Article
`
`pubs.acs.org/bc
`
`68Ga-Complex Lipophilicity and the Targeting Property of a Urea-
`Based PSMA Inhibitor for PET Imaging
`†
`†
`Matthias Eder,*,†
`Martin Schäfer,
`Ulrike Bauder-Wüst,
`William-Edmund Hull,
`†
`∥
`∥
`Walter Mier,
`Uwe Haberkorn,
`and Michael Eisenhut
`†
`‡
`Radiopharmaceutical Chemistry,
`Molecular Structure Analysis, German Cancer Research Center, Heidelberg, Germany
`§Department of Nuclear Medicine, University Hospital Munich, Munich, Germany
`∥
`Department of Nuclear Medicine, University of Heidelberg, Heidelberg, Germany
`*S Supporting Information
`
`‡
`
`Carmen Wängler,
`
`§
`
`ABSTRACT: Urea-based inhibitors of
`the prostate-specific
`low-molecular-weight
`membrane antigen (PSMA) represent
`pepidomimetics showing the ability to image PSMA-expressing
`prostate tumors. The highly efficient, acyclic Ga(III) chelator
`N,N′-bis [2-hydroxy-5-(carboxyethyl)benzyl] ethylenediamine-
`N,N′- diacetic acid (HBED-CC) was introduced as a lipophilic
`side chain into the hydrophilic pharmacophore Glu-NH-CO-NH-
`Lys which was found favorable to interact with the PSMA “active
`binding site”. This report describes the syntheses, in vitro binding
`analyses, and biodistribution data of the radiogallium labeled
`PSMA inhibitor Glu-NH-CO-NH-Lys(Ahx)-HBED-CC in com-
`parison to the corresponding DOTA conjugate. The binding
`properties were analyzed using competitive cell binding and
`enzyme-based assays followed by internalization experiments. Compared to the DOTA-conjugate, the HBED-CC derivative
`showed reduced unspecific binding and considerable higher specific internalization in LNCaP cells. The 68Ga complex of the
`HBED-CC ligand exhibited higher specificity for PSMA expressing tumor cells resulting in improved in vivo properties. 68Ga
`labeled Glu-NH-CO-NH-Lys(Ahx)-HBED-CC showed fast blood and organ clearances, low liver accumulation, and high specific
`uptake in PSMA expressing organs and tumor. It could be demonstrated that the PET-imaging property of a urea-based PSMA
`inhibitor could significantly be improved with HBED-CC.
`
`■ INTRODUCTION
`
`It is still a challenge to select appropriate treatment options for
`disseminated prostate cancer because of the lack of sensitive
`imaging agents for diagnosis and therapy monitoring.1 Prostate-
`specific membrane antigen (PSMA) is expressed in nearly all
`prostate cancers with increased expression in poorly differ-
`entiated, metastatic, and hormone-refractory carcinomas.2 Its
`expression level
`is about 1000-fold higher compared to the
`physiologic levels found in other tissues such as kidney, small
`intestine, or brain.3 PSMA is primarily restricted to the prostate,
`it is abundantly expressed at all stages of disease, it is presented
`on the cell surface, and it is not shed into the circulation.4 As a
`consequence, PSMA can be considered a promising target for
`specific prostate cancer imaging and therapy.5−7
`The radiohalogenated inhibitors of PSMA exhibiting Glu-
`NH-CO-NH-Lys as a pharmacophore showed the ability to
`image PSMA-expressing prostate tumor xenografts.8,9 More
`recently, a corresponding DOTA-conjugate was labeled with
`68Ga which represents an attractive generator-based alternative
`to cyclotron-based PET radiopharmaceuticals.10,11 The readily
`available 68Ga decays with 89% probability by positron emission
`allowing high-resolution PET images with the option of
`
`accurate quantification. N,N′-Bis[2-hydroxy-5-(carboxyethyl)-
`benzyl]ethylenediamine-N,N′- diacetic acid (HBED-CC) was
`68Ga chelator with fast
`recently proposed as an efficient
`complexing kinetics and a high in vitro as well as in vivo
`complex stability.12,13
`Besides the efficient Ga(III) complexing characteristics,
`HBED-CC was selected because of
`its lipophilic nature. It
`was found that the “active binding site” of PSMA is composed
`of two structural motifs, one representing a lipophilic pocket
`and the other interacting with urea-based inhibitors.14 A study
`comparing a series of linkers located between the urea-based
`motif and a 99mTc chelator revealed a clear dependency of
`binding properties in favor of the more lipophilic compounds.14
`Further
`studies described the binding site as a pocket
`interacting with the carboxylic groups and the zinc complexing
`urea on one hand and with hydrophobic, mostly aromatic
`groups on the other hand.15
`
`Received:
`June 1, 2011
`Revised:
`February 6, 2012
`Published: February 28, 2012
`
`© 2012 American Chemical Society
`
`688
`
`dx.doi.org/10.1021/bc200279b | Bioconjugate Chem. 2012, 23, 688−697
`
`Downloaded via FURMAN DEMARIS on March 1, 2024 at 20:32:23 (UTC).
`
`See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
`
`Petitioner GE Healthcare – Ex. 1030, p. 688
`
`

`

`Bioconjugate Chemistry
`
`Table 1. Analytical and PSMA-Binding Data
`
`Article
`
`affinity related Ki [nM] determined in
`affinity related IC50 [nM] determined in
`analytical HPLC
`m/z calculated as
`Ga-peptide
`cell-based assayc
`enzyme-based assayc
`retentionb
`m/za
`[M+H]+
`complexes
`12.0 ± 2.8
`7.5 ± 2.2
`[Ga]7
`3.1 min
`947.4250
`947.4257
`37.6 ± 14.3
`19.4 ± 7.1
`[Ga]8
`2.4 min
`819.4100
`819.4104
`11.1 ± 1.8
`8.7 ± 3.9
`[Ga]12
`3.7 min
`1284.6364
`1284.6368
`aHigh-resolution mass spectrometry data of the free ligands ([M+H]+). bCompounds were labeled with 68Ga; runs were performed using a linear
`A−B gradient (0% B to 100% B in 6 min) at a flow rate of 4 mL/min. Solvent A was 0.1% aqueous TFA and solvent B was MeOH. cCompounds
`were complexed with natGa.
`
`This report describes the syntheses of Glu-NH-CO-NH-
`Lys(Ahx)-HBED-CC (7) and Glu-NH-CO-NH-Lys(Ahx)-
`DOTA (8) together with a reference PSMA inhibitor
`previously published by Banerjee et al.10 After 68Ga complex-
`ation, these compounds were evaluated in vitro and in vivo to
`study the influence of these two chelators and side chain
`variations.
`
`■ MATERIALS AND METHODS
`Reagents. All commercially available chemicals were of
`analytical grade and were used without further purification.
`68Ga was obtained from a 68Ge/68Ga generator based on
`resin support.12,16 Protected amino acids were
`pyrogallol
`obtained from Novabiochem (Merck, Darmstadt, Germany)
`or IRIS Biotech (Marktredwitz, Germany).
`The preparations were analyzed using reversed-phase high-
`performance liquid chromatography (RP-HPLC; Chromolith
`RP-18e, 100 × 4.6 mm; Merck, Darmstadt, Germany). Runs
`were performed using a linear A−B gradient (0% B to 100% B
`in 6 min) at a flow rate of 4 mL/min. Solvent A consisted of
`0.1% aqueous TFA and solvent B was 0.1% TFA in CH3CN or
`MeOH (in case of determination of radiochemical yield (RCY)
`or analysis of serum stability).
`For all preparative purifications, a Chromolith RP-18e
`column (100 × 10 mm; Merck, Darmstadt, Germany) was
`used with a 6 min gradient starting at 0% raised to 50% and
`followed by a 1 min increase to 100% B. The flow rate was 6
`mL/min. The HPLC system (L6200 A; Merck-Hitachi,
`Darmstadt, Germany) was equipped with a variable UV and a
`gamma detector (Bioscan; Washington, USA). UV absorbances
`were measured at 214 and 254 nm. Mass spectrometry was
`performed with a MALDI-MS Daltonics Microflex (Bruker
`Daltonics, Bremen, Germany) system. High-resolution mass
`spectrometry was performed using a system equipped with a
`mass spectrometer supporting Orbitrap technology (Exactive,
`Thermo Fisher Scientific). Full-scan single mass spectra were
`obtained by scanning m/z = 200−4000. NMR data were
`obtained with Bruker Avance NMR Spectrometers, AV(I)-600
`(600 MHz) and AV(III)-400 (400 MHz). The chemical shifts
`are referenced to solvent signals (DMSO-d5 = 2.50/39.757
`ppm, DMSO-d6 = 39.50).
`Synthesis of Glu-NH-CO-NH-Lys(Ahx)-HBED-CC (7). In
`a first step, the isocyanate 2 of the glutamyl moiety was
`generated in situ by adding a mixture of 3 mmol of bis(tert-
`butyl) L-glutamate hydrochloride (Bachem, Switzerland) (1)
`and 1.5 mL of N-ethyldiisopropylamine (DIPEA) in 200 mL of
`dry CH2Cl2 to a solution of 1 mmol triphosgene in 10 mL of
`dry CH2Cl2 at 0 °C over 4 h. After agitation of the reaction
`mixture for one further hour at 25 °C, 0.5 mmol of a resin-
`immobilized (2-chloro-tritylresin, Merck, Darmstadt) ε-allylox-
`ycarbonyl protected lysine was added in one portion (in 4 mL
`DCM) and reacted for 16 h with gentle agitation leading to
`
`compound 3. The resin was filtered off and the allyloxy-
`protecting group was
`removed using 100 mg tetrakis-
`(triphenyl)palladium(0) (Sigma-Aldrich, Germany) and 400
`μL morpholine in 4 mL CH2Cl2 for 3 h resulting in compound
`4. The following coupling of the aminohexanoic moiety was
`performed using 2 mmol of the Fmoc-protected 6-amino-
`hexanoic acid (Sigma-Aldrich, Germany), 1.96 mmol of HBTU
`(Merck, Darmstadt, Germany), and 2 mmol of N-ethyl-
`diisopropylamine in a final volume of 4 mL DMF. The product
`5 was cleaved from the resin by reacting with 4 mL of a 30%
`1,1,1,3,3,3-hexafluoroisopropanole (HFIP) in CH2Cl2 for two
`hours at ambient
`temperature resulting in the tert-butyl
`protected crude product 6 which was purified via RP-HPLC.
`To conjugate HBED-CC, the purified product 6 was reacted
`with an equimolar amount of HBED-CC-TFP-ester synthesized
`as previously described12 in the presence of 2 equiv of DIPEA
`in N,N-dimethylformamide (DMF;
`final volume of 1 mL).
`After HPLC purification, the remaining tert-butyl groups were
`cleaved at
`room temperature for one hour using 2 mL
`trifluoroacetic acid to obtain 7 in ∼35% yield after purification
`by HPLC. High-resolution mass spectrometry was used to
`confirm the identity (Table 1), and purity was analyzed via
`analytical HPLC at λ = 206 nm (Supporting Information).
`Complete and unequivocal NMR 1H and 13C signal assign-
`ments were obtained (Supporting Information).
`Synthesis of Glu-NH-CO-NH-Lys(Ahx)-DOTA (8). Pre-
`cursor 5 was synthesized as mentioned before. After activation
`with 3.95 equiv of HBTU and DIPEA for 2 h, 4 equiv of tris(t-
`bu)-DOTA (Chematech, Dijon, France) relative to the resin
`loading were reacted with 5 after removal of
`the Fmoc-
`protecting group in a final volume of 3 mL DMF. The product
`was cleaved from the resin in a 2 mL mixture consisting of
`trifluoroacetic acid, triisopropylsilane, and water (95:2.5:2.5).
`The product was purified via semipreparative RP-HPLC
`resulting in compound 8 (∼30% yield). High-resolution mass
`spectrometry was used to confirm the identity (Table 1) and
`purity was analyzed via analytical HPLC at λ = 206 nm
`(Supporting Information). Complete and unequivocal 1H and
`13C NMR signal assignments were obtained (Supporting
`Information).
`Synthesis of Reference Compound (12). Synthesis of
`the DOTA conjugate previously published by Banerjee et al.10
`was performed using a modified solid-phase method. Five
`equivalents of
`suberic-acid-bis-(N-hydroxysuccinimideester)
`(Sigma-Aldich, Germany) was conjugated to the side chain of
`the lysine moiety of compound 4 in the presence of 5 equiv of
`triethylamine in 3 mL DMF. After 2 h, the resin was washed
`with DMF. In the presence of five equiv of triethylamine, five
`equiv of Fmoc-Lys-oAll was coupled to the immobilized NHS-
`ester of intermediate 9 in a final volume of 3 mL DMF for 16 h
`resulting in 10. The remaining conjugations of two phenyl-
`alanine building blocks and tris(t-bu)-DOTA were performed
`
`689
`
`dx.doi.org/10.1021/bc200279b | Bioconjugate Chem. 2012, 23, 688−697
`
`Petitioner GE Healthcare – Ex. 1030, p. 689
`
`

`

`Bioconjugate Chemistry
`
`according to standard Fmoc-protocols resulting in 11. Cleavage
`from resin was performed using a 3 mL mixture of
`trifluoroacetic acid, triisopropylsilane, and water (95:2.5:2.5).
`The product was purified via RP-HPLC to obtain 12 in ∼8%
`yield. High-resolution mass spectrometry was used to confirm
`the identity (Table 1), and purity was analyzed via analytical
`HPLC at λ = 206 nm (Supporting Information).
`68Ga-Labeling. Typically, 0.1−1 nmol of Glu-NH-CO-NH-
`Lys(Ahx)-HBED-CC (7, in 0.1 M HEPES buffer pH 7.5) or 1
`nmol of the DOTA conjugates (8/12, in 0.1 M HEPES buffer
`pH 7.5) were added in a volume of 100 μL to a mixture of 10
`μL 2.1 M HEPES solution and 10 μL [68Ga]Ga3+ eluate (50−
`100 MBq). The pH of the labeling solution was adjusted to 4.2.
`Depending on the chelator, the reaction mixture was incubated
`either at ambient temperature or at 80 °C for 2 min. The
`radiochemical yield (RCY) was determined using RP-HPLC.
`67Ga-Labeling. 67Ga was purchased from MDS Nordion
`(Fleurus, Belgium) as [67Ga]GaCl3 in 0.1 N HCl. Typically, 0.1
`nmol of Glu-NH-CO-NH-Lys(Ahx)-HBED-CC (7) was added
`in a volume of 100 μL to a mixture of 10 μL 2.1 M HEPES
`solution, 2 μL 1 N HCl, and 10 μL [67Ga]GaCl3 (∼100 MBq)
`in 0.1 N HCl resulting in a solution with a pH of 4.2. The
`reaction mixture was
`incubated for 2 min at ambient
`temperature. The RCY was determined using RP-HPLC.
`natGa-Complexes. A 10× molar excess of Ga(III)-nitrate
`(Sigma Aldrich, Germany) in 0.1 N HCl (10 μL) was reacted
`with the compounds 7, 8, or 12 (1 mM in 0.1 M HEPES buffer
`pH 7.5, 40 μL) in a mixture of 10 μL 2.1 M HEPES solution
`and 2 μL 1 N HCl for 2 min at 80 °C.
`Radiochemical Stability. The radiochemical stability of
`the 68Ga-labeled compounds 7, 8, and 12 was determined by
`incubating in both PBS and human serum for 2 h at 37 °C. An
`equal volume of MeCN was added to the samples to precipitate
`serum proteins. Subsequently, the samples were centrifuged for
`5 min at 13 000 rpm. An aliquot of the supernatant and the PBS
`sample was analyzed by RP-HPLC. In addition, serum samples
`of compound [68Ga]7 were run on a Superdex 200 GL 5/150
`gel filtration column (GE Healthcare, Munich, Germany) to
`analyze protein binding. To analyze the complex stability
`against human transferrin, a 400 μL aliquot of [68Ga]7 was
`added to 250 μg apo-transferrin in PBS at pH 7 and incubated
`at 37 °C (water bath) for 2 h. The complex stability was
`determined using a Superdex 200 GL 5/150 short column with
`PBS pH 7 as eluent.
`Naaladase Assay. Recombinant human PSMA (rhPSMA,
`R&D systems, Wiesbaden, Germany) was diluted in assay
`buffer (50 mM HEPES, 0.1 M NaCl, pH 7.5) to 0.4 μg/mL.
`The substrate Ac-Asp-Glu (Sigma, Taufkirchen, Germany, 40
`μM final concentration) was mixed with [natGa]7, [natGa]8, or
`[natGa]12 at concentrations ranging from 0.05 nM to 1000 nM
`in a final volume of 125 μL assay buffer. The mixtures were
`combined with 125 μL of the rhPSMA solution (0.4 μg/mL)
`and incubated for one hour at 37 °C. The reaction was stopped
`by heating at 95 °C for 5 min. 250 μL of a 15 mM solution of
`ortho-phthaldialdehyde (Sigma, Taufkirchen, Germany) was
`added to all vials and incubated for 10 min at ambient
`temperature. Finally, 200 μL aliquots of the reaction solutions
`were loaded onto a F16 Black Maxisorp Plate (Nunc,
`Langenselbold, Germany) and read at excitation and emission
`wavelengths of 330 and 450 nm,
`respectively, using a
`microplate reader (DTX-880, Beckman Coulter, Krefeld,
`Germany). The data were analyzed by a one-site total binding
`
`Article
`
`regression algorithm of GraphPad Prism (GraphPad Software,
`California, USA).
`Cell Binding Studies. Cell binding studies were performed
`using PSMA+ LNCaP cells (metastatic lesion of human
`prostatic adenocarcinoma, ATCC CRL-1740) and PSMA−
`PC-3 cells (bone metastasis of a grade IV prostatic
`adenocarcinoma, ATCC CRL-1435) cultured in DMEM
`medium supplemented with 10% fetal calf serum and 2
`mmol/L L-glutamine (all from Invitrogen). During cell culture,
`cells were grown at 37 °C in an incubator with humidified air,
`equilibrated with 5% CO2. The cells were harvested using
`trypsin-ethylenediaminetetraacetic acid (trypsin-EDTA; 0.25%
`trypsin, 0.02% EDTA, all from Invitrogen).
`In order to determine the binding affinity, a competitive cell
`binding assay was performed. LNCaP cells (105 per well) were
`incubated with a 0.2 nM solution of [67Ga]7 in the presence of
`12 different concentrations of [natGa]7, [natGa]8, or [natGa]12
`(0−5000 nM, 200 μL/well). After incubation at ambient
`temperature for 1 h with gentle agitation, the binding buffer
`was removed using a multiscreen vacuum manifold (Millipore,
`Billerica, MA). After washing twice with 100 μL and once with
`200 μL ice-cold binding buffer, the cell-bound radioactivity was
`measured with a gamma counter (Packard Cobra II, GMI,
`Minnesota, USA). The 50% inhibitory concentration (IC50)
`values were calculated by fitting the data using a nonlinear
`regression algorithm (GraphPad Software). Experiments were
`performed at least three times including quadruplicate sample
`measurements.
`Determination of Binding Specificity and Internal-
`ization.
`Internalization experiments were performed as
`previously described.17 Briefly, 105 LNCaP or PC-3 cells were
`seeded in poly(L-lysine)-coated 24-well cell culture plates 24 h
`before incubation. After washing with PBS, the cells were
`incubated with the radiolabeled compounds [68Ga]7, [68Ga]8,
`and [68Ga]12 (25 nM final concentration) for 45 min at 37 °C
`and at 4 °C, respectively. To determine specific cell uptake,
`cells were blocked with 2-(phosphonomethyl)-pentanedioic
`acid (PMPA, Axxora, Loerrach, Germany) to a final
`concentration of 100 μM. Cellular uptake was terminated by
`washing 4 times with 1 mL of ice-cold PBS. To remove surface-
`bound radioactivity, cells were incubated twice with 0.5 mL
`glycine-HCl
`in PBS (50 mM, pH 2.8) for 5 min at room
`temperature. The cells were washed with 1 mL of ice-cold PBS
`and lysed using 0.5 mL of 0.3 M NaOH. The surface-bound
`and internalized fractions were measured in a gamma counter.
`Specificity of binding was additionally analyzed in a
`concentration-dependent cell uptake experiment. Solutions of
`[68Ga]7, [68Ga]8, or [68Ga]12 at final concentrations of 2.5, 25,
`and 250 nM were added to 7 × 105 cells suspended in 50 μL
`OPTIMEM medium (Gibco, Auckland, New Zealand). After
`45 min at 37 °C, the samples were briefly vortexed and a 10 μL
`aliquot was transferred to a 400 μL microcentrifuge tube (Roth,
`Germany) containing 350 μL of a 75:25 mixture of silicon oil,
`density 1.05 (Sigma Aldrich, Germany), and mineral oil, density
`0.872 (Acros, Nidderau, Germany). Separation of cells from the
`medium was performed by centrifugation at 12 000 rpm for 2
`min. After freezing the tubes using liquid nitrogen, the bottom
`tips containing the cell pellet were cut off. The cell pellets and
`the supernatants were separately counted in a gamma counter.
`Biodistribution and PET Imaging. Five ×106 cells of
`either LNCaP or PC-3 in 50% Matrigel (Becton Dickinson,
`Heidelberg, Germany) were subcutaneously inoculated into the
`right trunk of male 7- to 8-week-old BALB/c nu/nu mice
`
`690
`
`dx.doi.org/10.1021/bc200279b | Bioconjugate Chem. 2012, 23, 688−697
`
`Petitioner GE Healthcare – Ex. 1030, p. 690
`
`

`

`Bioconjugate Chemistry
`
`Article
`
`Scheme 1. Syntheses of Glu-NH-CO-NH-Lys(Ahx)-HBED-CC (7) and Glu-NH-CO-NH-Lys(Ahx)-DOTA (8)
`
`(a) Triphosgene, DIPEA, CH2Cl2, 0 °C; (b) H-Lys(Alloc)-2CT-Resin, CH2Cl2; (c) Pd[P(C6H5)3]4, morpholine, CH2Cl2; (d) Fmoc-6-Ahx-OH,
`HBTU, DIPEA, DMF; (e) 20% piperidine, DMF; (f) hexafluoroisopropanol/CH2Cl2; (g) HBED-CC-TFP ester, DIPEA, DMF; (h) TFA; (i) tris(t-
`bu)DOTA, HBTU, DIPEA; (j) TFA.
`
`(Charles River Laboratories). The tumors were allowed to grow
`for 3 to 4 weeks until approximately 1 cm3 in size.
`The 68Ga-radiolabeled compounds of 7, 8, and 12 were
`injected via tail vein (1−2 MBq per mouse; 0.1−0.2 nmol). At
`1 h after injection, the animals were sacrificed. Organs of
`interest were dissected, blotted dry, and weighed. The
`radioactivity was measured with a gamma counter and
`calculated as % ID/g.
`For the microPET studies, 10−25 MBq of compounds
`[68Ga]7 and [68Ga]8 in a volume of 0.15 mL (∼0.5 nmol) were
`injected via a lateral tail vein into mice bearing LNCaP tumor
`xenografts. The anesthetized animals (2% sevoflurane, Abbott,
`Wiesbaden, Germany) were placed in prone position into the
`Inveon small animal PET scanner (Siemens, Knoxville, Tenn,
`USA) to perform a 50 min dynamic microPET scan starting at
`1 min post injection followed by a 20 min static scan.
`Statistical Aspects. All experiments were performed at
`least in triplicate. Quantitative data were expressed as mean ±
`SD. If applicable, means were compared using Student’s t-test.
`P values of <0.05 were considered statistically significant.
`
`■ RESULTS
`Chemistry. The synthesis of the resin-bound intermediate 5
`was performed using solid-phase chemistry as outlined in
`Scheme 1. To couple the respective chelators, compound 5 was
`either cleaved from resin and reacted with an equimolar
`amount of HBED-CC-TFP-ester resulting in 7 or coupled with
`HBTU activated tris(t-bu)DOTA resulting in the DOTA-
`conjugate 8. The preparation of the reference compound 12
`previously published by Banerjee et al.10 was performed using a
`modified solid-phase method as outlined in Scheme 2. Briefly,
`disuccinimidyl suberate was conjugated to the ε-amino group of
`the lysine group of 4. After Fmoc-Lys-OAll was coupled to the
`resulting immobilized NHS-ester,
`the conjugation of
`two
`phenylalanine building blocks and tris(t-bu)-DOTA were
`performed according to standard Fmoc-protocols. Analytical
`data of compounds 7, 8, and 12 are summarized in Table 1.
`Radiolabeling and Stability. Radiolabeling with 68Ga was
`performed at pH 4.2 by incubating the conjugates 7, 8, or 12 in
`a mixture consisting of 50−100 MBq [68Ga]Ga3+ and HEPES.
`Radiochemical yields were determined by RP-HPLC analysis of
`the reaction mixtures. The retention times obtained for each
`compound are shown in Table 1. An amount of 0.1 nmol of the
`
`691
`
`dx.doi.org/10.1021/bc200279b | Bioconjugate Chem. 2012, 23, 688−697
`
`Petitioner GE Healthcare – Ex. 1030, p. 691
`
`

`

`Bioconjugate Chemistry
`
`Article
`
`Scheme 2. Modified Solid-Phase Synthesis of the DOTA Conjugate Previously Published by Banerjee et al.10
`
`(a) Disuccinimidyl suberate (DSS), TEA, DMF; (b) Fmoc-Lys-oAll, TEA, DMF; (c) 20% piperidine, DMF; (d) Fmoc-Phe-OH, HBTU, DIPEA,
`DMF; (e) tris(t-bu)DOTA, HBTU, DMF; (f) Pd[P(C6H5)3]4, morpholine, CH2Cl2; (g) TFA, triisopropylsilane, H2O (95/2.5/2.5).
`
`Figure 1. Determination of binding affinity of compounds [natGa]7, [natGa]8, and [natGa]12 by competitive titration on LNCaP cells (A) and
`purified receptor using an enzyme-based assay (B). Data are expressed as mean ± SD (n = 4). cpm = counts per minute; FI = fluorescence intensity.
`
`HBED-CC conjugate 7 in a final concentration of 1.7 μM led
`to a radiochemical yield of more than 99% in less than 1 min at
`room temperature. As a consequence, specific activities in the
`range 500−1000 GBq/μmol were obtained. In order to achieve
`comparable high radiochemical yields with the DOTA-
`
`conjugates 8 and 12, the compounds were incubated for 2
`min at 80 °C using 1 nmol precursor.
`Incubation of the 68Ga-labeled compounds 7, 8, and 12 in
`human serum for 2 h at 37 °C resulted in no detectable changes
`in the radiograms. Stability of the 68Ga-labeled HBED-CC-
`
`692
`
`dx.doi.org/10.1021/bc200279b | Bioconjugate Chem. 2012, 23, 688−697
`
`Petitioner GE Healthcare – Ex. 1030, p. 692
`
`

`

`Bioconjugate Chemistry
`
`conjugate 7 could additionally be demonstrated by transferrin
`challenging and incubation in human serum. No activity was
`exchanged in presence of 250 μg transferrin or transferred to
`other serum proteins.
`Determination of PSMA Affinity. The variants were
`competitively analyzed in terms of their binding capacity by
`performing an enzyme-based assay on rhPSMA (Naaladase-
`Assay) and a binding assay on LNCaP cells with 67Ga-labeled 7
`(RCY > 99%; specific activity ∼1000 GBq/μmol) as radio-
`ligand. The affinity related IC50 and the calculated Ki values of
`both assays were determined by analyzing the respective
`binding curves (Figure 1) with GraphPad Prism. The data are
`summarized in Table 1. Compared to the DOTA conjugate
`[natGa]8, the HBED-CC conjugate [natGa]7 and the more
`lipophilic reference compound [natGa]12 showed slightly better
`affinities to both the purified extracellular receptor rhPSMA
`and LNCaP cells.
`Specific Cell Binding and Internalization. Comparative
`cell uptake and internalization experiments with LNCaP
`(PSMA positive) and PC-3 (PSMA negative) cells revealed
`unspecific cell uptake of [68Ga]8 and [68Ga]12 in both cell lines
`which could not be blocked with excess PMPA (Figure 2A). In
`contrast, the HBED-CC conjugate ([68Ga]7) showed almost
`no uptake in PSMA-negative PC-3 cells or in LNCaP cells
`blocked with PMPA. Furthermore, [68Ga]7 significantly
`revealed higher
`specific internalization and cell
`surface
`accumulation as compared to the DOTA conjugates (Figure
`2B).
`In another cell uptake experiment,
`three different
`concentrations of the radiolabeled compounds (2.5 nM, 25
`nM, and 250 nM) were incubated with either LNCaP (Figure
`3A) or PC-3 cells (Figure 3B). In accordance with the results
`received from the internalization experiments, [68Ga]7 showed
`a considerably higher uptake in LNCaP cells. Furthermore, the
`unspecific uptake in PC-3 cells was lower as compared to the
`DOTA derivatives [68Ga]8 and [68Ga]12.
`Biodistribution Studies. The organ and tumor uptake
`the complexes [68Ga]7, [68Ga]8, and [68Ga]12
`values of
`differed in accordance with the aforementioned in vitro data
`(Figure 4). The HBED-CC conjugated compound [68Ga]7 was
`cleared rapidly from the circulation and PSMA negative tissue.
`Remarkably, the liver activity of [68Ga]7 amounted to only 0.87
`± 0.05% ID/g as early as one hour after injection. The high
`kidney, spleen, and lung uptake of 139.4 ± 21.4% ID/g, 17.90
`± 2.87% ID/g, and 2.49 ± 0.27% ID/g was nearly completely
`blocked to 4.02 ± 1.14% ID/g, 1.54 ± 0.33% ID/g, and 0.64 ±
`0.32% ID/g, respectively, after the coinjection of 2 mg/kg 2-
`PMPA (Figure 4B). The tumor uptake amounted to 7.70 ±
`1.45% ID/g on LNCaP and 1.30 ± 0.12% ID/g on PC-3.
`The DOTA complex [68Ga]8 showed a completely different
`uptake pattern. The liver uptake of the DOTA-conjugate was
`enhanced by a factor of 5.7, the tumor uptake was reduced by
`factor 2.6, and the kidney uptake was surprisingly lowered to
`background values. The uptake values of PSMA-negative PC-3
`tumors were low for all compounds (1.30 ± 0.13% ID/g
`([68Ga]7), 0.60 ± 0.06% ID/g ([68Ga]8), and 0.58 ± 0.07%
`ID/g ([68Ga]12), respectively) (Figure 4C).
`PET-Imaging. The tumor-targeting efficiency of [68Ga]7
`and [68Ga]8 was evaluated by 50 min dynamic microPET scans
`followed by a 20 min static scan. The differences observed in
`biodistribution could have also been visualized by PET imaging.
`The complexes [68Ga]7 and [68Ga]8 were injected intra-
`venously into LNCaP tumor-bearing mice, and microPET scans
`were carried out for up to 70 min p.i. A coronal PET slice after
`
`Article
`
`Figure 2. Cell binding and internalization of compounds [68Ga]7,
`[68Ga]8, and [68Ga]12 (A). Specific cell uptake was evaluated by
`blockade with 100 μM PMPA. Specificity of cell uptake (B) was
`calculated by subtracting the respective signals resulting from PMPA-
`blockage. Values are expressed as % of applied radioactivity bound to
`106 cells. Data are expressed as mean ± SD (n = 3).
`
`injection of [68Ga]7 is shown in Figure 5a. The DOTA complex
`[68Ga]8 failed to image the tumor proven by the time activity
`curves over tumor and muscle (Figure 5B).
`Comparison of Biological Properties of L- and D-
`Forms of [68Ga]7. The 68Ga complex of D-Glu-NH-CO-NH-
`Lys(Ahx)-HBED-CC was synthesized and compared with the
`corresponding L-isomer. Time−activity curves obtained from
`dynamic PET measurements revealed D-[68Ga]7 to be cleared
`from the kidneys into the urinary bladder while L-[68Ga]7 is
`retained in this organ (Figure 6). This finding is supported by
`the ∼103 times lower PSMA affinity of D-[68Ga]7 (data not
`shown).
`
`■ DISCUSSION
`PSMA is strongly expressed on prostate cancer cells18 and
`therefore represents a promising target for the development of
`imaging agents for tumor staging and followup. Among the
`currently used radioisotopes, 68Ga exhibits several
`favorable
`properties as there are the availability and the facile labeling
`technology by means of complexation.11 A series of chelators
`are available for the complexation of 68Ga which are conjugated
`to targeting carrier molecules. Besides some reported stabilizing
`
`693
`
`dx.doi.org/10.1021/bc200279b | Bioconjugate Chem. 2012, 23, 688−697
`
`Petitioner GE Healthcare – Ex. 1030, p. 693
`
`

`

`Bioconjugate Chemistry
`
`Article
`
`Figure 3. Binding of [68Ga]7, [68Ga]8, and [68Ga]12 to LNCaP (PSMA+) cells (A) and PC-3 (PSMA−) cells (B) depending on 68Ga-complex
`concentration. Values were normalized with respect to added radioactivity of each compound. Data are expressed as mean ± SD (n = 3).
`
`Figure 4. Organ distribution expressed as % ID/g tissue one hour post injection. (A) Comparison of the compounds [68Ga]7 (PSMA-binding motif
`coupled to HBED-CC), [68Ga]8 (PSMA-binding motif coupled to DOTA), and [68Ga]12. (B) PSMA-blocking by coadministration of 2 mg/kg
`body weight PMPA indicating PSMA-specific uptake in kidney and spleen. (C) Uptake in PC-3 tumors and LNCaP tumors 1 h after injection. Data
`are expressed as mean ± SD (n = 3).
`
`Figure 5. Whole-body coronal microPET image of an athymic male nude mice bearing LNCaP tumor xenografts. The tumor-targeting efficacy of
`[68Ga]7 was demonstrated by dynamic microPET scan (B) followed by a static scan (A). The static scan up to 1 h post injection of [68Ga]7 is shown
`in (A). Approximately 15 MBq/mouse was injected. Graph B shows the respective time−activity curves in muscle and tumor for [68Ga]7 and [68Ga]
`8.
`
`effects or pretargeting approaches,17,19 the function of the
`chelator in radiopharmaceuticals is usually restricted to serving
`as a complexing agent without influencing the receptor binding.
`However, the size-demanding radiometal complexes often exert
`their
`influence on the targeting molecule by changing
`lipophilicity or charge.
`In particular,
`the pharmacological
`property of small molecules can strongly be influenced or
`abolished with respect to its target binding properties.
`An interesting study comparing various linkers located
`between the PSMA binding group 2-[3-(1,3-dicarboxyprop-
`yl)-ureido]pentanedioic acid (DUPA) and a 99mTc complex
`showed that
`the lipophilicity of
`the linker
`significantly
`correlates with improved binding properties.14 Crystal structure
`investigations also indicated that, besides the electrostatic
`
`interactions of urea and carboxylic groups at the active, Zn(II)-
`containing center of PSMA, there are lipophilic interactions
`resulting from a hydrophobic pocket located next to the active
`site.8,20 A further study substantiates the PSMA active site as a
`pocket with multiple interactions as well. The pharmacophore
`was proposed to present three carboxylic groups able to interact
`with the respective side chains of PSMA, an oxygen as part of
`zinc complexation in the active center and an aromatic structure
`able to interact with a hydrophobic part of the binding pocket
`tyrosines.15 This work suggests that
`these
`composed of
`interactions are additive in terms of
`functional efficiency.
`Only inhibitors which interact with all parts of the binding site
`displayed a tight binding mode and a high rate of internal-
`
`694
`
`dx.doi.org/10.1021/bc200279b | Bioconjugate Chem. 2012, 23, 688−697
`
`Petitioner GE Healthcare – Ex. 1030, p. 694
`
`

`

`Bioconjugate Chemistry
`
`Article
`
`Figure 6. Time−activity curves taken from dynamic PET measurements and expressed as Bq/mL organ normalized to injected dose. The D-form of
`[68Ga]7 (A) shows ∼103 reduced binding capacity to PSMA as compared to the L-form (B). As a consequence, the L-form accumulates in kidneys
`whereas the D-isomer is rapidly cleared from kidneys into the bladder.
`
`ization. Molecules lacking one of these interactions showed
`other modes of action and weaker binding.15
`These results led to the development of the amphiphilic,
`PSMA-specific tracer [68Ga]7 consisting of
`the urea-based
`pharmacophore and the 68Ga-HBED-CC complex able to
`interact with the hydrophobic binding pocket. Two 68Ga-
`DOTA complexes, the hydrophilic [68Ga]8 and the amphiphilic
`[68Ga]12, were compared with [68Ga]7 in terms of PSMA-
`binding characteristics and internalization.
`Compared to the DOTA complexes [68Ga]8 and [68Ga]12,
`the HBED-CC complex [68Ga]7 showed a significantly
`enhanced PSMA-specific internalization in LNCaP cells (Figure
`2B). In another cell uptake experiment, where three different
`concentrations of the radiolabeled compounds were given to
`either LNCaP (Figure 3A) or PC-3 cells (Figure 3B), [68Ga]7
`showed a considerably higher uptake in LNCaP cells, while the
`unspecific uptake in PC-3 cells was significantly lower as
`compared to the DOTA derivatives [68Ga]8 and [68Ga]12. The
`binding data are characteristic for specific and unspecific
`binding because the values rise in an exponential manner on
`PSMA+ LNCaP cells (Figure 3A) and linearly on PSMA− PC-3
`cells (Figure 3B).
`Despite clear differences regarding unspecific cell uptake, the
`obtained binding affinity data of all com