In Vitro and in Vivo Evaluation of the Bifunctional Chelator NODIA-Me in Combination with a Prostate-Specific Membrane Antigen Targeting Vector

Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch

Year: 2018

In vitro and in vivo evaluation of the bifunctional chelator NODIA-Me in combination with a prostate-specific membrane antigen targeting vector

Läppchen, Tilman ; Kiefer, Yvonne ; Holland, Jason P ; Bartholomä, Mark D

Abstract: Introduction We recently developed a chelating platform based on the macrocycle 1,4,7- triazacyclononane with up to three five-membered azaheterocyclic arms for complexation of the PET nuclides gallium-68 and copper-64. The main objective of this study was to evaluate the stability and pharmacokinetics of 68Ga- and 64Cu-complexes of the bifunctional chelator NODIA-Me 1 covalently bound to a PSMA targeting vector in vivo. Methods NODIA-Me 1 was conjugated to the PSMA targeting Glu-NH-CO-NH-Lys moiety to give the bioconjugate NODIA-Me-NaI-Ahx-PSMA 4. The stability of [68Ga]4 and [64Cu]4 was assessed in vitro by serum stability studies. The PSMA binding aﬀinity was determined in competitive cell experiments in LNCaP cells using 68Ga-PSMA-HBED-CC as radioligand. The stability and pharmacokinetics of [68Ga]4 and [64Cu]4 was evaluated by PET imaging and ex vivo biodistribution studies in mice bearing subcutaneous LNCaP tumors. Results In human serum, [68Ga]4 and [64Cu]4 remained intact to 85% (3 h) and 92% (24 h), respectively. Nature of the metal chelate influenced PSMA binding aﬀinity with IC50 of 233 ± 10 nM for uncomplexed 4, 681 ±7nMforCu-4 and 176 ± 10 nM for Ga-4. In animal studies, [68Ga]4 and [64Cu]4 revealed low uptake (฀1% IA g−1) in the majority of organs. Kidney uptake at 1 h p.i. was 6.28 ± 0.92% IA g−1 and 4.96 ± 0.79% IA g−1 and specific tumor uptake was 1.33 ± 0.46% IA g−1 and 2.15 ± 0.38% IA g−1 for [68Ga]4and [64Cu]4, respectively. Conclusion The bifunctional chelator NODIA-Me 1 was successfully conjugated to a PSMA targeting moiety. In small-animal PET imaging and ex vivo biodistribution studies, 68Ga- and 64Cu-labelled conjugates specifically delineated PSMA-positive LNCaP tumors and exhibited rapid renal clearance from non-target tissues with no significant demetallation/transchelation in vivo. The results support further development of this novel chelating platform for production of 68Ga- and 64Cu-labelled radiopharmaceuticals.

DOI: https://doi.org/10.1016/j.nucmedbio.2018.03.002

Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-167583 Journal Article Accepted Version

The following work is licensed under a Creative Commons: Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) License.

Originally published at: Läppchen, Tilman; Kiefer, Yvonne; Holland, Jason P; Bartholomä, Mark D (2018). In vitro and in vivo evaluation of the bifunctional chelator NODIA-Me in combination with a prostate-specific membrane antigen targeting vector. Nuclear medicine and biology, 60:45-54. DOI: https://doi.org/10.1016/j.nucmedbio.2018.03.002

2 1 2 3 4 In vitro and in vivo evaluation of the bifunctional chelator 5 6 7 NODIA-Me in combination with a prostate-specific membrane 8 9 10 antigen targeting vector 11 12 13 14 15 Tilman Läppchen1,2, Yvonne Kiefer1, Jason P. Holland1,3, and Mark D. Bartholomä1,* 16 17 18 19 1 Department of Nuclear Medicine, Medical Center – University of Freiburg, Faculty of 20 21 Medicine, University of Freiburg, Hugstetterstrasse 55, D-79106, Freiburg, Germany 22 23 2 Department of Nuclear Medicine, Inselspital, Bern University Hospital and University of 24 25 26 Bern, Freiburgstrasse, CH-3010 Bern, Switzerland 27 3 28 Department of Chemistry, University of Zürich, Winterthurerstrasse 190, CH-8057, Zürich, 29 30 Switzerland 31 32 33 34 35 36 37 38 * Corresponding Author: 39 40 Dr. Mark D. Bartholomä 41 42 43 Tel: +49-(761)-270-39600 44 45 Fax: +49-(761)-270-39300 46 47 E-mail: [email protected] 48 49 50 51 52 53 54 55 56 57 58 Page 1 59 60 61 62 63 Key words: Gallium-68, Copper-64, prostate-specific membrane antigen (PSMA), PET 64 65 imaging, bifunctional chelator. 66 67 68 69 70 71 Running Title: NODIA-Me-PSMA 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 Page 2 118 119 120 121 122 Abstract 123 124 Introduction: We recently developed a chelating platform based on the macrocycle 1,4,7- 125 126 triazacyclononane with up to three five-membered azaheterocyclic arms for complexation of 127 128 the PET nuclides gallium-68 and copper-64. The main objective of this study was to evaluate 129 130 the stability and pharmacokinetics of 68Ga- and 64Cu-complexes of the bifunctional chelator 131 132 NODIA-Me 1 covalently bound to a PSMA targeting vector in vivo. 133 134 Methods: NODIA-Me 1 was conjugated to the PSMA targeting Glu-NH-CO-NH-Lys moiety 135 136 to give the bioconjugate NODIA-Me-NaI-Ahx-PSMA 4. The stability of [68Ga]4 and [64Cu]4 137 138 139 was assessed in vitro by serum stability studies. The PSMA binding affinity was determined 140 68 141 in competitive cell experiments in LNCaP cells using Ga-PSMA-HBED-CC as radioligand. 142 143 The stability and pharmacokinetics of [68Ga]4 and [64Cu]4 was evaluated by PET imaging 144 145 and ex vivo biodistribution studies in mice bearing subcutaneous LNCaP tumors. 146 147 Results: In human serum, [68Ga]4 and [64Cu]4 remained intact to 85% (3 h) and 92% (24 h), 148 149 respectively. Nature of the metal chelate influenced PSMA binding affinity with IC50 of 233 150 151 ± 10 nM for uncomplexed 4, 681 ± 7 nM for Cu-4 and 176 ± 10 nM for Ga-4. In animal 152 153 studies, [68Ga]4 and [64Cu]4 revealed low uptake IA g-1) in the majority of organs. 154 (≤1% 155 -1 -1 156 Kidney uptake at 1 h p.i. was 6.28 ± 0.92% IA g and 4.96 ± 0.79% IA g and specific tumor 157 -1 -1 68 64 158 uptake was 1.33 ± 0.46% IA g and 2.15 ± 0.38% IA g for [ Ga]4 and [ Cu]4, 159 160 respectively. 161 162 Conclusion: The bifunctional chelator NODIA-Me 1 was successfully conjugated to a PSMA 163 164 targeting moiety. In small-animal PET imaging and ex vivo biodistribution studies, 68Ga- and 165 166 64Cu-labelled conjugates specifically delineated PSMA-positive LNCaP tumors and exhibited 167 168 rapid renal clearance from non-target tissues with no significant demetallation/transchelation 169 170 in vivo. The results support further development of this novel chelating platform for 171 172 68 64 173 production of Ga- and Cu-labelled radiopharmaceuticals. 174 175 176 Page 3 177 178 179 180 181 1. Introduction 182 183 Obtaining maximum information from positron emission tomography (PET) imaging relies 184 185 on the development of target-specific radiotracers. Besides compounds labelled with the PET 186 187 nuclides carbon-11 and fluorine-18, radiopharmaceuticals labelled with positron-emitting 188 189 radiometals are an important part of ongoing research. In particular, the two radiometals 190 191 gallium-68 and copper-64 have received increased attention in the development of new PET 192 193 radiopharmaceuticals. One general advantage over carbon-11 and fluorine-18 is that labelling 194 195 reactions using radiometals can typically be carried out in aqueous media and are therefore 196 197 198 amenable to kit formulations. The interest in gallium-68 originates from its physical 199 + + 200 properties (t1/2 = 67.71 min, ß 89%, E(ß )max = 1.9 MeV), which are suitable for use with 201 202 targeting vectors possessing rapid pharmacokinetics such as radiotracers based on small 203 204 molecules and peptides [1-4]. In contrast to carbon-11 and fluorine-18, the radiometal 205 206 gallium-68 is generator produced and readily accessible without the need for an on-site 207 208 cyclotron. More recently, a GMP-compliant pharmaceutical grade generator produced by 209 210 Eckert & Ziegler has received marketing authorization. Other clinical grade gallium-68 211 212 generators are in development, which will spur the continued use of 68Ga-based 213 214 215 radiopharmaceuticals in the near future [5]. 216 217 The cyclotron-produced radiometal copper-64 has also shown promise as both a 218 219 suitable PET imaging and a therapeutic radionuclide due to its favorable decay characteristics 220 + + - - 221 (t1/2 = 12.7 h, ß 17.4%, E(ß )max = 0.656 MeV; ß 39%, E(ß )max = 0.573 MeV) and 222 223 commercial availability in large quantities with high specific activity [6-8]. The intermediate 224 225 half-life of copper-64 allows visualization of processes at time points up to 48 h, which is 226 227 ideal for developing radiopharmaceuticals that require longer circulation times to achieve 228 229 optimal uptake. The half-life of copper-64 also allows central production and shipment. 230 231 64 232 Furthermore, Cu-labelled radiopharmaceuticals may be used to provide pharmacokinetic 233 234 235 Page 4 236 237 238 239 - 240 profiles of therapeutic radiotracers that are labelled with copper-67 (t1/2 = 61.5 h, ß 100%, 241 242 Emax = 0.121 MeV) as this isotope has promise in endoradiotherapy applications [9]. 243 244 Radiolabelling of target-specific radiopharmaceuticals with a radiometal generally 245 246 requires the use of a bifunctional chelator (BFC) that (1) forms kinetically stable complexes 247 248 with the radiometal and (2) possesses a function for the covalent attachment of a target- 249 250 specific molecule. High kinetic and metabolic stability of the metal chelate is essential to 251 252 avoid accumulation of radioactivity in background tissues that may result from degradation or 253 254 transmetallation in vivo. However, introduction of a metal chelate into a biomolecule is a 255 256 257 non-innocent modification. Chelates/metal complexes often have an impact on the biological 258 259 properties of the radiopharmaceutical. The use of different radiometals with the same BFC 260 261 can alter binding affinities as illustrated by reported studies on various gallium and yttrium 262 263 labelled somatostatin analogs such as DOTA-[Tyr3]-octreotate and DOTA-octreotide [10]. 264 265 Variation of the metal chelate by using different BFCs with the same radiometal can also 266 267 result in altered pharmacokinetic profiles [11-12]. Interestingly, the influence of the metal 268 269 270 chelate is not limited to radiotracers based on small molecules or peptides. For example, the 271 64 272 biodistribution profiles of Cu-labelled antibody fragments as well as full antibody 273 274 conjugates can also be influenced by the nature of the metal chelate [13-14]. Evidently the 275 276 important role that chemical structure has on the biological performance of a radiotracer 277 278 demands continued development and optimization of new BFCs. The dependency of 279 280 radiotracer performance on the chemical structure also presents an opportunity for 281 282 radiochemists to tune the pharmacokinetics. Radiotracer design incorporates control over the 283 284 metal complex properties such as overall charge, lipophilicity and size as indispensable 285 286 287 handles for optimizing target-specific binding and distribution in vivo. 288 289 Numerous BFCs have been developed for both gallium-68 and copper-64 (Scheme 1), 290 291 such as NOTA (1,4,7-triazacyclononane-1,4,7-triacetic acid), NODAGA (1,4,7- 292 293 294 Page 5 295 296 297 298 299 triazacyclononane-1-glutaric acid-4,7-diacetic acid) [15-18] [1, 11, 19-21], TRAP (1,4,7- 300 301 triazacyclononane-1,4,7-tris[methyl(2-carboxyethyl)phosphinic acid) [22-24], PCTA 302 303 (3,6,9,15-tetraazabicyclo[9.3.1]pentadeca-1(15),11,13-triene-3,6,9-triacetic acid) [25-27], and 304 305 rigid cage-like sarcophagine type ligands such as diamsar [3, 28-32]. The most prominent 306 307 BFC for gallium-68 is the ligand DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic 308 309 acid [6], while copper-specific ligands include CB-TE2A (2,2'-(1,4,8,11- 310 311 tetraazabicyclo[6.6.2]hexadecane-4,11-diyl)diacetic acid) [28, 33-38], and bispidines [39-41]. 312 313 For a more comprehensive overview, the reader is referred to recent reviews [6, 42-45]. 314 315 68 316 The clinical success of Ga-labelled PSMA-HBED-CC (also known as DKFZ- 317 318 PSMA-11) represents an archetypal example of a system where the nature of the metal 319 320 chelate forms an integral part of the radiotracer’s binding mode. The aromatic rings of the 321 322 [68Ga]Ga-HBED-CC chelate form a key element for docking hydrophobic patches near the 323 324 binding pocket of the target protein prostate-specific membrane antigen (PSMA) [46]. PSMA 325 326 is an attractive molecular target for the diagnosis due to its upregulation in primary and 327 328 metastatic prostate cancer [47-48]. The PSMA pocket is comprised of two sub-pockets with 329 330 one surface-exposed highly positively charged pharmacophore region containing two Zn2+ 331 332 333 ions, and a second buried hydrophobic non-pharmacophore sub-pocket [49-50]. While the 334 335 urea motif of PSMA-HBED-CC binds to the pharmacophore region, the BFC HBED-CC 336 337 interacts advantageously with the lipophilic part of the PSMA pocket [46, 51]. In case of 338 339 traditional macrocyclic BFCs like DOTA, which lack aromatic rings, lipophilic substituents 340 341 were consequently introduced into the linker region in order to mimic the proven additional 342 343 biologic interactions of the chelator HBED-CC [52-56]. 344 345 We recently described a chelating platform based on the macrocycle 1,4,7- 346 347 triazacyclononane (TACN) with additional five-membered azaheterocyclic arms including 348 349 350 imidazole, methylimidazole and thiazole residues [57]. These ligands are characterized by 351 352 353 Page 6 354 355 356 357 358 their excellent complexation properties for radioactive, divalent copper cations. They label 359 64 360 instantaneously with [ Cu]CuCl2 under very mild conditions in the pH range of 4 – 8 with 361 362 high molar activities of 120 – 180 MBq nmol-1. Experiments showed that corresponding 363 364 methylimidazole-type ligands (e.g. NODIA-Me in Scheme 1) that incorporate borderline 365 366 azaheterocyclic N donors according to the HSAB (hard and soft acids and bases) principle 367 368 form rigid and stable Ga3+ complexes allowing their application also with the PET-nuclide 369 370 gallium-68 [58]. This study reports the in vitro and in vivo evaluation of the BFC NODIA-Me 371 372 (2-(4,7-bis((1-methyl-1H-imidazol-2-yl)methyl)-1,4,7-triazonan-1-yl)acetic acid) 1 labelled 373 374 375 with gallium-68 and copper-64 in combination with the glutamate-urea-lysine moiety for 376 377 targeting PSMA. The aim of this study was to demonstrate that NODIA-Me is suitable for 378 379 future radiopharmaceutical development. 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 Page 7 413 414 415 416 417 2. Materials and Methods 418 419 2.1. General 420 421 Chemicals and solvents were purchased from Sigma-Aldrich and TCI Europe, and used as 422 423 received. The precursor Glu-NH-CO-NH-Lys(Ahx)-(tBu)3 ester was purchased from ABX 424 425 (Radeberg, Germany). Fmoc-3-(2-naphthyl)-L-alanine was obtained from IRIS Biotech 426 427 (Marktredwitz, Germany). The bifunctional chelator NODIA-Me (2-(4,7-bis((1-methyl-1H- 428 429 imidazol-2-yl)methyl)-1,4,7-triazonan-1-yl)acetic acid) 1 was prepared as previously 430 431 described [58]. The radiotracer [68Ga]Ga-PSMA-HBED-CC was prepared as previously 432 433 434 described [59]. Low resolution electrospray ionisation mass spectrometry (LR-ESI(+)-MS) 435 436 was performed on a PerkinElmer Flexar SQ 300 MS Detector. High resolution mass 437 438 spectrometry (HR-ESI(+)-MS) was performed on a Thermo Scientific Exactive mass 439 1 13 440 spectrometer. H and C NMR spectra were measured in D2O at room temperature using a 441 442 Bruker Avance III HD 500MHz NMR spectrometer (Rheinstetten, Germany). Chemical 443 444 shifts are given in parts per million (ppm) and are reported relative to 4,4-dimethyl-4- 445 446 silapentane-1-sulfonic acid (DSS). Coupling constants are reported in hertz (Hz). The 447 448 multiplicity of the NMR signals is described as follows: s = singlet, d = doublet, t = triplet, q 449 450 68 451 = quartet, m = multiplet. Radiolabelling with [ Ga]GaCl3 was accomplished using a fully 452 453 automated synthesis module (Pharmtracer, Eckert & Ziegler, Berlin, Germany) with an 454 64 455 IGG100 generator (Eckert & Ziegler, Berlin, Germany). [ Cu]CuCl2 was purchased from the 456 457 Department of Preclinical Imaging and Radiopharmacy of the Eberhard-Karls-University 458 64 64 64 459 (Tübingen, Germany). [ Cu]CuCl2 was produced via the Ni(p,n) Cu route. The 460 461 corresponding molar activity was 300–400 MBq nmol-1 (18 h EOB). High performance liquid 462 463 chromatography (HPLC) was conducted on an Agilent 1260 Infinity System equipped with 464 465 an Agilent 1200 DAD UV detector (UV detection at 220 nm) and a gamma-radiation detector 466 467 468 (Ramona, Raytest GmbH, Straubenhardt, Germany) in series. A Phenomenex Jupiter Proteo 469 470 471 Page 8 472 473 474 475 476 (250 × 4.60 mm) column was used for analytical HPLC. The solvent system was A = H2O 477 478 (0.1% TFA) and B = acetonitrile (0.1% TFA). The gradient was 0–1 min 5% B, 1–20 min 5– 479 480 40% B at a flow rate of 1 mL min-1. Semi-preparative HPLC was performed on a Knauer 481 482 Smartline 1000 HPLC system in combination with a Macherey Nagel VP 250/21 Nucleosil 483 484 120-5 C18 column. Conditions A were 0–40 min 5–60% B at a flow rate of 12 mL min-1. 485 486 Conditions B were 0–25 min 5–95% B at a flow rate of 12 mL min-1. Samples were 487 488 lyophilized using a Christ Alpha 1-2 LD plus lyophilizer. All instruments measuring 489 490 radioactivity were calibrated and maintained in accordance with previously reported routine 491 492 493 quality-control procedures [60]. Radioactivity was measured using an Activimeter ISOMED 494 495 2010 (Nuklear-Medizintechnik, Dresden, Germany). For accurate quantification of 496 497 radioactivity, experimental samples were counted for 1 min on a calibrated PerkinElmer 498 499 (Waltham, MA) 2480 Automatic Wizard Gamma Counter by using a dynamic energy 500 501 window of 400–600 keV for gallium-68 and copper-64 (511 keV emission). A Thermo Fisher 502 503 Heraeus Fresco 21 centrifuge was used for centrifugation. 504 505 506 507 2.2. Bioconjugate Synthesis 508 509 2 510 2.2.1. Glu-NH-CO-NH-Lys(Ahx-NaI)-(tBu)3 ester ( ) 511 512 Fmoc-L-2NaI-OH (Fmoc-3-(2-naphthyl)-L-alanine) (16.6 mg, 0.038 mmol), HATU (1- 513 514 [bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium-3-oxide 515 516 hexafluorophosphate) (14.5 mg, 0.038 mmol) and DIPEA (N,N-diisopropylethylamine) (9.8 517 518 mg, 14 µL, 0.076 mmol) were mixed in 500 µL anhydrous N,N-dimethylformamide (DMF) 519 520 and allowed to stir for 1 h at r.t.. Then, Glu-NH-CO-NH-Lys(Ahx)-(tBu)3 (20.7 mg, 0.034 521 522 mmol) in 100 µL DMF was added and stirring was continued for additional 16 h. The solvent 523 524 was removed by rotary evaporation and the residue treated with 20% piperidine/DMF for 30 525 526 527 min at r.t.. After removal of the solvent, the residue was taken up in water/acetonitrile (50:50 528 529 530 Page 9 531 532 533 534 535 v/v) with 0.1 % trifluoroacetic acid (TFA) and purified by semi-preparative HPLC. Fractions 536 537 containing the product were combined and lyophilized to give compound 2 as white powder 538 539 (26.3 mg, 0.033 mmol, 97%). LR-ESI-MS calcd m/z for C43H68N5O9 (M + H): 798.5, found: 540 541 799.1. HR-ESI-MS calcd m/z for C43H68N5O9 (M + H): 798.5012, found: 798.5007. RP- 542

543 HPLC (semi-preparative, conditions A): tR = 44 min. 544 545 546 547 2.2.2. NODIA-Me-NaI-Ahx-PSMA-(tBu) ester (3) 548 3 549 NODIA-Me (13.6 mg, 0.036 mmol), HATU (15.2 mg, 0.040 mmol) and DIPEA (10.3 mg, 14 550 551 552 µL, 0.080 mmol) were mixed in 500 µL anhydrous DMF and allowed to stir for 30 min at r.t. 553 554 Then, compound 2 (26.3 mg, 0.033 mmol) in 100 µL DMF was added and stirring was 555 556 continued for additional 16 h. The solvent was removed by rotary evaporation and the residue 557 558 was taken up in water/acetonitrile (50:50 v/v) with 0.1 % TFA and purified by semi- 559 560 preparative HPLC. Fractions containing the product were combined and the solvent removed 561 562 by rotary evaporation to give compound 3 as colorless solid (26.5 mg, 0.023 mmol, 70%). 563 564 LR-ESI-MS calcd m/z for C61H95N12O10 (M + H): 1155.7, found: 1156.3. HR-ESI-MS calcd 565 566 m/z for C H N O (M + H): 1155.7289, found: 1155.7301. RP-HPLC (semi-preparative, 567 61 95 12 10 568 569 conditions B): tR = 25 min. 570 571 572 573 2.2.3. NODIA-Me-NaI-Ahx-PSMA (4) 574 575 Compound 3 (26.5 mg, 0.023 mmol) was dissolved in 1 mL 5 M hydrochloric acid and 576 577 allowed to stir for 16 h at room temperature. The reaction mixture was concentrated by rotary 578 579 evaporation to a volume of 1 mL and further purified by semi-preparative HPLC to give 580 581 NODIA-Me-NaI-Ahx-PSMA ((4S,18S,22S)-1-(4,7-bis((1-methyl-1H-imidazol-2-yl)methyl)- 582 583 1,4,7-triazonan-1-yl)-4-(naphthalen-2-ylmethyl)-2,5,12,20-tetraoxo-3,6,13,19,21- 584 585 586 pentaazatetracosane-18,22,24-tricarboxylic acid) 4 as colorless powder after lyophilization 587 588 589 Page 10 590 591 592 593 1 594 (20.6 mg, 0.021 mmol, 91%). H-NMR (D2O, pD ≤ 2): δ = 7.79 – 7.73 (4H, m), 7.48 – 7.45 595 596 (4H, m), 7.42 (1H, d, J = 1.7), 7.33 (2H, s), 4.89 (1H, dd, J = 10.7, 4.7), 4.21 (3H, m), 4.06 – 597 598 3.95 (3H, m), 3.69 (3H, s), 3.67 (3H, s), 3.47 (2H, m), 3.36 – 3.28 (4H, m), 3.12 – 2.94 (7H, 599 600 m), 2.80 (1H, m), 2.67 (1H, d, J = 16), 2.53 (1H, m), 2.54 (4H, m), 2.43 (1H, m), 2.12 (3H, 601 602 m), 2.01 (1H, m), 1.92 (1H, m), 1.67 (1H, m), 1.55 (1H, m), 1.49 – 1.38 (6H, m), 1.26 (2H, 603 604 m), 1.15 (2H, m). 13C-NMR (D2O, pD 2): δ = 177.2, 176,7, 176.2, 171.5, 165.0, 163.0 (q, 605 ≤ 606 607 TFA), 159.2, 142.6, 134.8, 133.0, 132.0, 128.2, 128.1, 127.9, 127.4, 127.2, 126.8, 126.3, 608 609 124.2, 124.1, 118.8, 118.6, 116.3 (q, TFA), 56.4, 55.9, 53.3, 53.2, 52.6, 51.6, 50.0, 48.4, 47.8, 610 611 47.0, 45.3, 39.3, 39.0, 35.6, 34.4, 30.5, 30.0, 28.0, 27.8, 26.3, 25.4, 24.9, 22.3. LR-ESI-MS 612 613 calcd m/z for C49H71N12O10 (M + H): 987.5, found: 987.7. HR-ESI-MS calcd m/z for 614 615 C49H71N12O10 (M + H): 987.5411, found: 987.5399. RP-HPLC (semi-preparative, conditions 616 617 B): tR = 14 min; RP-HPLC (analytical, UV: 220 nm): tR = 15:20 min. 618 619 620 621 2.3. Preparation of non-radioactive references 622 623 624 For each reference compound, the bioconjugate 4 (500 µg, 0.51 µmol) in 250 µL H2O was 625 626 mixed with 250 µL of metal stock solutions containing 1.5 equivalents of the corresponding 627 628 metal salt (CuCl2⋅2 H2O, Ga(NO3)3 hydrate) and heated at 95 °C for 15 min. After cooling to 629 630 r.t., the complexes were purified using C18 Sep Pak cartridges, which were preconditioned 631 632 with EtOH and H2O (5 mL each). After loading, the cartridge was washed with H2O (2 mL) 633 634 and the product was eluted off using EtOH:H O (50:50 v/v; 2mL). After evaporation of EtOH 635 2 636 at ambient temperature, samples were lyophilized to give Cu-4 (479 µg, 0.46 µmol, 90%) and 637 638 + 639 Ga-4 (461 µg, 0.44 µmol, 86%). HR-ESI-MS of Cu-4 calcd m/z for C49H69CuN12O10 (M – 640 641 H): 1048.4550, found: 1048.4551. RP-HPLC of Cu-4 (analytical, UV: 220 nm): tR = 17:42 642 + 643 min. HR-ESI-MS of Ga-4 calcd m/z for C49H68GaN12O10 (M – 2H): 1053.4432, found: 644 645 1053.4426. RP-HPLC of Ga-4 (analytical, UV: 220 nm): tR = 15:33 min. 646 647 648 Page 11 649 650 651 652 653 654 655 2.4. Radiolabelling 656 657 2.4.1. Radiosynthesis of [68Ga]Ga-NODIA-Me-NaI-Ahx-PSMA ([68Ga]4) 658 659 68Ga radiolabelling was accomplished by using the Modular-Lab PharmTracer automated 660 661 synthesis module (Eckert&Ziegler, Berlin, Germany) as previously described [58]. The 662 663 reaction vial contained 20 µg of 4 in a mixture of 400 µL sodium acetate buffer (pH 4.5) and 664 665 2 mL water. Radiolabelling was performed at 95°C for 10 min. The radiochemical purity 666 667 (RCP) was >98% and the decay corrected radiochemical yield (RCY) was >98%. The mean 668 669 -1 670 molar activity was Am = 20.2 ± 2.3 MBq nmol (n = 10). RP-HPLC (analytical, radioactivity 671 68 672 detector): tR ([ Ga]4) = 15:50 min. 673 674 675 676 2.4.2. Radiosynthesis of [64Cu]Cu-NODIA-Me-NaI-Ahx-PSMA ([64Cu]4) 677 64 678 Radiolabelling with [ Cu]CuCl2 was accomplished by dissolving 6 µg (6 nmol) of 4 in 200 679 680 64 µL ammonium acetate (0.1 M, pH 8.2) followed the addition of ~120 MBq [ Cu]CuCl2 in 681 682 0.1 M HCl. The solution was then incubated for 15 min at r.t. The RCP was >99% and the 683 684 decay corrected RCY was >99%. The mean molar activity was A = 20.5 ± 1.8 MBq nmol-1 685 m 686 64 687 (n = 5). RP-HPLC (analytical, radioactivity detector): tR ([ Cu]4) = 17:43 min. 688 689 690 691 692 2.5. Lipophilicity (log Doct/PBS) measurements 693 694 68 64 68 For log Doct/PBS measurements, 1–2 MBq of [ Ga]4, [ Cu]4 and [ Ga]Ga-PSMA-HBED-CC 695 696 (20 µL) were added to a pre-saturated mixture of phosphate buffered saline pH 7.4 (PBS) 697 698 699 (480 µL) and octanol (500 µL). Samples were shaken for 30 min at room temperature, 700 701 centrifuged at 21,100 g for 5 min and 100 µL of each phase were counted using a Packard 702 703 Cobra gamma counter. Experiments were performed in triplicate. 704 705 706 707 Page 12 708 709 710 711 712 2.6. Serum stability measurements 713 714 For each experiment, 5–10 MBq (50 µL) of [68Ga]4 and [64Cu]4 were added to human serum 715 716 (500 µL, human male AB plasma, Sigma-Aldrich), which was pre-equilibrated at 37 °C in a 717 718 cell incubator for 1 h. Samples were vortexed briefly and stored at 37 °C in a cell incubator. 719 720 721 At selected time points, 100 µL aliquots were taken from the serum solution and serum 722 723 proteins removed by centrifugation using centrifuge tubes (molecular weight cutoff 30 kDa) 724 725 at 4,000 g for 5 min at 4 °C. 100 µL PBS were added to the filter followed by an additional 726 727 centrifugation step. For determination of protein bound fraction, the filters and the filtrates 728 729 were transferred into plastic tubes, and samples were counted using a Packard Cobra gamma 730 731 counter. A sample of the filtrate was kept on ice until HPLC analysis. The percentage of 732 733 intact [68Ga]4 and [64Cu]4 was calculated from the HPLC chromatograms. All experiments 734 735 were performed in triplicate. 736 737 738 739 740 2.7. Cell culture 741 742 PSMA-positive LNCaP cells (ATCC, Manassas, VA, USA) were cultured at 37 °C in a 5% 743 -1 744 CO2 atmosphere (RPMI Medium 1640 GlutaMAX containing 10% FBS, 1% 100 U mL 745 746 penicillin and 100 μg mL-1 streptomycin, 1% sodium-pyruvate 1 mM, 1% L-glutamin 2 mM). 747 748 749 750 2.8. Competitive binding assay 751 752 753 The binding affinity of the bioconjugate 4 and the corresponding metallated compounds Ga-4 754 755 and Cu-4 was determined using a cell-based competitive binding assay with 68Ga-labelled 756 757 PSMA-HBED-CC (PSMA-DKFZ-11) according to the literature with minor modifications 758 759 [61]. Briefly, each compound at different concentrations (0 – 5000 nM) was incubated for 1 h 760 761 at 37 °C with 0.15 nM [68Ga]Ga-PSMA-HBED-CC together with 2×105 LNCaP cells per 762 763 well. After incubation, cells were washed three times with ice-cold PBS and cell-associated 764 765 766 Page 13 767 768 769 770 771 activity recovered by addition of 1 M NaOH. Radioactivity was measured by a gamma 772 773 counter and data fitted using non-linear regression (GraphPad Prism). Experiments were 774 775 performed two times in triplicate. 776 777 778 779 2.9. Small-animal PET imaging 780 781 782 All animal experiments complied with the current laws of the Federal Republic of Germany 783 784 and were conducted according to German Animal welfare guidelines. Normal female athymic 785 786 Balb/c nude mice (17–20 g, 4–6 weeks old) were obtained from Janvier SAS (St. Berthevin 787 788 Cedex, France). Mice were provided with food and water ad libitum. LNCaP tumors were 789 790 induced on the shoulder by sub-cutaneous injection of 5×106 cells in a 100 μL cell suspension 791 792 of a 1:1 v/v mixture of media with reconstituted basement membrane (GFR BD MatrigelTM, 793 794 795 Corning BV, Amsterdam, Holland). After an average of 4 weeks, tumor size reached ~200 – 796 797 300 mg and the animals were used for PET imaging studies. 798 799 For PET imaging studies, mice were injected with 100 µL sterile filtered phosphate 800 801 buffered saline formulations pH 7.4 of [64Cu]4 (6–8 MBq) or [68Ga]4 (18–23 MBq) by 802 803 intravenous tail-vein injection and anesthetized with isoflurane (2–4% in air). PET Imaging 804 805 was performed on a Focus 120 microPET scanner at 30 min, 1 h, 2 h and 3 h post-injection 806 807 (p.i.). For copper-64, additional scans at 24 h p.i. were performed. Data were acquired for 10, 808 809 15 or 60 min in list mode. Reconstruction was performed using unweighted OSEM2D. Image 810 811 812 analysis was performed using AMIDE. Image counts per second per voxel (cps/voxel) were 813 -1 814 calibrated to activity concentrations (Bq mL ) by measuring a 3.5 cm cylinder phantom filled 815 816 with a known concentration of radioactivity. Time-activity curves (TACs) were constructed 817 818 by manually drawing regions of interest (ROI) in the tumor, liver and the kidneys. All ROIs 819 820 were then copied on each of the frames, and time-activity curves of the ROI mean values 821 822 were generated. 823 824 825 Page 14 826 827 828 829 830 Competitive inhibition (blocking) studies were also performed in vivo and measured 831 832 using static PET imaging to investigate the specificity of the radiotracers for PSMA. As a 833 834 blocking agent, non-radiolabelled 2-(phosphonomethyl)pentane-1,5-dioic acid (PMPA; 20 835 836 nmol mouse-1) was co-injected with the radiotracers (n = 3) [62]. 837 838 839 840 2.10. Ex vivo biodistribution 841 842 843 For each compound, a total of five animals were injected with [64Cu]4 (6–8 MBq) or [68Ga]4 844 845 (18–23 MBq) in 100 µL sterile filtered phosphate buffered saline via a tail vein. At 60 min 846 847 p.i., animals were sacrificed by isoflurane anesthesia. Organs of interest were dissected, 848 849 weighed and assayed for radioactivity in a gamma counter. The percent injected activity per 850 851 gram (%IA g-1) for each tissue was calculated by comparison of the tissue counts to a 852 853 standard sample prepared from the injection solution. Specificity of the radiotracers for 854 855 -1 856 PSMA was determined by co-injection of PMPA (20 nmol mouse ) [62]. 857 858 859 860 3. Results and discussion 861 862 3.1. Bioconjugate synthesis 863 864 Our previously described triply substituted ligands lack a functionality for the covalent 865 866 conjugation of biologically active targeting vectors [57]. For this reason, we developed the 867 868 BFC NODIA-Me 1, in which we replaced one methylimidazole arm with an acetic acid group 869 870 (Scheme 2) [58]. This acetic acid group serves as site for the attachment of appropriate 871 872 873 targeting vectors via peptide bond formation. The glutamate-urea-lysine (Glu-NH-CO-NH- 874 875 Lys) structural motif is a well-known building block for the development of potent PSMA 876 877 inhibitors [63-64]. Besides interactions of the urea and the carboxylic groups with the Zn2+ 878 879 active site of the PSMA binding pocket, there are also lipophilic interactions with an 880 881 accessible hydrophobic pocket [63, 65]. Therefore, a lipophilic naphthyl-alanine spacer was 882 883 884 Page 15 885 886 887 888 889 introduced via standard Fmoc chemistry to the commercially available precursor Glu-NH- 890 891 CO-NH-Lys(Ahx)-tBu3 ester to obtain the intermediate 2 in 97% yield (Scheme 2). 892 893 Conjugation of NODIA-Me 1 was accomplished using HATU as coupling reagent to give 894 895 compound 3 in 70% yield. Cleavage of the tBu-esters was accomplished by reaction of 3 in 5 896 897 M hydrochloric acid for 16 h to give the final PSMA-bioconjugate 4 in 61% overall yield. 898 899 The identity and purity of compound 4 (>98%) was determined by NMR spectroscopy, mass 900 901 spectrometry and analytical RP-HPLC. Corresponding data are given in the Supporting 902 903 Information. 904 905 906 907 908 3.2. Radiochemistry 909 910 Radiosynthesis of [64Cu]4 was accomplished by mixing the bioconjugate 4 with the 911 64 912 appropriate radioactivity amount of [ Cu]CuCl2 in ammonium acetate (pH 8.2) followed by 913 914 incubation for 15 min at ambient temperatures. These mild conditions concur with the results 915 916 obtained in previous 64Cu-labelling experiments of our ligands bearing three azaheterocyclic 917 918 arms [57]. The RCP and the decay corrected RCY for [64Cu]4 were both determined to be 919 920 >98%. 921 922 68 923 Radiosynthesis of [ Ga]4 was performed in an automated, GMP-compliant process by 924 68 925 heating compound 4 with [ Ga]GaCl3(aq.) at 95 °C for 10 min [58]. The RCP and the decay 926 927 corrected RCY for [68Ga]4 were measured to >98% and for both radiometals molar activities 928 929 of ~20 MBq nmol-1 were obtained. 930 931 932 933 3.3. Lipophilicity 934 935 68 64 The log Doct/PBS values of [ Ga]4 and [ Cu]4 were measured in an octanol-phosphate 936 937 buffered saline system (Table 1). Compounds are highly hydrophilic with log D values 938 oct/PBS 939 68 64 940 of -4.27 ± 0.08 and -3.99 ± 0.05 for [ Ga]4 and [ Cu]4, respectively. The order of 941 942 943 Page 16 944 945 946 947 948 lipophilicity was corroborated chromatographically by RP-HPLC (Table 1). The lipophilicity 949 68 950 of both compounds is comparable to [ Ga]Ga-PSMA-HBED-CC with log Doct/PBS = -4.06 ± 951 952 0.10. 953 954 We recently showed that the remaining coordination site of the gallium NODIA-Me 955 956 complex can be occupied by monodentate ligands such as hydroxide or chloride [58]. The 957 958 NODIA-Me complex charge can thus be estimated to +2 for [68Ga]4 and [64Cu]4. The 959 960 deprotonated carboxylic acid functions of the Glu-NH-CO-NH-Lys PSMA-binding motif 961 962 provide a 3-fold negative charge resulting in an estimated overall charge of both [68Ga]4 and 963 964 64 68 965 [ Cu]4 of -1. In contrast, the PSMA-targeting reference ligand [ Ga]Ga-PSMA-HBED-CC 966 967 possesses an estimated overall charge of -5 (-1 for the hexadentate N2O4-complex to Ga(III) 968 969 with two carboxylates and two phenolates of the HBED-CC, -1 for the free terminal 970 971 carboxyethyl-group of HBED-CC, and -3 for deprotonated carboxylic acid functions of the 972 973 Glu-NH-CO-NH-Lys PSMA-binding motif) [66-68]. Despite the lower overall charge of 974 975 [68Ga]4 and [64Cu]4 at physiological pH, the total number of charges is comparable to 976 977 [68Ga]Ga-PSMA-HBED-CC. Furthermore, the positively charged NODIA-Me metal chelates 978 979 result in a zwitterionic nature of the [68Ga]4 and [64Cu]4 PSMA-targeting probes. 980 981 982 983 984 985 986 3.4. In vitro studies 987 988 3.4.1. Serum stability studies 989 990 The stability of [68Ga]4 and [64Cu]4 was assessed in vitro by incubation at 37°C in human 991 992 serum. For [68Ga]4, the amount of intact compound decreased from 94 ± 2% at 1 h to 85 ± 993 994 3% after 4 h incubation time. The major metabolite of [68Ga]4 eluted after the dead volume of 995 996 the system, consistent with the retention time of “free” gallium-68 and/or [68Ga]1. The 64Cu- 997 998 64 999 labelled conjugate [ Cu]4 was found to be more stable with 92 ± 2% intact tracer after 24 h 1000 1001 1002 Page 17 1003 1004 1005 1006 68 64 1007 incubation. The protein associated fractions of [ Ga]4 and [ Cu]4 in the serum stability 1008 1009 studies were determined to 48 ± 5% and 29 ± 2%, respectively. The protein bound fractions 1010 1011 of [68Ga]4 and [64Cu]4 compare well to that of [68Ga]Ga-PSMA-HBED-CC, which was 1012 1013 determined to 35 ± 3%. Finally, exchange of monodentate ligands such as chloride or 1014 1015 hydroxide coordinated to the metal center, as reported previously, was not observed for 1016 1017 [68Ga]4 [58]. 1018 1019 1020 1021 3.4.2. Competitive binding assay 1022 1023 1024 The binding affinities of 4, Ga-4 and Cu-4 were determined in a competitive binding assay on 1025 68 1026 LNCaP cells using [ Ga]Ga-PSMA-HBED-CC as radioligand, which yielded IC50-values of 1027 1028 233 ± 10 nM, 176 ± 10 nM and 681 ± 7 nM (Table 1). The IC50-values were converted to Ki- 1029 68 1030 values using the Cheng-Prusoff equation [69] and the reported Kd value for [ Ga]Ga-PSMA- 1031

1032 HBED-CC of 2.9 ± 0.6 nM [55]. With Ki values of 222 nM, 167 nM, and 648 nM, 1033 1034 respectively, the NODIA-Me based PSMA-targeting ligands 4, Ga-4, and Cu-4 revealed 1035 1036 substantially lower PSMA-binding affinities than [68Ga]Ga-PSMA-HBED-CC. Clearly, the 1037 1038 size and charge of the metal chelating unit as well differences in the linker region have a 1039 1040 1041 major impact on the PSMA-binding affinity of the whole construct. 1042 1043 Different from HBED-CC, the BFC NODIA-Me 1 does not possess any charge 1044 1045 compensating donor groups and, consequently, forms metal complexes with Cu2+ and Ga3+ 1046 1047 that have a positive overall charge. Under physiological conditions, the compounds [68Ga]4 1048 1049 and [64Cu]4 presumably possess an estimated overall charge of -1. Even if no metal cation is 1050 1051 coordinated to 4, the NODIA-Me chelating moiety is partially protonated at physiologic pH 1052 1053 as the pKa values for the TACN amino groups are 7.0 and 11.3 and the pKa value of 1054 1055 imidazole is 7.0 [70]. This partial protonation of the metal binding moiety of uncomplexed 4 1056 1057 1058 at physiological pH seems to translate to a PSMA binding affinity comparable to Ga-4. 1059 1060 1061 Page 18 1062 1063 1064 1065 1066 Even though other factors cannot be excluded such as differences in the competition 1067 1068 assay (different radioligand), the rather large charge difference between our NODIA-Me 1069 1070 based PSMA-targeting probes and [68Ga]Ga-PSMA-HBED-CC may be the primary reason 1071 99m 1072 for their lower affinity. Indeed, recent studies by Babich and coworkers on Tc(CO3)- 1073 1074 labeled PSMA inhibitors containing functionalized imidazole rings have shown that charge 1075 1076 has a major impact on binding affinity [71-72]. In a series of probes based on the Glu-NH- 1077 1078 CO-NH-Lys PSMA-binding motif and single amino acid chelate (SAAC) systems containing 1079 1080 substituted di-imidazoles, increasing derivatization of the imidazole rings with negatively 1081 1082 1083 charged carboxylic acid functional groups resulted in significantly improved affinity for 1084 1085 PSMA [71]. Building on these findings, addition of carboxylate groups to our NODIA-Me 1086 1087 chelating platform may pave the way to a second generation of PSMA-targeting probes with 1088 1089 higher affinity. 1090 1091 Interestingly, however, the PSMA binding affinity of [64Cu]4 is almost 4-times lower 1092 1093 than that of [68Ga]4 (both potentially having an overall charge of -1), suggesting that binding 1094 1095 affinity of the probes is not a simple function of complex charge, but instead results from an 1096 1097 interplay of several factors. Similar to our findings, the 64Cu-complex of the recently reported 1098 1099 1100 PSMA-targeting probe (R)-NODAGA-Phe-Phe-D-Lys(suberoyl)-Lys-urea-Glu (CC34) 1101 68 1102 showed a higher Kd-value than its Ga-labeled counterpart [55]. 1103 1104 1105 1106 3.5. In vivo studies 1107 1108 The aim of this study was to demonstrate the radiochemical scope and utility of the BFC 1109 1110 NODIA-Me 1 for radiopharmaceutical design. Therefore, the stability and pharmacokinetic 1111 1112 profiles of the 68Ga- and 64Cu-labelled PSMA conjugate 4 were assessed in vivo. The 1113 1114 distribution and targeting of [68Ga]4 and [64Cu]4 in mice bearing subcutaneous PSMA- 1115 1116 1117 1118 1119 1120 Page 19 1121 1122 1123 1124 1125 positive LNCaP tumors was measured by small-animal PET imaging and ex vivo 1126 1127 biodistribution studies. 1128 1129 1130 1131 3.5.1. Small-animal PET imaging 1132 1133 The pharmacokinetics of [68Ga]4 and [64Cu]4 were assessed by small-animal PET imaging of 1134 1135 nude mice bearing LNCaP xenografts at 30 min, 1, 2 and 3 h post injection (p.i.) as well as 1136 1137 after 24 h for [64Cu]4. Corresponding maximum intensity projections (MIPs) for [68Ga]4 and 1138 1139 [64Cu]4 at 1 h p.i. are given in Figure 1. In addition, volume-of-interest (VOI) analysis was 1140 1141 1142 performed for both compounds on the LNCaP tumor, liver and kidneys (Figure 2). 1143 64 1144 The distribution of radiotracers differs from each other with [ Cu]4 exhibiting a slightly 1145 1146 higher uptake than [68Ga]4 in the majority of organs, which is potentially a consequence of its 1147 1148 higher lipophilicity compared to [68Ga]4 (Table 1). Overall, uptake of both compounds in 1149 1150 background organs was low due to their rapid blood pool clearance and renal excretion. Both 1151 1152 conjugates [68Ga]4 and [64Cu]4 showed specific accumulation in PSMA positive LNCaP 1153 1154 tumors (Figure 2A+C). Tumor uptake peaked at 30 min p.i. with 0.92% and 1.53% injected 1155 1156 activity per gram (% IA g-1) for [68Ga]4 and [64Cu]4, respectively. A slow washout from the 1157 1158 -1 1159 LNCaP tumors occurred for both compounds declining from 0.75% IA g at 1 h p.i. to 0.41% 1160 -1 68 64 68 1161 IA g at 3 h p.i. for [ Ga]4. The conjugate [ Cu]4 exhibited a slower washout than [ Ga]4 1162 1163 with 1.14% IA g-1 at 1 h p.i. decreasing to 0.88% IA g-1 at 3 h p.i. and a prolonged retention 1164 1165 with 0.39% IA g-1 still present in the tumor after 24 h. Compared to 68Ga-PSMA-HBEC-CC 1166 1167 with a tumor uptake of ~3.5% IA g-1 at 1 h p.i. in similar PET-imaging studies [46], the tumor 1168 1169 uptake of [68Ga]4 and [64Cu]4 in our studies at the same time point (1 h p.i.) was lower. 1170 1171 However, tumor uptake is known to be dependent on the PSMA expression level of the 1172 1173 specific batch of LNCaP-tumors and the total amount of the radioligand injected (~1 nmol of 1174 1175 1176 radioligand / mouse in our study), e.g., tumor uptake values in the report of Eder et al. [46] 1177 1178 1179 Page 20 1180 1181 1182 1183 -1 -1 1184 were ~3.5% IA g in their PET study (~0.5 nmol / mouse) compared to 7.70 ± 1.45% IA g 1185 1186 in their ex vivo biodistribution study (0.1 – 0.2 nmol / mouse), while others even reported 1187 1188 values of 15.8 ± 1.4% IA g-1 for an ex vivo biodistribution study with 10 pmol per mouse 1189 1190 [55]. Specific uptake of [68Ga]4 and [64Cu]4 was demonstrated by blockade of PSMA using 1191 1192 co-administration of PMPA. Blocking experiments resulted in a statistically significant 1193 1194 reduction of tumor uptake (P < 0.05 over all time points) (Figure 2B + D). 1195 1196 Liver uptake was very low for [68Ga]4 with 0.37% IA g-1 at 30 min p.i. reaching 1197 1198 essentially background levels at 3 h p.i. (Figure 2A). For [64Cu]4, uptake peaked at 30 min 1199 1200 -1 -1 1201 p.i. with 1.31% IA g and then plateaued at ~0.7–1.0% IA g up to 24 h (Figure 2C). At all 1202 64 68 1203 time points studied, the liver uptake of [ Cu]4 was slightly higher than that of [ Ga]4, which 1204 1205 might be attributed to differences in lipophilicity (Table 1) or, alternatively, suggest 1206 1207 instability of [64Cu]4 and superoxide dismutase (SOD1) dependent transchelation [73]. 1208 1209 Importantly, however, the liver uptake of about 1.0% IA g-1 for [64Cu]4 is comparable to the 1210 1211 liver uptake of the best candidates identified in an extensive study directed towards the 1212 1213 development of novel 64Cu-labeled PSMA-inhibitors based on the same lysine-urea- 1214 1215 glutamate scaffold, and stability of these probes was considered suitable for clinical 1216 1217 1218 translation [12]. 1219 68 64 1220 Both [ Ga]4 and [ Cu]4 exhibited rapid blood pool clearance and renal excretion. 1221 1222 Corresponding activity levels in the kidneys declined from 4.39% IA g-1 at 30 min p.i. to 1223 1224 0.88% IA g-1 after 3 h for [68Ga]4, and from 3.31% IA g-1 at 30 min p.i. to 1.05% IA g-1 after 1225 1226 3 h for [64Cu]4 (Figure 2A+C). Both compounds showed much lower kidney accumulation 1227 1228 and faster washout compared to [68Ga]Ga-PSMA-HBED-CC and [68Ga]Ga-PSMA-617, 1229 1230 consistent with their lower affinity for PSMA [46, 56]. In corresponding blocking studies, the 1231 1232 kidney uptake was significantly reduced for [68Ga]4, whereas only a slight decrease was 1233 1234 64 1235 noted for [ Cu]4 (Figure 2B+D). Despite the low affinity of both tracers for PSMA, the low 1236 1237 1238 Page 21 1239 1240 1241 1242 68 64 1243 accumulation and retention of [ Ga]4 and [ Cu]4 in the kidneys is still noteworthy. In 1244 64 1245 previous PET imaging studies of our Cu-labelled trisubstituted chelators (in absence of a 1246 1247 targeting vector), we found high uptake in the liver (~11–17% IA g-1) and the kidneys (~20– 1248 1249 40% IA g-1) at 1 h p.i. for the imidazole-type ligands, respectively [57]. It remained unclear 1250 1251 whether the uptake in liver and kidneys was a result of the positive charge or due to 1252 1253 demetallation/transchelation. The data of the current PET imaging study shows that the 1254 1255 biodistribution of both radiotracers is mainly determined by the PSMA targeting vector rather 1256 1257 than the positive overall charge of the metal chelates and that rapid clearance and low uptake 1258 1259 1260 in background organs is consistent with kinetically stable metal chelates. 1261 1262 1263 1264 3.5.2. Ex vivo biodistribution 1265 1266 In addition to the PET imaging studies, we also assayed the pharmacokinetics and the 1267 1268 stability of [68Ga]4 and [64Cu]4 by ex vivo biodistribution studies (Table 2). Corresponding 1269 1270 results of selected organs at 1 h p.i. are summarized in Figure 3. In general, results follow the 1271 1272 same trend as seen in the PET imaging experiments, however, absolute uptake values 1273 1274 obtained from ex vivo biodistribution studies were slightly higher than those calculated from 1275 1276 68 -1 1277 the PET imaging data. Uptake of [ Ga]4 was low in all organs with < 1% IA g at 1 h p.i. 1278 64 1279 with exception of the kidneys, which are known to express PSMA [74]. For [ Cu]4, low but 1280 1281 overall higher accumulation compared to that of [68Ga]4 was noted in the kidneys, liver, lung 1282 1283 and intestine. The kidney uptake for [68Ga]4 and [64Cu]4 with 6.28% IA g-1 and 4.96% IA g-1 1284 1285 was significantly lower than that reported for [68Ga]Ga-PSMA-HBED-CC and [68Ga]Ga- 1286 1287 PSMA-617 with 139.4% IA g-1 and 113.3% IA g-1, respectively [46, 56, 61]. Accumulation in 1288 1289 the kidneys was PSMA-specific and could be reduced by co-injection of PMPA by ~80% and 1290 1291 ~72% for [68Ga]4 and [64Cu]4, respectively (Table 2). Liver uptake of [64Cu]4 was higher 1292 1293 68 1294 than that of [ Ga]4 but was reduced by almost 50% after co-injection of PMPA. Compared 1295 1296 1297 Page 22 1298 1299 1300 1301 68 -1 68 -1 1302 to [ Ga]Ga-PSMA-HBED-CC (0.87% IA g ) and [ Ga]Ga-PSMA-617 (1.17% IA g ) [46, 1303 68 -1 1304 61], liver accumulation was lower for [ Ga]4 (0.30% IA g ), while it was about two-fold 1305 1306 higher for [64Cu]4 (2.69 % IA g-1). Nevertheless, this rather low value is not indicative of 1307 1308 complex instability and substantial transchelation in vivo. 1309 1310 Tumor accumulation of [68Ga]4 with 1.33% IA g-1 was lower than that of [64Cu]4 with 1311 1312 2.15% IA g-1. In comparison to [68Ga]Ga-PSMA-HBED-CC and [68Ga]Ga-PSMA-617, which 1313 1314 have reported tumor uptake of 7.70 and 8.47% IA g-1 [46, 61], respectively, [68Ga]4 and 1315 1316 [64Cu]4 exhibited lower accumulation in the tumor. This is in line with the lower PSMA 1317 1318 68 1319 binding affinity of our NODIA-Me based PSMA targeting probes compared to [ Ga]Ga- 1320 68 1321 PSMA-HBED-CC and [ Ga]Ga-PSMA-617 [55-56, 61]. A promising strategy for 1322 1323 improvement of binding affinity and tumor targeting properties may be the addition of 1324 1325 pendant carboxylic acids to the imidazole-groups of the NODIA-Me chelating moiety, as 1326 1327 these negatively charged functional groups were reported to lead to additional PSMA binding 1328 1329 interactions for a series of similar imidazole-containing PSMA targeting probes (vide supra) 1330 1331 [71]. PSMA specificity of [68Ga]4 and [64Cu]4 was confirmed by co-injection of PMPA 1332 1333 resulting in reduction of tumor uptake by ~88% and ~70%, respectively. Interestingly, the 1334 1335 68 64 1336 relatively low kidney accumulation and retention of [ Ga]4 and [ Cu]4 resulted in 1337 1338 significantly improved tumor-to-kidney ratios of 0.21 and 0.43, respectively. These values 1339 1340 are ~3-8fold higher than those of [68Ga]Ga-PSMA-HBED-CC and [68Ga]Ga-PSMA-617 with 1341 1342 0.055 and 0.074 [46, 56, 61]. 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 Page 23 1357 1358 1359 1360 1361 4. Conclusions 1362 1363 In the present work, the novel chelator NODIA-Me 1 was conjugated to the PSMA targeting 1364 1365 Glu-NH-CO-NH-Lys moiety to give the final bioconjugate 4, which was labelled with 1366 1367 gallium-68 and copper-64 in molar activities of ~20 MBq nmol-1. The metal-free as well as 1368 1369 the Ga and Cu complexes of 4 revealed different affinities to PSMA in competitive binding 1370 1371 studies underscoring the influence of the metal chelate on biological properties. Evaluation of 1372 1373 the 68Ga- and 64Cu-labelled conjugates by PET imaging and ex vivo biodistribution studies 1374 1375 demonstrated PSMA-specific tumor uptake. Both compounds exhibited low kidney uptake, 1376 1377 1378 rapid renal clearance, and only low uptake in non-target organs. Liver accumulation was very 1379 68 64 1380 low for [ Ga]4, while it was moderately increased for [ Cu]4, which could be either due to 1381 1382 the higher lipophilicity of the 64Cu-complex or suggest a very limited level of transchelation. 1383 1384 Strategies towards further improvement of binding affinity and tumor targeting of the 1385 1386 NODIA-Me based PSMA-binding probes may include introduction of additional negative 1387 1388 charges in the chelating unit by functionalization of the imidazoles with pendant carboxylic 1389 1390 acid arms and optimization of the spacer moiety. Overall, the results indicate that imidazole- 1391 1392 based TACN chelators show promise for the future development of 68Ga- and 64Cu-labelled 1393 1394 1395 radiopharmaceuticals. 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 Page 24 1416 1417 1418 1419 1420 Acknowledgements 1421 1422 JPH thanks the Department of Nuclear Medicine, University Hospital Freiburg, the German 1423 1424 Cancer Consortium (DKTK), and the German Cancer Research Center (DKFZ) for financial 1425 1426 support as well as the Swiss National Science Foundation (SNSF Professorship 1427 1428 PP00P2_163683) and the European Research Council (ERC-StG-2015, NanoSCAN – 1429 1430 676904). MDB thanks the Chemistry Department of the Albert-Ludwigs-University Freiburg 1431 1432 (Germany) for technical support, in particular Christoph Warth for MS measurements, 1433 1434 Manfred Keller for NMR measurements and Prof. Philipp Kurz for his continuous support. 1435 1436 1437 MDB also thanks the Research Commission of the University Freiburg as well as the Fonds 1438 1439 der chemischen Industrie for financial support. 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 Page 25 1475 1476 1477 1478 1479 References 1480 1481 [1] Bartholoma, M. D.; Louie, A. S.; Valliant, J. F.; Zubieta, J. Technetium and Gallium 1482 Derived Radiopharmaceuticals: Comparing and Contrasting the Chemistry of Two 1483 Important Radiometals for the Molecular Imaging Era. Chem Rev 2010; 110: 2903- 1484 2920. 1485 [2] Decristoforo, C.; Pickett, R. D.; Verbruggen, A. Feasibility and availability of 1486 (6)(8)Ga-labelled peptides. Eur J Nucl Med Mol Imaging 2012; 39 Suppl 1: S31-40. 1487 [3] Breeman, W. A.; de Blois, E.; Sze Chan, H.; Konijnenberg, M.; Kwekkeboom, D. J.; 1488 Krenning, E. P. (68)Ga-labeled DOTA-peptides and (68)Ga-labeled 1489 radiopharmaceuticals for positron emission tomography: current status of research, 1490 clinical applications, and future perspectives. Semin Nucl Med 2011; 41: 314-321. 1491 [4] Fani, M.; Andre, J. P.; Maecke, H. R. 68Ga-PET: a powerful generator-based 1492 alternative to cyclotron-based PET radiopharmaceuticals. Contrast Media Mol Imaging 1493 2008; 3: 67-77. 1494 1495 [5] Velikyan, I. 68Ga-Based radiopharmaceuticals: production and application relationship. 1496 Molecules 2015; 20: 12913-12943. 1497 [6] Bartholomä, M. D. Recent developments in the design of bifunctional chelators for 1498 metal-based radiopharmaceuticals used in Positron Emission Tomography. Inorg Chim 1499 Acta 2012; 389: 36-51. 1500 [7] Wadas, T. J.; Wong, E. H.; Weisman, G. R.; Anderson, C. J. Copper chelation 1501 chemistry and its role in copper radiopharmaceuticals. Curr Pharm Des 2007; 13: 3-16. 1502 [8] Ikotun, O. F.; Lapi, S. E. The rise of metal radionuclides in medical imaging: copper- 1503 64, zirconium-89 and yttrium-86. Future Med Chem 2011; 3: 599-621. 1504 [9] Blower, P. J.; Lewis, J. S.; Zweit, J. Copper radionuclides and radiopharmaceuticals in 1505 nuclear medicine. Nucl Med Biol 1996; 23: 957-980. 1506 [10] Dumont, R. A.; Tamma, M.; Braun, F.; Borkowski, S.; Reubi, J. C.; Maecke, H., et al. 1507 Targeted Radiotherapy of Prostate Cancer with a Gastrin-Releasing Peptide Receptor 1508 Antagonist Is Effective as Monotherapy and in Combination with Rapamycin. J Nucl 1509 Med 2013; 54: 762-769. 1510 [11] Dumont, R. A.; Deininger, F.; Haubner, R.; Maecke, H. R.; Weber, W. A.; Fani, M. 1511 1512 Novel Cu-64- and Ga-68-Labeled RGD Conjugates Show Improved PET Imaging of 1513 alpha(v)beta(3) Integrin Expression and Facile Radiosynthesis. J Nucl Med 2011; 52: 1514 1276-1284. 1515 [12] Banerjee, S. R.; Pullambhatla, M.; Foss, C. A.; Nimmagadda, S.; Ferdani, R.; 1516 Anderson, C. J., et al. 64Cu-Labeled Inhibitors of Prostate-Specific Membrane Antigen 1517 for PET Imaging of Prostate Cancer. J Med Chem 2014; 57: 2657-2669. 1518 [13] Dearling, J. L.; Paterson, B. M.; Akurathi, V.; Betanzos-Lara, S.; Treves, S. T.; Voss, S. 1519 D., et al. The ionic charge of copper-64 complexes conjugated to an engineered 1520 antibody affects biodistribution. Bioconjug Chem 2015; 26: 707-717. 1521 [14] Dearling, J. L.; Voss, S. D.; Dunning, P.; Snay, E.; Fahey, F.; Smith, S. V., et al. 1522 Imaging cancer using PET--the effect of the bifunctional chelator on the biodistribution 1523 of a (64)Cu-labeled antibody. Nucl Med Biol 2011; 38: 29-38. 1524 [15] Liu, Z.; Yan, Y.; Liu, S.; Wang, F.; Chen, X. 18F, 64Cu, and 68Ga Labeled RGD- 1525 Bombesin Heterodimeric Peptides for PET Imaging of Breast Cancer. Bioconjugate 1526 Chem 2009; 20: 1016-1025. 1527 [16] Li, Z.-B.; Chen, K.; Chen, X. 68Ga-labeled multimeric RGD peptides for microPET 1528 1529 imaging of integrin αvβ3 expression. Eur J Nucl Med Mol Imaging 2008; 35: 1100- 1530 1108. 1531 1532 1533 Page 26 1534 1535 1536 1537 1538 [17] Velikyan, I.; Maecke, H.; Langstrom, B. Convenient Preparation of 68Ga-Based PET- 1539 Radiopharmaceuticals at Room Temperature. Bioconjugate Chem 2008; 19: 569-573. 1540 [18] Jeong, J. M.; Hong, M. K.; Chang, Y. S.; Lee, Y.-S.; Kim, Y. J.; Cheon, G. J., et al. 1541 Preparation of a Promising Angiogenesis PET Imaging Agent: 68Ga-Labeled 1542 c(RGDyK)–Isothiocyanatobenzyl-1,4,7-Triazacyclononane-1,4,7-Triacetic Acid and 1543 Feasibility Studies in Mice. J Nucl Med 2008; 49: 830-836. 1544 [19] Ait-Mohand, S.; Fournier, P.; Dumulon-Perreault, V.; Kiefer, G. E.; Jurek, P.; Ferreira, 1545 C. L., et al. Evaluation of 64Cu-labeled bifunctional chelate-bombesin conjugates. 1546 Bioconjug Chem 2011; 22: 1729-1735. 1547 [20] Lane, S. R.; Nanda, P.; Rold, T. L.; Sieckman, G. L.; Figueroa, S. D.; Hoffman, T. J., et 1548 al. Optimization, biological evaluation and microPET imaging of copper-64-labeled 1549 bombesin agonists, [64Cu-NO2A-(X)-BBN(7-14)NH2], in a prostate tumor 1550 xenografted mouse model. Nucl Med Biol 2010; 37: 751-761. 1551 [21] Wieser, G.; Mansi, R.; Grosu, A. L.; Schultze-Seemann, W.; Dumont-Walter, R. A.; 1552 Meyer, P. T., et al. Positron emission tomography (PET) imaging of prostate cancer 1553 1554 with a gastrin releasing peptide receptor antagonist--from mice to men. Theranostics 1555 2014; 4: 412-419. 1556 [22] Notni, J.; Hermann, P.; Havlickova, J.; Kotek, J.; Kubicek, V.; Plutnar, J., et al. A 1557 triazacyclononane-based bifunctional phosphinate ligand for the preparation of 1558 multimeric 68Ga tracers for positron emission tomography. Chem Eur J 2010; 16: 1559 7174-7185. 1560 [23] Notni, J.; Šimeček, J.; Hermann, P.; Wester, H.-J. TRAP, a Powerful and Versatile 1561 Framework for Gallium-68 Radiopharmaceuticals. Chem Eur J 2011; 17: 14718-14722. 1562 [24] Notni, J.; Pohle, K.; Wester, H. J. Be spoilt for choice with radiolabelled RGD peptides: 1563 preclinical evaluation of (6)(8)Ga-TRAP(RGD)(3). Nucl Med Biol 2013; 40: 33-41. 1564 [25] Ferreira, C. L.; Lamsa, E.; Woods, M.; Duan, Y.; Fernando, P.; Bensimon, C., et al. 1565 Evaluation of Bifunctional Chelates for the Development of Gallium-Based 1566 Radiopharmaceuticals. Bioconjugate Chem 2010; 21: 531-536. 1567 [26] Tircsó, G.; Benyó, E. T.; Suh, E. H.; Jurek, P.; Kiefer, G. E.; Sherry, A. D., et al. (S)-5- 1568 (p-Nitrobenzyl)-PCTA, a Promising Bifunctional Ligand with Advantageous Metal Ion 1569 Complexation Kinetics. Bioconjugate Chem 2009; 20: 565-575. 1570 1571 [27] Ferreira, C. L.; Yapp, D. T. T.; Mandel, D.; Gill, R. K.; Boros, E.; Wong, M. Q., et al. 1572 68Ga Small Peptide Imaging: Comparison of NOTA and PCTA. Bioconjugate Chem 1573 2012; 23: 2239-2246. 1574 [28] Boal, A. K.; Rosenzweig, A. C. Structural biology of copper trafficking. Chem Rev 1575 2009; 109: 4760-4779. 1576 [29] Cai, H.; Fissekis, J.; Conti, P. S. Synthesis of a novel bifunctional chelator AmBaSar 1577 based on sarcophagine for peptide conjugation and (64)Cu radiolabelling. Dalton Trans 1578 2009, 10.1039/b902210d [doi] 5395-5400. 1579 [30] Cai, H.; Li, Z.; Huang, C. W.; Park, R.; Shahinian, A. H.; Conti, P. S. An improved 1580 synthesis and biological evaluation of a new cage-like bifunctional chelator, 4-((8- 1581 amino-3,6,10,13,16,19-hexaazabicyclo[6.6.6]icosane-1-ylamino)methyl)benzoic acid, 1582 for 64Cu radiopharmaceuticals. Nucl Med Biol 2010; 37: 57-65. 1583 [31] Di Bartolo, N.; Sargeson, A. M.; Smith, S. V. New 64Cu PET imaging agents for 1584 personalised medicine and drug development using the hexa-aza cage, SarAr. Org 1585 Biomol Chem 2006; 4: 3350-3357. 1586 [32] Di Bartolo, N. M.; Sargeson, A. M.; Donlevy, T. M.; Smith, S. V. Synthesis of a new 1587 1588 cage ligand, SarAr, and its complexation with selected transition metal ions for 1589 potential use in radioimaging. J Chem Soc, Dalton Trans 2001, 10.1039/b103242a 1590 2303-2309. 1591 1592 Page 27 1593 1594 1595 1596 1597 [33] Boswell, C. A.; Sun, X.; Niu, W.; Weisman, G. R.; Wong, E. H.; Rheingold, A. L., et 1598 al. Comparative in vivo stability of copper-64-labeled cross-bridged and conventional 1599 tetraazamacrocyclic complexes. J Med Chem 2004; 47: 1465-1474. 1600 [34] Garrison, J. C.; Rold, T. L.; Sieckman, G. L.; Figueroa, S. D.; Volkert, W. A.; Jurisson, 1601 S. S., et al. In vivo evaluation and small-animal PET/CT of a prostate cancer mouse 1602 model using 64Cu bombesin analogs: side-by-side comparison of the CB-TE2A and 1603 DOTA chelation systems. J Nucl Med 2007; 48: 1327-1337. 1604 [35] Hausner, S. H.; Kukis, D. L.; Gagnon, M. K.; Stanecki, C. E.; Ferdani, R.; Marshall, J. 1605 F., et al. Evaluation of [64Cu]Cu-DOTA and [64Cu]Cu-CB-TE2A chelates for targeted 1606 positron emission tomography with an alphavbeta6-specific peptide. Mol Imaging 1607 2009; 8: 111-121. 1608 [36] Sprague, J. E.; Peng, Y.; Fiamengo, A. L.; Woodin, K. S.; Southwick, E. A.; Weisman, 1609 G. R., et al. Synthesis, Characterization and In Vivo Studies of Cu(II)-64-Labeled 1610 Cross-Bridged Tetraazamacrocycle-amide Complexes as Models of Peptide Conjugate 1611 Imaging Agents. J Med Chem 2007; 50: 2527-2535. 1612 1613 [37] Sun, X.; Wuest, M.; Weisman, G. R.; Wong, E. H.; Reed, D. P.; Boswell, C. A., et al. 1614 Radiolabeling and in vivo behavior of copper-64-labeled cross-bridged cyclam ligands. 1615 J Med Chem 2002; 45: 469-477. 1616 [38] Wadas, T. J.; Anderson, C. J. Radiolabeling of TETA- and CB-TE2A-conjugated 1617 peptides with copper-64. Nat Protoc 2006; 1: 3062-3068. 1618 [39] Juran, S.; Walther, M.; Stephan, H.; Bergmann, R.; Steinbach, J.; Kraus, W., et al. 1619 Hexadentate bispidine derivatives as versatile bifunctional chelate agents for copper(II) 1620 radioisotopes. Bioconjug Chem 2009; 20: 347-359. 1621 [40] Stephan, H.; Walther, M.; Fahnemann, S.; Ceroni, P.; Molloy, J. K.; Bergamini, G., et 1622 al. Bispidines for dual imaging. Chemistry (Easton) 2014; 20: 17011-17018. 1623 [41] Comba, P.; Hunoldt, S.; Morgen, M.; Pietzsch, J.; Stephan, H.; Wadepohl, H. 1624 Optimization of pentadentate bispidines as bifunctional chelators for 64Cu positron 1625 emission tomography (PET). Inorg Chem 2013; 52: 8131-8143. 1626 [42] Cai, Z. X.; Anderson, C. J. Chelators for copper radionuclides in positron emission 1627 tomography radiopharmaceuticals. J Labelled Compd Radiopharm 2014; 57: 224-230. 1628 [43] Spang, P.; Herrmann, C.; Roesch, F. Bifunctional Gallium-68 Chelators: Past, Present, 1629 1630 and Future. Seminars in Nuclear Medicine 2016; 46: 373-394. 1631 [44] Zeglis, B. M.; Houghton, J. L.; Evans, M. J.; Viola-Villegas, N.; Lewis, J. S. 1632 Underscoring the Influence of Inorganic Chemistry on Nuclear Imaging with 1633 Radiometals. Inorg Chem 2014; 53: 1880-1899. 1634 [45] Holland, J. P.; Williamson, M. J.; Lewis, J. S. Unconventional Nuclides for 1635 Radiopharmaceuticals. Molecular Imaging 2010; 9: 1-20. 1636 [46] Eder, M.; Schäfer, M.; Bauder-Wüst, U.; Hull, W.-E.; Wängler, C.; Mier, W., et al. 1637 68Ga-Complex Lipophilicity and the Targeting Property of a Urea-Based PSMA 1638 Inhibitor for PET Imaging. Bioconjugate Chem 2012; 23: 688-697. 1639 [47] Israeli, R. S.; Powell, C. T.; Fair, W. R.; Heston, W. D. Molecular cloning of a 1640 complementary DNA encoding a prostate-specific membrane antigen. Cancer Res 1641 1993; 53: 227-230. 1642 [48] Horoszewicz, J. S.; Kawinski, E.; Murphy, G. P. Monoclonal antibodies to a new 1643 antigenic marker in epithelial prostatic cells and serum of prostatic cancer patients. 1644 Anticancer Res 1987; 7: 927-935. 1645 [49] Rong, S. B.; Zhang, J.; Neale, J. H.; Wroblewski, J. T.; Wang, S.; Kozikowski, A. P. 1646 1647 Molecular modeling of the interactions of glutamate carboxypeptidase II with its potent 1648 NAAG-based inhibitors. J Med Chem 2002; 45: 4140-4152. 1649 1650 1651 Page 28 1652 1653 1654 1655 1656 [50] Zhang, A. X.; Murelli, R. P.; Barinka, C.; Michel, J.; Cocleaza, A.; Jorgensen, W. L., et 1657 al. A remote arene-binding site on prostate specific membrane antigen revealed by 1658 antibody-recruiting small molecules. J Am Chem Soc 2010; 132: 12711-12716. 1659 [51] Eder, M.; Neels, O.; Muller, M.; Bauder-Wust, U.; Remde, Y.; Schafer, M., et al. Novel 1660 Preclinical and Radiopharmaceutical Aspects of [68Ga]Ga-PSMA-HBED-CC: A New 1661 PET Tracer for Imaging of Prostate Cancer. Pharmaceuticals (Basel) 2014; 7: 779-796. 1662 [52] Weineisen, M.; Schottelius, M.; Simecek, J.; Baum, R. P.; Yildiz, A.; Beykan, S., et al. 1663 68Ga- and 177Lu-Labeled PSMA I&T: Optimization of a PSMA-Targeted Theranostic 1664 Concept and First Proof-of-Concept Human Studies. J Nucl Med 2015; 56: 1169-1176. 1665 [53] Herrmann, K.; Bluemel, C.; Weineisen, M.; Schottelius, M.; Wester, H.-J.; Czernin, J., 1666 et al. Biodistribution and Radiation Dosimetry for a Probe Targeting Prostate-Specific 1667 Membrane Antigen for Imaging and Therapy. J Nucl Med 2015; 56: 855-861. 1668 [54] Chatalic, K. L.; Heskamp, S.; Konijnenberg, M.; Molkenboer-Kuenen, J. D.; Franssen, 1669 G. M.; Clahsen-van Groningen, M. C., et al. Towards Personalized Treatment of 1670 Prostate Cancer: PSMA I&T, a Promising Prostate-Specific Membrane Antigen- 1671 1672 Targeted Theranostic Agent. Theranostics 2016; 6: 849-861. 1673 [55] Gourni, E.; Canovas, C.; Goncalves, V.; Denat, F.; Meyer, P. T.; Maecke, H. R. (R)- 1674 NODAGA-PSMA: A Versatile Precursor for Radiometal Labeling and Nuclear 1675 Imaging of PSMA-Positive Tumors. PLoS One 2015; 10: e0145755. 1676 [56] Benešová, M.; Schäfer, M.; Bauder-Wüst, U.; Afshar-Oromieh, A.; Kratochwil, C.; 1677 Mier, W., et al. Preclinical Evaluation of a Tailor-Made DOTA-Conjugated PSMA 1678 Inhibitor with Optimized Linker Moiety for Imaging and Endoradiotherapy of Prostate 1679 Cancer. J Nucl Med 2015; 56: 914-920. 1680 [57] Gotzmann, C.; Braun, F.; Bartholomä, M. D. Synthesis, 64Cu-labeling and PET 1681 imaging of 1,4,7-triazacyclononane derived chelators with pendant azaheterocyclic 1682 arms. RSC Advances 2016; 6: 119-131. 1683 [58] Schmidtke, A.; Läppchen, T.; Weinmann, C.; Bier-Schorr, L.; Keller, M.; Kiefer, Y., et 1684 al. Gallium Complexation, Stability, and Bioconjugation of 1,4,7-Triazacyclononane 1685 Derived Chelators with Azaheterocyclic Arms. Inorg Chem 2017, 1686 10.1021/acs.inorgchem.7b01129. 1687 [59] Jilg, C. A.; Drendel, V.; Rischke, H. C.; Beck, T.; Vach, W.; Schaal, K., et al. 1688 1689 Diagnostic Accuracy of Ga-68-HBED-CC-PSMA-Ligand-PET/CT before Salvage 1690 Lymph Node Dissection for Recurrent Prostate Cancer. Theranostics 2017; 7: 1770- 1691 1780. 1692 [60] Zanzonico, P. Routine quality controlof clinical nucliear medicine instrumentation: a 1693 brief review. J Nucl Med 2009; 49: 1114-1131. 1694 [61] Benešová, M.; Bauder-Wüst, U.; Schäfer, M.; Klika, K. D.; Mier, W.; Haberkorn, U., et 1695 al. Linker Modification Strategies To Control the Prostate-Specific Membrane Antigen 1696 (PSMA)-Targeting and Pharmacokinetic Properties of DOTA-Conjugated PSMA 1697 Inhibitors. J Med Chem 2016; 59: 1761-1775. 1698 [62] Kratochwil, C.; Giesel, F. L.; Leotta, K.; Eder, M.; Hoppe-Tich, T.; Youssoufian, H., et 1699 al. PMPA for Nephroprotection in PSMA-Targeted Radionuclide Therapy of Prostate 1700 Cancer. J Nucl Med 2015; 56: 293-298. 1701 [63] Barinka, C.; Byun, Y.; Dusich, C. L.; Banerjee, S. R.; Chen, Y.; Castanares, M., et al. 1702 Interactions between human glutamate carboxypeptidase II and urea-based inhibitors: 1703 structural characterization. J Med Chem 2008; 51: 7737-7743. 1704 [64] Banerjee, S. R.; Pullambhatla, M.; Shallal, H.; Lisok, A.; Mease, R. C.; Pomper, M. G. 1705 1706 A modular strategy to prepare multivalent inhibitors of prostate-specific membrane 1707 antigen (PSMA). Oncotarget 2011; 2: 1244-1253. 1708 1709 1710 Page 29 1711 1712 1713 1714 1715 [65] Mesters, J. R.; Barinka, C.; Li, W.; Tsukamoto, T.; Majer, P.; Slusher, B. S., et al. 1716 Structure of glutamate carboxypeptidase II, a drug target in neuronal damage and 1717 prostate cancer. EMBO J 2006; 25: 1375-1384. 1718 [66] Zoller, M.; Schuhmacher, J.; Reed, J.; Maierborst, W.; Matzku, S. Establishment and 1719 Characterization of Monoclonal-Antibodies against an Octahedral Gallium Chelate 1720 Suitable for Immunoscintigraphy with Pet. J Nucl Med 1992; 33: 1366-1372. 1721 [67] Leplatte.F; Murase, I.; Martell, A. E. New Multidentate Ligands .6. Chelating 1722 Tendencies of N,N'-Di(2-Hydroxybenzyl) Ethylenediamine-N,N'-Diacetic Acid. J Am 1723 Chem Soc 1967; 89: 837-&. 1724 [68] Banerjee, S. R.; Chen, Z. P.; Pullambhatla, M.; Lisok, A.; Chen, J.; Mease, R. C., et al. 1725 Preclinical Comparative Study of Ga-68-Labeled DOTA, NOTA, and HBED-CC 1726 Chelated Radiotracers for Targeting PSMA. Bioconjugate Chem 2016; 27: 1447-1455. 1727 [69] Cheng, Y.; Prusoff, W. H. Relationship between Inhibition Constant (K1) and 1728 Concentration of Inhibitor Which Causes 50 Per Cent Inhibition (I50) of an Enzymatic- 1729 Reaction. Biochem Pharmacol 1973; 22: 3099-3108. 1730 1731 [70] Vandermerwe, M. J.; Boeyens, J. C. A.; Hancock, R. D. Crystallographic and 1732 Thermodynamic Study of Metal-Ion Size Selectivity in the Ligand 1,4,7- 1733 Triazacyclononane-N,N',N''-Triacetate. Inorg Chem 1985; 24: 1208-1213. 1734 [71] Maresca, K. P.; Hillier, S. M.; Lu, G. L.; Marquis, J. C.; Zimmerman, C. N.; Eckelman, 1735 W. C., et al. Small molecule inhibitors of PSMA incorporating technetium-99m for 1736 imaging prostate cancer: Effects of chelate design on pharmacokinetics. Inorg Chim 1737 Acta 2012; 389: 168-175. 1738 [72] Hillier, S. M.; Maresca, K. P.; Lu, G. L.; Merkin, R. D.; Marquis, J. C.; Zimmerman, C. 1739 N., et al. Tc-99m-Labeled Small-Molecule Inhibitors of Prostate-Specific Membrane 1740 Antigen for Molecular Imaging of Prostate Cancer. J Nucl Med 2013; 54: 1369-1376. 1741 [73] Zarschler, K.; Kubeil, M.; Stephan, H. Establishment of two complementary in vitro 1742 assays for radiocopper complexes achieving reliable and comparable evaluation of in 1743 vivo stability. RSC Advances 2014; 4: 10157-10164. 1744 [74] Silver, D. A.; Pellicer, I.; Fair, W. R.; Heston, W. D. W.; CordonCardo, C. Prostate- 1745 specific membrane antigen expression in normal and malignant human tissues. Clinical 1746 Cancer Research 1997; 3: 81-85. 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 Page 30 1770 1771 1772 1773 1774 Figure and Scheme Legends 1775 1776 1777 1778 Scheme 1. Selected chelators for copper-64 and gallium-68. 1779 1780 1781 1782 Scheme 2. Synthesis of PSMA targeting conjugate 4. i) HATU, DIPEA, DMF, r.t., 70%; ii) 5 1783 1784 M HCl, r.t., 91%. 1785 1786 1787 1788 Figure 1. Representative maximum intensity projections (MIPs) of (A) [68Ga]4 and (C) 1789 1790 64 1791 [ Cu]4 at 1 h p.i. in LNCaP tumor bearing mice. Blocking studies confirmed the specificity 1792 68 64 1793 of (B) [ Ga]4 and (D) [ Cu]4 for PSMA expression. White arrows indicate organs of 1794 1795 interest: T = tumor, K = kidney, L = liver. 1796 1797 1798 1799 Figure 2. Bar diagram of the volume-of-interest (VOI) analysis for (A) [68Ga]4 and (C) 1800 1801 [64Cu]4 in the tumor, the liver, and the kidneys from 30 min – 3 h post administration. 1802 1803 Corresponding VOI analysis after co-injection of PSMA blocking agent (PMPA) for (B) 1804 1805 [68Ga]4 and (D) [64Cu]4. For [64Cu]4, data is also provided for 24 h p.i. Legend applies for 1806 1807 1808 charts A-D. 1809 1810 1811 1812 Figure 3. Ex vivo biodistribution of [68Ga]4 and [64Cu]4 for selected organs in LNCaP tumor 1813 1814 bearing mice at 1 h p.i along with blocking studies. Data are expressed as %IA g-1 and 1815 1816 represent mean ± SD (n = 5). 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 Page 31 1829 1830 1831 1832 1833 Table 1. Analytical data of prepared compounds. 1834 HPLCUV/vis HPLCrad IC50 Compound Mass calc. Mass found d) Log Doct/PBS 1835 tR [min] tR [min] [nM] 1836 [nat/68Ga]4 1053.4432a) 1053.4426 15:33 15:50 -4.27 ± 0.08 176 ± 10 1837 nat/64 b) 1838 [ Cu]4 1048.4550 1048.4551 17:42 17:43 -3.99 ± 0.05 681 ± 7 + + + + 1839 a) [M – 2H] = [C49H68N12O10Ga] , b) [M – H] = [C49H69N12O10Cu] , UV/vis and 1840 radioactivity detector in series. 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 Page 32 1888 1889 1890 1891 68 64 1892 Table 2. Ex vivo biodistribution of [ Ga]4 and [ Cu]4 in mice bearing PSMA-positive 1893 -1 1894 LNCaP tumors at 1 h p.i along with blocking studies. Data are expressed as %IA g and 1895 1896 represent mean ± SD (n = 5). 1897 1898 1899 [68Ga]-4 [64Cu]-4 1900 1901 1 h 1 h blockade 1 h 1 h blockade 1902 1903 blood 0.37 ± 0.15 0.31 ± 0.16 0.31 ± 0.07 0.16 ± 0.01 1904 1905 1906 heart 0.17 ± 0.06 0.15 ± 0.08 0.45 ± 0.09 0.27 ± 0.04 1907 1908 lung 0.50 ± 0.13 0.29 ± 0.08 1.22 ± 0.56 0.49 ± 0.07 1909 1910 spleen 0.46 ± 0.44 0.13 ± 0.04 0.49 ± 0.05 0.21 ± 0.05 1911 1912 liver 0.30 ± 0.22 0.19 ± 0.02 2.69 ± 0.64 1.48 ± 0.27 1913 1914 pancreas 0.17 ± 0.10 0.11 ± 0.05 0.26 ± 0.01 0.14 ± 0.03 1915 1916 stomach 0.30 ± 0.39 0.13 ± 0.08 0.49 ± 0.22 0.25 ± 0.19 1917 1918 intestine 0.34 ± 0.20 0.20 ± 0.15 1.05 ± 0.10 0.83 ± 0.35 1919 1920 kidney 6.28 ± 0.92 1.25 ± 0.23 4.96 ± 0.79 1.39 ± 0.05 1921 1922 muscle 0.16 ± 0.16 0.14 ± 0.11 0.19 ± 0.01 0.62 ± 0.39 1923 1924 bone 0.21 ± 0.23 0.14 ± 0.04 0.19 ± 0.03 0.33 ± 0.28 1925 1926 tumor 1.33 ± 0.46 0.17 ± 0.13 2.15 ± 0.38 0.65 ± 0.04 1927 1928 tumor:blood 3.59 6.32 1929 1930 tumor:kidney 0.21 0.43 1931 1932 tumor:liver 4.43 0.80 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 Page 33 1947

SUPPORTING INFORMATION

In vitro and in vivo evaluation of the bifunctional chelator NODIA-

Me in combination with a prostate-specific membrane antigen

targeting vector

Tilman Läppchen1,2, Yvonne Kiefer1, Jason P. Holland1,3, and Mark D. Bartholomä1,*

1 Department of Nuclear Medicine, Medical Center – University of Freiburg, Faculty of

Medicine, University of Freiburg, Hugstetterstrasse 55, D-79106, Freiburg, Germany

2 Department of Nuclear Medicine, Inselspital, Bern University Hospital and University of Bern,

Freiburgstrasse, CH-3010 Bern, Switzerland

3 Department of Chemistry, University of Zürich, Winterthurerstrasse 190, CH-8057, Zürich,

Switzerland

* Corresponding Author:

Dr. Mark D. Bartholomä

Tel: +49-(761)-270-39600

Fax: +49-(761)-270-39300

E-mail: [email protected] Content

NMR data for compound 4 S3

Mass spectrometry S4

HPLC chromatograms S9

Competitive cell binding experiments S13

Page S2 NMR characterization

1 Figure S1. H-NMR spectrum of compound 4 in D2O (pD ≤ 2).

Figure S2. 13C-NMR of compound 4 in D2O (pD ≤ 2) Page S3 Mass spectrometry

Figure S3. LR-ESI(+)-mass spectrum of 2.

D:\data_2017\klind13s_hr02 9/28/2017 3:24:12 PM nai.psma.tbu3

klind13s_hr02 #1 RT: 0.01 AV: 1 NL: 2.00E7 T: FTMS + p ESI Full lock ms [85.00-1700.00] 798.5007 C 43 H 68 O9 N 5 = 798.5012 -0.6057 ppm 100

50 799.5035

45 RelativeAbundance 40

35 820.4823 C 43 H 67 O9 N 5 Na = 820.4831 30 -0.9749 ppm

15 821.4857

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 m/z

+ + Figure S4. HR-ESI(+)-mass spectrum of 2. Calc. for [M+H] = [C43H68N5O9] = 798.5012.

Page S4 - 3 tBu

- 2 tBu

- tBu

Figure S5. LR-ESI(+)-mass spectrum of 3.

D:\data_2018\klind25s_hr02 2/5/2018 3:00:24 PM nodia.me.nai.ahx.psma.tbu

klind25s_hr02 #1 RT: 0.01 AV: 1 NL: 5.95E5 T: FTMS + p ESI Full ms [125.00-2000.00] 1155.7301 C 61 H 95 O10 N 12 = 1155.7289 1.0721 ppm 100

45 RelativeAbundance 40

219.1248 15 C 13 H 17 O2 N = 219.1254 -2.8497 ppm 10 1158.7386

0 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 m/z

+ + Figure S6. HR-ESI(+)-mass spectrum of 3. Calc. for [M+H] = [C61H95N12O10] = 1155.7289.

Page S5 Figure S7. LR-ESI(+)-mass spectrum of 4.

D:\data_2018\klind26s_hr05 2/7/2018 4:42:20 PM nodia.ma.nai.ahx.psma

klind26s_hr05 #1 RT: 0.02 AV: 1 NL: 1.62E6 T: FTMS + p ESI Full lock ms [100.00-2000.00] 987.5399 C 49 H 71 O10 N 12 = 987.5411 -1.1602 ppm 100

45 RelativeAbundance 40

30 840.4886 C 46 H 64 O7 N 8 = 840.4892 25 -0.7142 ppm

668.4019 989.5461 15 C 36 H 54 O7 N 5 = 668.4018 0.2111 ppm 10

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 m/z

+ + Figure S8. HR-ESI(+)-mass spectrum of 4. Calc. for [M+H] = [C49H71N12O10] = 987.5411.

Page S6 D:\data_2018\klind27s_hr05 2/9/2018 1:49:26 PM Ga.nodia.me.naui.ahx.psma

klind27s_hr05 #1 RT: 0.00 AV: 1 NL: 1.49E7 T: FTMS + p ESI Full lock ms [100.00-2000.00] 1053.4426 100

85 1055.4429

45 RelativeAbundance

528.5118 20

15 506.5299

10 529.5151 304.2614 906.3911 453.6981 880.4109 908.3902 1075.4249 1077.4243 5

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 m/z

+ + Figure S9. HR-ESI(+)-mass spectrum of Ga-4. Calc. for [M – 2H] = [C49H68N12O10Ga] =

1053.4432.

Page S7 D:\data_2018\klind28s_hr06 2/9/2018 2:03:11 PM Cu.nodia.me.naui.ahx.psma

klind28s_hr06 #1 RT: 0.00 AV: 1 NL: 2.78E7 T: FTMS + p ESI Full ms [100.00-2000.00]

1048.4551 z=1 100

1050.4559 65 z=1

45 RelativeAbundance 40

20 528.5115 451.2038 z=1 15 z=2 1052.4611 C 49 H 56 O10 N 7 = 451.2039 10 945.4518 z=1 1070.4357 -0.2628 ppm 529.5149 589.2457 z=1 z=1 z=1 z=1 5

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 m/z

+ + Figure S10. HR-ESI(+)-mass spectrum of Cu-4. Calc. for [M – H] = [C49H69N12O10Cu] =

1048.4550.

Page S8 HPLC chromatography

Figure S11. UV/vis trace of HPLC chromatogram of 4.

Figure S12. UV/vis trace of HPLC chromatogram of Ga-4.

Page S9 Figure S13. Radioactivity trace of HPLC chromatogram of [68Ga]4.

Figure S14. Radio-HPLC chromatogram of [68Ga]4 after 4 h incubation in human serum.

Page S10 Figure S15. UV/vis trace of HPLC chromatogram of Cu-4.

Figure S16. Radioactivity trace of HPLC chromatogram of [64Cu]4.

Page S11 Figure S17. Radio-HPLC chromatogram of [64Cu]4 after 24 h incubation in human serum.

Page S12 Figure S18. Determination of binding affinity of compounds 4, Ga-4 and Cu-4 by competitive titration on LNCaP cells using [68Ga]Ga-PSMA-HBED-CC as radioligand.

Page S13