, in contrast to sst2- or sst3-expressing cells, even untreated sst5 cells have an intracellular pool of sst5, as shown in Figure 5. Interestingly, somatostatin-28 induces greater sst5 internalization at 100 nmol/L than does somatostatin-14, perhaps as a result of the higher binding affinity of somatostatin-28 than of somatostatin-14 to sst5. This finding is also summarized in Table 1. Because well-characterized sst5 antagonists are not available, blocking studies with an sst5 antagonist could not be performed. However, we showed that the internalization process was abolished in the presence of 0.45 mol/L sucrose, a result that strongly supports the specificity of the observation. Unexpectedly, several high-affinity synthetic sst5 agonists such as L-817,818, BIM-23244, or KE108 are unable to elicit an sst5 internalization response, even at doses of up to 1,000 nmol/L, as shown in Figure 5. Therefore, in contrast to sst2 and sst3, not all agonists are able to stimulate sst5 receptor endocytosis.

FIGURE 5. 
FIGURE 5. 

sst5 internalization is induced by native somatostatin-28 but not by high-affinity synthetic sst5 agonists. HEK-sst5 cells were treated either with vehicle (no peptide); with 1 nmol/L, 10 nmol/L, 100 nmol/L, or 1 μmol/L somatostatin-28; or with 1 μmol/L KE108, 1 μmol/L BIM-23244, or 1 μmol/L L-817,818. After incubation with peptides, cells were processed for immunocytochemistry. Dose-response experiment with somatostatin-28 shows that it can elicit sst5 internalization. This is, however, not true for sst5 agonists KE108, BIM-23244, and L-817,818. Note also presence of intracellular sst5 even when cells were treated with vehicle alone (no peptide). All images are composed of 2 single pictures.

DISCUSSION

In this study, sensitive immunocytochemical methods were applied to examine sst2, sst3, and sst5 receptor internalization after agonist or antagonist treatment, using a variety of somatostatin analogs with established or potential interest for nuclear medicine. To our knowledge, this was the first time that a method monitoring receptor trafficking rather than radioligand trafficking was used to preclinically evaluate new G-protein–coupled receptor ligands for potential use in nuclear medicine. The immunocytochemistry-based internalization assay has several important advantages over assays using radiolabeled ligands. Because we are monitoring receptor trafficking, we are not restricted to the use of radiolabeled ligands but can test any nonlabeled compound. Further, using this method, receptor internalization can be monitored at a broad range of agonist and antagonist concentrations rather than at subsaturating concentrations, as usually occur for radioligands. Moreover, unlike radioactive isotopes, ligands to be tested for internalization will not experience alteration, which might affect the structure of the ligand and thus its biologic activity.

The study also showed that sst2 agonists, but not sst2 antagonists, can trigger sst2 internalization, in agreement with our earlier study (14). Moreover, the study demonstrated that sst2 antagonists can selectively antagonize the sst2 agonistic effect on internalization. It thus seems likely that compounds such as Y-DOTA-NOC, Y-DOTA-TATE, Lu-DOTA-BOC-ATE, Lu-DOTA-NOC-ATE, or the sugar-containing octreotide analogs, each of which induces internalization, are agonists at the sst2.

The study further showed that high-affinity sst2 binding is a prerequisite for an agonist to trigger sst2 internalization. Thus, the sst5-selective L-817,818, with a low affinity for sst2, is unable to trigger sst2 internalization. Conversely, all agonists with high-affinity sst2-binding properties were able to internalize sst2. The agonist with the highest sst2-binding affinity, BIM-23244, had the highest sst2 internalization potency. We should emphasize that the second-generation compounds foreseen for in vivo sst2 targeting, such as octreotides or octreotates modified in position 3 and linked to DOTA or sugars, often have considerably better internalization capabilities than do the first-generation compound DTPA-octreotide.

Factors other than ligand binding also play an important role in receptor internalization, and these can have cell-specific effects. Previous studies have demonstrated that sst2 and sst3 are rapidly phosphorylated on agonist binding, most probably by G-protein–coupled receptor kinases (32–34). Receptor phosphorylation is followed by the recruitment of β-arrestins to the receptor (14,35), and the bound arrestins then link the receptors to the endocytosis machinery. The nature and concentration of G-protein–coupled receptor kinase subtypes, as well as the relative concentrations of the 2 arrestins, β-arrestin-1 and β-arrestin-2, are known to vary among cell types and are thought to produce differences in the efficiency and extent of receptor internalization (36). Therefore, in an attempt to generalize the data obtained in HEK-sst2 cells, we have tested several clinically relevant compounds for their ability to stimulate receptor internalization in a different cell line, namely in CHO-sst2 cells, using our previously described quantitative ELISA method (14). Our results demonstrated that all tested compounds produced internalization in both cell types. However, the interesting and clinically important possibility exists that receptor internalization will differ in certain tumor types.

To our knowledge, this was the first time that the internalization capability of novel somatostatin receptor ligands potentially useful in nuclear medicine has been examined for sst3 and sst5. Highly promising drugs with high binding affinity to sst3, such as KE108, Y-DOTA-NOC, Lu-DOTA-BOC-ATE, and Lu-DOTA-NOC-ATE, show excellent sst3 internalization properties. Moreover, these compounds can be antagonized by the selective sst3 antagonist sst3-ODN-8, strongly indicating that Y-DOTA-NOC, Lu-DOTA-BOC-ATE, and Lu-DOTA-NOC-ATE are agonists at sst3—information that was not previously available.

In contrast to sst2 and sst3, the cellular distribution characteristics of sst5 are unusual. Even in untreated cells, a distinct intracellular perinuclear staining of sst5 is observed in addition to cell-surface sst5 staining. One possible reason for this combined cell surface and intracellular staining may be the particular cellular distribution and trafficking of sst5, which was first reported by Stroh et al. (37). They elegantly showed, by biochemical, confocal, and electron microscopy methods, that in COS-7 cells exogenously expressing the rat sst5, functional sst5 is maintained at the cell surface even in the presence of somatostatin, both because of the rapid recycling of the internalized receptor to the cell surface and because of a massive recruitment of sst5 to the cell surface from an intracellular sst5 reserve pool. In our HEK-sst5 cells, the balance between depletion of the intracellular sst5 pool and sst5 internalization in the presence of somatostatin resulted in a net internalization, because the intracellular staining became more prominent after somatostatin treatment. Whether the internalized receptor colocalizes with the intracellular receptor pool is unknown. Although the presence of an intracellular sst5 pool makes the evaluation of an agonist-triggered internalization more difficult for sst5 than for sst2 or sst3, somatostatin-28 and somatostatin-14 are, without doubt, able to induce sst5 internalization. The better sst5 internalization by 100 and 1,000 nmol/L of somatostatin-28 than by somatostatin-14 may be explained by the significantly higher binding affinity of somatostatin-28. However, the observation that some of the most potent sst5 agonists, such as BIM-23244, KE108, or L-817,818, with binding affinities below 1 nmol/L were unable to trigger any net sst5 internalization was not expected. We previously showed a similar dissociation between the binding of certain somatostatin analogs and their efficacy at promoting sst2 internalization (14). Further, we observed that the sst2–arrestin complex was less stable in the presence of an analog that was a poor inducer of internalization than in the presence of somatostatin-14, which promoted internalization effectively (14). These results support the conclusion that the ability of analogs to facilitate the formation of a stable receptor–arrestin complex determines their effectiveness for stimulating endocytosis. Although arrestins are not recruited as strongly to sst5 as to sst2 (34), it is possible that receptor–arrestin interactions are also important for determining the ability of sst5 analogs to stimulate receptor internalization. The differences among analogs for producing sst5 endocytosis, described here for what is to our knowledge the first time, may not only help explain the biologic actions of these analogs but also demonstrate the importance of individual assessment of agonists for their effect on somatostatin receptor internalization.

CONCLUSION

These present data are likely to be important for the preclinical evaluation of the internalization properties of new ligands aimed at tumor targeting and for the interpretation of future imaging data using labeled somatostatin analogs. The described methodology may be used in future to screen such novel peptide ligands for their agonistic and antagonistic properties on internalization. One of the most novel and unexpected results from these studies is that compounds with high affinity at both sst2 and sst5, such as BIM-23244 or KE108, or the chelated analogs Y-DOTA-NOC or Lu-DOTA-BOC-ATE, show distinct internalization properties at these 2 receptors, namely a strong sst2 but no sst5 internalization. Such data will have to be considered when one is interpreting clinical studies using these compounds for tumor targeting.

Acknowledgments

This work was supported in part by NIH grants DK059953 and DK032234 and SNF grant 3100A0-100390. We thank Merck Pharmaceuticals for the generous gift of L-779,976 and L-817,818.

References

  1. 1.↵
    Reubi JC. Peptide receptors as molecular targets for cancer diagnosis and therapy. Endocr Rev. 2003;24:389–427.
    OpenUrlCrossRefPubMed
  2. 2.↵
    Kwekkeboom DJ, Krenning EP, de Jong M. Peptide receptor imaging and therapy. J Nucl Med. 2000;41:1704–1713.
    OpenUrlAbstract/FREE Full Text
  3. 3.↵
    Schottelius M, Poethko T, Herz M, et al. Featured article: first (18)F-labeled tracer suitable for routine clinical imaging of sst receptor-expressing tumors using positron emission tomography. Clin Cancer Res. 2004;10:3593–3606.
    OpenUrlAbstract/FREE Full Text
  4. 4.↵
    Wild D, Schmitt JS, Ginj M, et al. DOTA-NOC, a high-affinity ligand of somatostatin receptor subtypes 2, 3 and 5 for labelling with various radiometals. Eur J Nucl Med Mol Imaging. 2003;30:1338–1347.
    OpenUrlCrossRefPubMed
  5. 5.
    Nock B, Nikolopoulou A, Chiotellis E, et al. [99mTc]Demobesin 1, a novel potent bombesin analogue for GRP receptor-targeted tumour imaging. Eur J Nucl Med. 2003;30:247–258.
    OpenUrl
  6. 6.
    Wang M, Caruano AL, Lewis MR, Meyer LA, VanderWaal RP, Anderson CJ. Subcellular localization of radiolabeled somatostatin analogues: implications for targeted radiotherapy of cancer. Cancer Res. 2003;63:6864–6869.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    de Jong M, Bernard BF, De Bruin E, et al. Internalization of radiolabelled [DTPA0]octreotide and [DOTA0,Tyr3]octreotide: peptides for somatostatin receptor-targeted scintigraphy and radionuclide therapy. Nucl Med Commun. 1998;19:283–288.
    OpenUrlPubMed
  8. 8.↵
    Schottelius M, Reubi JC, Eltschinger V, Schwaiger M, Wester HJ. N-terminal sugar conjugation and C-terminal Thr-for-Thr(ol) exchange in radioiodinated Tyr3-octreotide: effect on cellular ligand trafficking in vitro and tumor accumulation in vivo. J Med Chem. 2005;48:2778–2789.
    OpenUrlCrossRefPubMed
  9. 9.↵
    Koenig JA, Edwardson JM. Endocytosis and recycling of G protein-coupled receptors. Trends Pharmacol Sci. 1997;18:276–287.
    OpenUrlCrossRefPubMed
  10. 10.↵
    Gray JA, Roth BL. Paradoxical trafficking and regulation of 5-HT(2A) receptors by agonists and antagonists. Brain Res Bull. 2001;56:441–451.
    OpenUrlCrossRefPubMed
  11. 11.
    Roettger BF, Ghanekar D, Rao R, et al. Antagonist-stimulated internalization of the G protein-coupled cholecystokinin receptor. Mol Pharmacol. 1997;51:357–362.
    OpenUrlAbstract/FREE Full Text
  12. 12.
    Bhowmick N, Narayan P, Puett D. The endothelin subtype A receptor undergoes agonist- and antagonist-mediated internalization in the absence of signaling. Endocrinology. 1998;139:3185–3192.
    OpenUrlCrossRefPubMed
  13. 13.↵
    Pheng LH, Dumont Y, Fournier A, Chabot JG, Beaudet A, Quirion R. Agonist- and antagonist-induced sequestration/internalization of neuropeptide Y Y1 receptors in HEK293 cells. Br J Pharmacol. 2003;139:695–704.
    OpenUrlCrossRefPubMed
  14. 14.↵
    Liu Q, Cescato R, Dewi DA, Rivier J, Reubi JC, Schonbrunn A. Receptor signaling and endocytosis are differentially regulated by somatostatin analogs. Mol Pharmacol. 2005;68:90–101.
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    Olias G, Viollet C, Kusserow H, Epelbaum J, Meyerhof W. Regulation and function of somatostatin receptors. J Neurochem. 2004;89:1057–1091.
    OpenUrlCrossRefPubMed
  16. 16.↵
    Csaba Z, Dournaud P. Cellular biology of somatostatin receptors. Neuropeptides. 2001;35:1–23.
    OpenUrlCrossRefPubMed
  17. 17.↵
    Saveanu A, Gunz G, Dufour H, et al. Bim-23244, a somatostatin receptor subtype 2- and 5-selective analog with enhanced efficacy in suppressing growth hormone (GH) from octreotide-resistant human GH-secreting adenomas. J Clin Endocrinol Metab. 2001;86:140–145.
    OpenUrlCrossRefPubMed
  18. 18.↵
    Reubi JC, Eisenwiener KP, Rink H, Waser B, Macke HR. A new peptidic somatostatin agonist with high affinity to all five somatostatin receptors. Eur J Pharmacol. 2002;456:45–49.
    OpenUrlCrossRefPubMed
  19. 19.↵
    Reubi JC. Evidence for two somatostatin-14 receptor types in rat brain cortex. Neurosci Lett. 1984;49:259–263.
    OpenUrlCrossRefPubMed
  20. 20.↵
    Cai RZ, Szoke B, Lu R, Fu D, Redding TW, Schally AV. Synthesis and biological activity of highly potent octapeptide analogs of somatostatin. Proc Natl Acad Sci U S A. 1986;83:1896–1900.
    OpenUrlAbstract/FREE Full Text
  21. 21.↵
    Taylor JE, Bogden AE, Moreau JP, Coy DH. In vitro and in vivo inhibition of human small cell lung carcinoma (NCI-H69) growth by a somatostatin analogue. Biochem Biophys Res Commun. 1988;153:81–86.
    OpenUrlCrossRefPubMed
  22. 22.
    Shimon I, Taylor JE, Dong JZ, et al. Somatostatin receptor subtype specificity in human fetal pituitary cultures: differential role of SSTR2 and SSTR5 for growth hormone, thyroid-stimulating hormone, and prolactin regulation. J Clin Invest. 1997;99:789–798.
    OpenUrlCrossRefPubMed
  23. 23.↵
    Hannon JP, Nunn C, Stolz B, et al. Drug design at peptide receptors: somatostatin receptor ligands. J Mol Neurosci. 2002;18:15–27.
    OpenUrlCrossRefPubMed
  24. 24.↵
    Rohrer SP, Birzin ET, Mosley RT, et al. Rapid identification of subtype-selective agonists of the somatostatin receptor through combinatorial chemistry. Science. 1998;282:737–740.
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    Ginj M, Chen J, Walter MA, Eltschinger V, Reubi JC, Maecke HR. Preclinical evaluation of new and highly potent analogues of octreotide for predictive imaging and targeted radiotherapy. Clin Cancer Res. 2005;11:1136–1145.
    OpenUrlAbstract/FREE Full Text
  26. 26.↵
    Reubi JC, Schaer JC, Waser B, et al. Affinity profiles for human somatostatin receptor sst1-sst5 of somatostatin radiotracers selected for scintigraphic and radiotherapeutic use. Eur J Nucl Med. 2000;27:273–282.
    OpenUrlCrossRefPubMed
  27. 27.↵
    Schottelius M, Rau F, Reubi JC, Schwaiger M, Wester HJ. Modulation of pharmacokinetics of radioiodinated sugar-conjugated somatostatin analogues by variation of peptide net charge and carbohydration chemistry. Bioconjug Chem. 2005;16:429–437.
    OpenUrlPubMed
  28. 28.↵
    Rajeswaran WG, Hocart SJ, Murphy WA, Taylor JE, Coy DH. Highly potent and subtype selective ligands derived by N-methyl scan of a somatostatin antagonist. J Med Chem. 2001;44:1305–1311.
    OpenUrlCrossRefPubMed
  29. 29.↵
    Reubi JC, Schaer JC, Wenger S, et al. sst3-selective potent peptidic somatostatin receptor antagonists. Proc Natl Acad Sci U S A. 2000;97:13973–13978.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    Gu Y-Z, Schonbrunn A. Coupling specificity between somatostatin receptor sst2A and G proteins: isolation of the receptor-G protein complex with a receptor antibody. Mol Endocrinol. 1997;11:527–537.
    OpenUrlCrossRefPubMed
  31. 31.↵
    Reubi JC, Kappeler A, Waser B, Laissue JA, Hipkin RW, Schonbrunn A. Immunohistochemical localization of somatostatin receptors sst2A in human tumors. Am J Pathol. 1998;153:233–245.
    OpenUrlPubMed
  32. 32.↵
    Hipkin RW, Friedman J, Clark RB, Eppler CM, Schonbrunn A. Agonist-induced desensitization, internalization and phosphorylation of the sst2A somatostatin receptor. J Biol Chem. 1997;272:13869–13876.
    OpenUrlAbstract/FREE Full Text
  33. 33.
    Roth A, Kreienkamp HJ, Meyerhof W, Richter D. Phosphorylation of four amino acid residues in the carboxyl terminus of the rat somatostatin receptor subtype 3 is crucial for its desensitization and internalization. J Biol Chem. 1997;272:23769–23774.
    OpenUrlAbstract/FREE Full Text
  34. 34.↵
    Tulipano G, Stumm R, Pfeiffer M, Kreienkamp HJ, Hollt V, Schulz S. Differential beta-arrestin trafficking and endosomal sorting of somatostatin receptor subtypes. J Biol Chem. 2004;279:21374–21382.
    OpenUrlAbstract/FREE Full Text
  35. 35.↵
    Kreuzer OJ, Krisch B, Dery O, Bunnett NW, Meyerhof W. Agonist-mediated endocytosis of rat somatostatin receptor subtype 3 involves beta-arrestin and clathrin coated vesicles. J Neuroendocrinol. 2001;13:279–287.
    OpenUrlCrossRefPubMed
  36. 36.↵
    Menard L, Ferguson SS, Zhang J, et al. Synergistic regulation of beta2-adrenergic receptor sequestration: intracellular complement of beta-adrenergic receptor kinase and beta-arrestin determine kinetics of internalization. Mol Pharmacol. 1997;51:800–808.
    OpenUrlAbstract/FREE Full Text
  37. 37.↵
    Stroh T, Jackson AC, Sarret P, et al. Intracellular dynamics of sst5 receptors in transfected COS-7 cells: maintenance of cell surface receptors during ligand-induced endocytosis. Endocrinology. 2000;141:354–365.
    OpenUrlCrossRefPubMed
  38. 38.↵
    Erchegyi J, Hoeger CA, Low W, et al. Somatostatin receptor 1 selective analogues: 2. N(alpha)-methylated scan. J Med Chem. 2005;48:507–514.
    OpenUrlPubMed
  39. 39.↵
    Rivier JE, Kirby DA, Erchegyi J, et al. Somatostatin receptor 1 selective analogues: 3. dicyclic peptides. J Med Chem. 2005;48:515–522.
    OpenUrlCrossRefPubMed
  • Received for publication July 6, 2005.
  • Accepted for publication September 23, 2005.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section

Related Articles

Cited By...

More in this TOC Section

Similar Articles