Sorbitol as a Polar Pharmacological Modifier to Enhance the Hydrophilicity of 99mTc-Tricarbonyl-Based Radiopharmaceuticals

The organometallic technetium-99m tricarbonyl core, [99mTc][Tc(CO)3(H2O)3]+, is a versatile precursor for the development of radiotracers for single photon emission computed tomography (SPECT). A drawback of the 99mTc-tricarbonyl core is its lipophilicity, which can influence the pharmacokinetic properties of the SPECT imaging probe. Addition of polar pharmacological modifiers to 99mTc-tricarbonyl conjugates holds the promise to counteract this effect and provide tumor-targeting radiopharmaceuticals with improved hydrophilicities, e.g., resulting in a favorable fast renal excretion in vivo. We applied the “Click-to-Chelate” strategy for the assembly of a novel 99mTc-tricarbonyl labeled conjugate made of the tumor-targeting, modified bombesin binding sequence [Nle14]BBN(7–14) and the carbohydrate sorbitol as a polar modifier. The 99mTc-radiopeptide was evaluated in vitro with PC-3 cells and in Fox-1nu mice bearing PC-3 xenografts including a direct comparison with a reference conjugate lacking the sorbitol moiety. The glycated 99mTc-tricarbonyl peptide conjugate exhibited an increased hydrophilicity as well as a retained affinity toward the Gastrin releasing peptide receptor and cell internalization properties. However, there was no significant difference in vivo in terms of pharmacokinetic properties. In particular, the rate and route of excretion was unaltered in comparison to the more lipophilic reference compound. This could be attributed to the intrinsic properties of the peptide and/or its metabolites. We report a novel glycated (sorbitol-containing) alkyne substrate for the “Click-to-Chelate” methodology, which is potentially of general applicability for the development of 99mTc-tricarbonyl based radiotracers displaying an enhanced hydrophilicity.


Introduction
Radiolabeled peptides have become an indispensable tool in nuclear medicine for the diagnosis (imaging) and therapy of cancer. In the majority of cases, radioactive metals are used in this context. Application of radiometal complexes for the radiolabeling of peptides has the advantage that the metal is exchangeable and thus, both diagnostic and therapeutic probes become accessible depending on the radionuclide employed. This approach has been termed "theranostics" [1]. Different combinations of radiometals have been described in the literature for theranostic approaches [2], for example the matched pair of the group 7 transition metals technetium-99m (γ-emitter for imaging) and rhenium-186/188 (β --emitter for therapy) [3]. Technetium-99m ( 99m Tc) is considered the workhorse of nuclear medicine among the single photon emitting radionuclides for molecular imaging by SPECT due to its optimal physical properties (t 1/2 = 6 h, E γ = 140 keV) and broad availability. After being eluted from the generator as pertechnetate, 99m Tc(VII)O 4 − must be reduced to a lower oxidation state for the coordination with a chelating system. Among different reported 99m Tc precursors, also referred to as 99m Tc cores [4], the 99m Tc-tricarbonyl core, [ 99m Tc][Tc(CO) 3 (H 2 O) 3 ] + , is an interesting and promising candidate that can be readily prepared via the reduction of 99m Tc(VII) to 99m Tc(I) using commercial kits [5]. However, a potential drawback of the organometallic 99m Tc-tricarbonyl precursor for radiometallation of (bio)molecules is its increased lipophilicity in comparison to other 99m Tc-cores and radiometals, which can impact the pharmacokinetic profile of a tumor-targeting radiotracer. Lipophilic radiolabeled peptides can show hepatic uptake and slow hepatobiliary excretion instead of fast renal elimination, which is a favored characteristic of tumor-targeting imaging probes [6].
To overcome such issues, a number of radiolabeled peptides have been modified synthetically by conjugating polar pharmacological modifiers such as carboxylates [7][8][9], oligo-and polyethylene glycols [10,11], or carbohydrates [12,13]. An elegant methodology to introduce a chelator for the 99m Tc-tricarbonyl core into (bio)molecules is via the "Click-to-Chelate" approach (Figure 1) [6,14]. This strategy includes an application of the Cu(I)-catalyzed azide-alkyne cycloaddition (CuAAC) [15,16] by which an efficient tridentate chelator for the 99m Tc-tricarbonyl core is formed and linked to (bio)molecules of interest while other moieties, e.g., carbohydrates, can be conjugated simultaneously [17]. We have previously applied this methodology to the assembly of a peptide conjugate comprising a modified analog of the binding sequence of the tumor-targeting peptide bombesin (BBN), namely [Nle 14 ]BBN (7)(8)(9)(10)(11)(12)(13)(14), a 99m Tc-tricarbonyl complex and the carbohydrate glucuronic acid [18]. BBN is an attractive peptide for targeting the Gastrin-releasing peptide receptor (GRPR), which is overexpressed by a variety of tumors, including prostate and breast cancer [19]. The aim of the study was to enhance the hydrophilicity of the radiolabeled peptide and therefore favoring renal over hepatobiliary excretion. While the hydrophilicity of the glycated peptide conjugate was found substantially increased, we were surprised that it accumulated nevertheless to a higher degree in the spleen and liver in comparison to a more lipophilic BBN reference compound. We assumed that glucuronated, negatively charged metabolites of the radiolabeled peptide can be recognized by organic anion transporting polypeptides (OATPs), which are expressed in these organs [18].
We thus resorted to the use of other carbohydrates as pharmacological modifiers. Reduced acyclic carbohydrates lacking potentially charged functional groups appeared as a promising alternative. For example, sorbitol is used safely as an artificial sweetener in food products. Weinstein et al. have reported that [ 18 F]fluorodeoxysorbitol has potential as an infection imaging agent for positron emission tomography (PET) and does not accumulate in the spleen and liver [20]. Also Leamon et al. showed that sorbitol derivatives are promising polar moieties for preventing hepatic clearance of drugs [21]. We thus hypothesized that the sorbitol moiety might present an appropriate innocent pharmacological modifier for our purpose. Herein, we report the "Click-to-Chelate" synthesis, radiolabeling and in vitro and in vivo evaluation of a novel glycated (sorbitol-containing), 99m Tc-tricarbonyl labeled [Nle 14 ]BBN (7)(8)(9)(10)(11)(12)(13)(14) conjugate including a side-by-side comparison with a reference compound lacking the carbohydrate moiety. hydrophilicity of the radiolabeled peptide and therefore favoring renal over hepatobiliary excretion. While the hydrophilicity of the glycated peptide conjugate was found substantially increased, we were surprised that it accumulated nevertheless to a higher degree in the spleen and liver in comparison to a more lipophilic BBN reference compound. We assumed that glucuronated, negatively charged metabolites of the radiolabeled peptide can be recognized by organic anion transporting polypeptides (OATPs), which are expressed in these organs [18]. We thus resorted to the use of other carbohydrates as pharmacological modifiers. Reduced acyclic carbohydrates lacking potentially charged functional groups appeared as a promising

Syntheses
The synthesis of a sorbitol containing CuAAC substrate for "Click-to-Chelate" started with the alkylation of 2,3;4,5-Di-O-isopropylidene D-glucitol (1) at the primary hydroxyl group with tert-butyl 4-bromobutanoate (2) by a reported 2-step procedure (Scheme 1) [22]. Removal of the acetonide groups of sorbitol derivative 3 followed by acetylation of the resulting pentaol 4 and subsequent cleavage of the tertBu-ester of intermediate 5 gave pentaacetate 6 in satisfying overall yield. Coupling of pentaacetate 6 to the N(ε)-amine of the previously reported propargyl lysine derivative 7 [17] provided product 8 in good yield. Stepwise deprotection of compound 8 yielded first intermediate 9, which upon saponification with NaOH gave the desired glycated and deprotected alkyne synthon 10 for subsequent CuAAC.  [20]. Also Leamon et al. showed that sorbitol derivatives are promising polar moieties for preventing hepatic clearance of drugs [21]. We thus hypothesized that the sorbitol moiety might present an appropriate innocent pharmacological modifier for our purpose. Herein, we report the "Click-to-Chelate" synthesis, radiolabeling and in vitro and in vivo evaluation of a novel glycated (sorbitol-containing), 99m Tc-tricarbonyl labeled [Nle 14 ]BBN (7)(8)(9)(10)(11)(12)(13)(14) conjugate including a side-by-side comparison with a reference compound lacking the carbohydrate moiety.

Syntheses
The synthesis of a sorbitol containing CuAAC substrate for "Click-to-Chelate" started with the alkylation of 2,3;4,5-Di-O-isopropylidene D-glucitol (1) at the primary hydroxyl group with tert-butyl 4-bromobutanoate (2) by a reported 2-step procedure (Scheme 1) [22]. Removal of the acetonide groups of sorbitol derivative 3 followed by acetylation of the resulting pentaol 4 and subsequent cleavage of the tertBu-ester of intermediate 5 gave pentaacetate 6 in satisfying overall yield. Coupling of pentaacetate 6 to the N(ε)-amine of the previously reported propargyl lysine derivative 7 [17] provided product 8 in good yield. Stepwise deprotection of compound 8 yielded first intermediate 9, which upon saponification with NaOH gave the desired glycated and deprotected alkyne synthon 10 for subsequent CuAAC. Azido-bombesin derivative 11 was prepared manually by standard Fmoc solid phase peptide synthesis (SPPS) as previously described (Scheme 2A) [23]. In order to avoid potential interference of the radiometal complex with the tumor-targeting BBN vector, a spacer consisting of three βAla was introduced between the moieties. The peptide was cleaved from the resin, deprotected, and reacted in aqueous solution with alkyne derivative 10 using Cu(OAc)2 and Na-ascorbate. This yielded, after HPLC purification, the glycated peptide conjugate BBN-12. Reference compound BBN-13, identical in all aspects to the investigational compound BBN-12 but lacking the carbohydrate moiety, was prepared by the same chemistry as previously reported [23]. Azido-bombesin derivative 11 was prepared manually by standard Fmoc solid phase peptide synthesis (SPPS) as previously described (Scheme 2A) [23]. In order to avoid potential interference of the radiometal complex with the tumor-targeting BBN vector, a spacer consisting of three βAla was introduced between the moieties. The peptide was cleaved from the resin, deprotected, and reacted in aqueous solution with alkyne derivative 10 using Cu(OAc) 2 and Na-ascorbate. This yielded, after HPLC purification, the glycated peptide conjugate BBN-12. Reference compound BBN-13, identical in all aspects to the investigational compound BBN-12 but lacking the carbohydrate moiety, was prepared by the same chemistry as previously reported [23].

TLC Analytics
In order to have an additional quality control with easy and rapid performance, a TLC system was developed to analyze the radiolabeled peptides conjugates (Table 1). The following retention

(Radio) Metal-Labeling and Characterization
The tricarbonyl precursor [ 99m Tc][Tc(CO) 3 3 ] + was carried out successfully either in a conventional heating block at 100 • C for 30 min or in a microwave reactor at 120 • C for 5 min (Scheme 2). The desired radiopeptides [ 99m Tc][Tc(CO) 3 (L)] (L = BBN-12, 13) were obtained independent of the reaction conditions in a radiochemical yield and purity of ≥ 95% as determined by analytical γ-HPLC and γ-TLC.

TLC Analytics
In order to have an additional quality control with easy and rapid performance, a TLC system was developed to analyze the radiolabeled peptides conjugates (Table 1). The following retention factors (R f ) were obtained for each analyzed sample and the radiochemical purity of the radiolabeled peptides was always in accordance to the results obtained using the γ-HPLC method. The impurities identified were either unreacted pertechnetate (~0.7%) and/or 99m Tc-tricarbonyl precursor (~1.8%), which is also in accordance to the γ-HPLC results.

In Vitro Evaluation
[ 99m Tc][Tc(CO) 3 (12)] was evaluated in vitro and compared with the reference compound [ 99m Tc][Tc(CO) 3 (13)]. Receptor binding and cell internalization properties of the 99m Tc(CO) 3 -labeled BBN derivatives were investigated with GRPR expressing PC-3 cells (n = 2-3 in triplicate). Despite the slightly slower uptake kinetics of [ 99m Tc][Tc(CO) 3 (12)], both radiolabeled compounds [ 99m Tc][Tc(CO) 3 (L)] (L = 12, 13) showed a comparable extent of cell binding and internalization: A plateau was reached after 1 h (22.9% ± 0.4% and 25.0% ± 0.1% of the applied radioactivity (AD)/10 6 cells, respectively; Table 2, Figure 2). The membrane bound fraction was less than 5% AD/10 6 cells for both compounds. The receptor-specific cell uptake was verified by incubating the cells with the radiopeptides in the presence of 1000-fold excess of natural bombesin as blocking agent. For both radiolabeled peptides this resulted in a significant decrease of cellular binding and uptake (<0.5% AD/10 6 cells). Molecules 2020, 25, x FOR PEER REVIEW 6 of 15  Table 2) [23]. Some variability in experimentally determined Kd and Bmax of structurally related radiolabeled peptides is not uncommon [18,23,27], yet in this case somewhat surprising as we did not expect the sorbitol moiety to have an influence on these parameters; this in particular because the results of cell binding and internalization experiments were similar for both radiolabeled peptides ( Figure 2). Differences in specific molar activities (MBq/mol) could account for this observation and cannot be entirely excluded even though both 99m Tc-radiopeptides were prepared under identical conditions. Because this study aims at the investigation of differences in the pharmacokinetics of the radiopeptides (e.g., rate and route of excretion) rather than in tumor uptake, we continued with the evaluation of both compounds in vivo.   Table 2) [23]. Some variability in experimentally determined K d and B max of structurally related radiolabeled peptides is not uncommon [18,23,27], yet in this case somewhat surprising as we did not expect the sorbitol moiety to have an influence on these parameters; this in particular because the results of cell binding and internalization experiments were similar for both radiolabeled peptides (Figure 2). Differences in specific molar activities (MBq/mol) could account for this observation and cannot be entirely excluded even though both 99m Tc-radiopeptides were prepared under identical conditions. Because this study aims at the investigation of differences in the pharmacokinetics of the radiopeptides (e.g., rate and route of excretion) rather than in tumor uptake, we continued with the evaluation of both compounds in vivo.  Table 2) [23]. Some variability in experimentally determined Kd and Bmax of structurally related radiolabeled peptides is not uncommon [18,23,27], yet in this case somewhat surprising as we did not expect the sorbitol moiety to have an influence on these parameters; this in particular because the results of cell binding and internalization experiments were similar for both radiolabeled peptides ( Figure 2). Differences in specific molar activities (MBq/mol) could account for this observation and cannot be entirely excluded even though both 99m Tc-radiopeptides were prepared under identical conditions. Because this study aims at the investigation of differences in the pharmacokinetics of the radiopeptides (e.g., rate and route of excretion) rather than in tumor uptake, we continued with the evaluation of both compounds in vivo.

In Vivo Evaluation
Biodistribution studies were performed in nude mice bearing PC-3 tumor xenografts at 1 h postinjection (p.i.) of the radiotracers (n = 6; Table 3, Figure 4). Table 3. Biodistribution data of radiolabeled peptide conjugates in nude mice bearing PC-3 xenografts.   in the GRPR positive pancreas and in the colon. Very low accumulation of radioactivity was observed in receptor-negative tissues and organs (e.g., the uptake in muscle and bone was less than 0.5% ID/g, Table 3). Blocking experiments resulted in a significant decrease of accumulation of radioactivity in GRPR-positive tumors and organs therefore demonstrating receptor specific uptake of the radiotracer (n = 5; Table 3, Figure 4). No significant differences in tumor uptake were observed as expected based on the in vitro cell binding and internalization experiments (Figure 2, Table 2). Despite glycation, the pharmacokinetics of [ 99m Tc][Tc(CO) 3 (12)] was not significantly altered in comparison to [ 99m Tc][Tc(CO) 3 (13)] and the enhanced renal excretion and lowered uptake in the liver was not as pronounced as expected. Therefore, the glycated radiopeptide did not perform significantly better in vivo. Nevertheless, as hypothesized, [ 99m Tc][Tc(CO) 3 (12)] bearing a sorbitol moiety did not show a pronounced accumulation in liver and spleen compared to the previously published glucuronated BBN derivative [18].

Materials and Methods
Caution: 99m Tc is a γ-emitter (140 keV) with a half-life of 6.01 h. All reactions involving 99m Tc were performed in a laboratory approved for the handling of radionuclides and appropriate safety procedures were followed at all times to prevent contamination.

General Procedures
Solvents and all other chemicals of at least of synthesis grade were purchased from B. Braun, Sigma-Aldrich, and Bachem. Buffers and stock solutions were prepared using Millipore water. [ 99m Tc]NaTcO 4 was eluted from an Ultra-TechneKow or TEKCIS 99 Mo/ 99m Tc generator. The precursor [ 99m Tc][Tc(CO) 3 (H 2 O) 3 ] + was prepared by using the CRS kit for 99m Tc-tricarbonyl (Center for Radiopharmaceutical Sciences, Paul Scherrer Institute, Villigen, Switzerland). The radiolabeling of the peptides was performed with a Biotage ® Initiator+ microwave using a 200-500 µL glass reactor or with a heating block in 1.5 mL LoProtein bind Eppendorf vessels. HPLC analyses were carried out with an Agilent system (Vienna, Austria) equipped with Autosampler Agilent 1100 Series, Iso Pump (Isocratic Pump) Agilent 1200 Series G1310A, UV-Monitor Agilent 1200 Series G1314B Variable Wavelength Detector (VWD) and Radioactivity Detector Elysia Raytest Gabi Star. Data acquisition and gradient control were performed using GINA StarTM, version 5.9. HPLC solvents were 0.1% TFA in H 2 O (A) and MeCN (B). Quality control of the radiometal labeled peptide was performed using a C12 Phenomenex Jupiter column (4u Proteo 90 Å, 4 µm, 250 × 4,6 mm) and a linear gradient from 80% to 50% of eluent A in 16 min with a flow rate of 1 mL/min. Low resolution (LR)-MS was performed on a Bruker maXis (UHR-TOF, Vienna, Austria) equipped with ESI ion source and Qq-TOF analyser or Bruker amaZon speed ETD supplied with ESI ion source and 3D ion trap. Sample centrifugation was done by Hettich Universal 30 RF with Hettich rotor 1412 24 × 3g.
In vitro and in vivo experiments were carried out with radiolabeled peptides with a purity of at least 95% (γ-HPLC). Samples were γ-counted for 30 sec using an energy window of 104-170 keV (2480 Wizard 2 , PerkinElmer, Waltham, Mass., USA).

Syntheses
Synthesis of compound 3. The compound was prepared according to a published procedure ref. [22] [17], the formation of rotamers due to the presence of the trisubstituted N(α)amine of the Lysine resulted in the duplication and/or broadening of NMR signals; thus, the compound was fully characterized after the Boc-deprotection (see compound 9).
Synthesis of pentaacetate 9. Compound 8 (80 mg, 0.11 mmol) was dissolved in CH 2 Cl 2 /TFA (3:1, 4 mL) and kept at rt for 3 h after which time TLC indicated completed conversion. The solution was evaporated under reduced pressure, the residue was dissolved in MeOH (2 mL), filtered through Celite TM and dried in vacuo to provide the TFA salt of compound 9 as a colorless oil (75 mg, 93%).  Table 1 and SM, Figure S16) and was at least 95%.

TLC
The purity of the 99m Tc-labeled peptides was additionally confirmed by thin-layer chromatography (TLC) using silica gel 60 F 254 aluminum plates (Merck, Darmstadt, Germany), as stationary phase (strips height 10 cm with 8 cm between application point and solvent front). A mixture 95.5:0.5 of MeOH:HCl 6 M was used as mobile phase. In a typical assay, 2 µL drops of freshly prepared [ 99m Tc][Tc(CO) 3 (12)] (n = 2 in duplicates) divided by the sum of counts in all ROIs.

LogD
The lipophilicities (LogD) of [ 99m Tc][Tc(CO) 3 (L)] (L = 12, 13) were determined by their partition coefficient between n-octanol and PBS (pH 7.4) utilizing the "shake-flask method" [25,26]. PBS and n-octanol were shaken overnight to saturate each phase. After separation of the layers by gravity, equal volumes (500 µL) of each layer were taken and transferred into an Eppendorf tube and 5 µL (~45 kBq) of the radiolabeled peptide solution were added to the PBS/n-octanol mixture. The resulting solutions were mixed in a shaker at room temperature for 20 min and centrifuged at 3000 rpm for 10 min. Aliquots of 300 µL were removed from the octanol and PBS phases and the radioactivity was measured in the γ-counter (n = 3 in triplicates). The lipophilicity was calculated as the log value of the average ratio between the radioactivity in the organic fraction (octanol) and the PBS fraction from the samples.

General Methods
Cell culture reagents were purchased from Thermo Fisher Scientific (Vienna, Austria) or Sigma-Aldrich (Vienna, Austria). Human prostate adenocarcinoma cells (PC-3) were purchased from ATCC and cultivated in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS), 2 mM L-glutamine, 100 units/mL penicillin and 100 µg/mL streptomycin under humidified atmosphere (37 • C, 5% CO 2 ) until minimum 80% at confluence. In order to determine the number of cells for each experiment, three extra wells were seeded for parallel cell counting using the Neubauer chamber (Ratiomed) and Trypan Blue solution, 0.4% (Gibco ®® ). Percentage of applied dose was normalized to 10 6 cells/well.

Internalization Studies
Internalization studies were performed as previously published [23]. In brief, approximately 10 6 PC-3 cells were seeded in 1% FBS RPMI-medium (RPMI-1640 medium containing 1% FBS, 2 mM L-glutamine, 100 units/mL penicillin and 100 µg/mL streptomycin) in 6-well plates on the day before the experiment and incubated overnight in humidified incubator (37 • C, 5% CO 2 ). Approx. one hour before the experiment, the medium was replaced with 1.3 mL fresh 1% FBS RPMI-medium. Subsequently, 100 µL of radiolabeled peptide [ 99m Tc][Tc(CO) 3 (L)] (L = 12, 13; 0.25 pmol;~1.0 kBq) were added to each well and cells were incubated for different time points (10, 60, and 120 min). Non-specific receptor binding was determined by incubating the cells with the radiotracers and 1000-fold excess of natural bombesin (250 pmol; 100 µL per well) as the receptor blocking agent. After the respective incubation time, the supernatant was collected and the cells were washed twice with 1 mL ice-cold DPBS (Dulbecco's phosphate-buffered saline). The combined fractions represent the unbound radiopeptide. The receptor-bound radioactivity was obtained by incubating the cells twice for 5 min with an ice-cold acidic glycine solution (100 mM NaCl, 50 mM glycine, pH 2.8; 1 mL) on ice followed by removal of the supernatant. Finally, the internalized fraction was collected after cell lysis using 1 M NaOH (1 mL; 10 min; 37 • C, 5% CO 2 ) and the wells containing the cell lysate were washed twice with NaOH (1 M, 1 mL). Standards of the radiolabeled peptides [ 99m Tc][Tc(CO) 3 (L)] (L = 12, 13; per well: 100 µL; 0.25 pmol;~1.0 kBq) for determination of the applied dose were prepared in triplicate. All fractions were measured in a γ-counter and calculated as percentage of the applied dose (n = 2-3 in triplicate).

Receptor Saturation Assay
Receptor binding assays were performed as previously published [23]. PC-3 cells were prepared in 6-well plates as described above for the internalization experiments. Approx. one hour before the experiment, the medium was replaced with 0.8 mL fresh 1% FBS RPMI-medium. PC-3 cells were kept on ice for 30 min to stop receptor internalization processes prior to the start of the experiment. Afterwards, the cells were incubated at 4 • C with increasing concentration (1, 5, 10, 25, 50, 75, and 100 nM in NaCl; 100 µL/well, 5-600 kBq) of the radiolabeled peptides [ 99m Tc][Tc(CO) 3 (L)] (L = 12, 13) to allow receptor saturation. Non-specific receptor binding was determined using an excess of natural bombesin (2.5 µM/well for concentrations of the radiopeptide <10 nM, and 10 µM/well for higher concentrations). After incubation at 4 • C for 2 h, the supernatant was collected and the cells were washed twice with 1 mL of ice-cold DPBS. The combined fractions represent the unbound radiopeptide. To determine the cell-bound fraction, the cells were lysed with 1 M NaOH (1 mL; 10 min; 37 • C, 5% CO 2 ) and the wells were washed twice with 1 M NaOH (1 mL). The obtained fractions were measured in a γ-counter. Dissociation constants (K d ) and maximum receptor occupancy (B max ) were calculated from the data for specific binding with nonlinear regression using GraphPad Prism 7 (n = 2-3 in triplicate).

Biodistribution Experiments
All animals were treated according to the European Union rules on animal care. Animal experiments were approved by the Austrian Ministry of Sciences (BMBWF-66.009/0122-V/3b/2019). Approximately 5 × 10 6 PC-3 cells in 100 µL serum-free medium were subcutaneously injected into flanks of 8-week-old, female, athymic mice (Fox-1 nu , Charles River Laboratories). Xenografts were allowed to grow for 16 days (approx. size: 250 mm 3 ) and mice were intravenously injected via the tail vein with 0.1 µM radiolabeled peptide (10 pmol,~42 kBq, 100 µL physiological saline) alone (baseline experiments) or co-injected with 2000-fold excess BBN(1-14) (20 nmol; blocking experiments). One hour p.i. of the radiotracers, mice were sacrificed and organs, tumor and blood were removed. Radioactivity in organs and tissues was quantified using a γ-counter (2480 Wizard 2 , PerkinElmer). Organs and tissues were wet-weighted and percentage of injected dose per gram was calculated (% ID/g). Statistical analysis (2-way ANOVA) was performed by GraphPad Prism.

Conclusions
We have applied the "Click-to-Chelate" methodology to the efficient synthesis of a 99m Tc-tricarbonyl labeled tumor-targeting BBN conjugate that contains sorbitol as an acyclic, reduced carbohydrate modifier for increasing the hydrophilicity of the radiopeptide and, as a result, to promote renal elimination in vivo. Indeed, addition of sorbitol to the 99m Tc-labeled BBN derivative increased its hydrophilicity (logD) by 0.7 log units in comparison to a non-glycated reference compound. However, both compounds showed comparable pharmacokinetic profiles in vivo with regards to rate and route of excretion. For example, unspecific uptake of the radiotracer in the excretory organs kidneys and liver was found almost identical for both compounds despite the increased hydrophilicity of the glycated peptide. This puzzling result might be explained by the inherent properties of the studied peptide, [Nle 14 ]BBN (7)(8)(9)(10)(11)(12)(13)(14), and/or its metabolites. Indeed, low unspecific uptake in the liver has been reported for other related agonistic BBN derivatives radiolabeled with the 99m Tc-tricarbonyl core [28]. However, this phenomenon will need to be verified with other peptides or (bio)molecules of interest for nuclear medicine. Further investigations in this direction are currently ongoing and will be reported in due time.
The herein reported sorbitol-containing lysine derivative 10 (Scheme 1) represents a promising general alkyne precursor for the "Click-to-Chelate" approach which might proof useful for overcoming the often stated but yet unresolved issue of the increased lipophilicity of radiotracers resulting from the radiolabeling with the 99m Tc-tricarbonyl core [6].