Towards Targeted Alpha Therapy with Actinium-225: Chelators for Mild Condition Radiolabeling and Targeting PSMA—A Proof of Concept Study

Simple Summary In nuclear medicine, therapeutic methods are used increasingly, in which tumors are destroyed by ionizing radiation that cannot or can only be treated insufficiently by other methods like surgery or chemotherapy. Targeted alpha therapy (TAT) is a promising method with increasing importance that facilitates new treatment options for advanced and late-stage cancer diseases. The effectiveness of alpha-emitting radionuclides is characterized by a higher linear energy transfer and a higher biological efficacy, compared to therapeutic approaches with beta emitters. 225Ac is an alpha emitter with favorable nuclear properties for radiopharmaceutical applications. The aim of our research was to find a universal chelator that enables the attachment of sensitive bio(macro)molecules and allows 225Ac-radiolabeling under mild conditions. An aza-macrocycle-derived mcp chelator with functional groups for universal connection of biomolecules using convenient click chemistry was developed for the 225Ac-labeling. The resulting 225Ac-radioconjugates were analyzed in vitro and in vivo, showing a high receptor affinity on tumor cells and a high tumor accumulation in tumor-bearing mice. Abstract Currently, targeted alpha therapy is one of the most investigated topics in radiopharmaceutical cancer management. Especially, the alpha emitter 225Ac has excellent nuclear properties and is gaining increasing popularity for the treatment of various tumor entities. We herein report on the synthesis of two universal 225Ac-chelators for mild condition radiolabeling and binding to conjugate molecules of pharmacological interest via the copper-mediated click chemistry. A convenient radiolabeling procedure was investigated as well as the complex stability proved for both chelators and two PSMA (prostate-specific membrane antigen)-targeting model radioconjugates. Studies regarding affinity and cell survival were performed on LNCaP cells followed by biodistribution studies, which were performed using LNCaP tumor-bearing mice. High efficiency radiolabeling for all conjugates was demonstrated. Cell binding studies revealed a fourfold lower cell affinity for the PSMA radioconjugate with one targeting motif compared to the radioconjugate owing two targeting motifs. Additionally, these differences were verified by in vitro cell survival evaluation and biodistribution studies, both showing a higher cell killing efficiency for the same dose, a higher tumor uptake (15%ID/g) and a rapid whole body clearance after 24 h. The synthesized chelators will overcome obstacles of lacking stability and worse labeling needs regarding 225Ac complexation using the DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetrayl)tetraacetic acid) chelator. Moreover, the universal functionalization expands the coverage of these chelators in combination with any sensitive bio(macro)molecule, thus improving treatment of any addressable tumor target.


Introduction
Targeted alpha therapy (TAT) provides new treatment options for advanced and late-stage cancer diseases but also offers new possibilities for radiopharmaceutical cancer research. Compared to other therapies (e.g., external radiotherapy and radioligand therapy (RLT) based on beta emitters), TAT is putatively more efficient in eliciting cell death due to the exceedingly high ionizing density. The short range of alpha particles minimizes the unwanted irradiation for surrounding healthy tissue and reduces possible side effects [1]. Unfortunately, only a few alpha-emitting radionuclides are currently available, owing appropriate qualities for medical applications.
Besides the alpha-emitting radionuclide astatine-211, which is covalently bound to the targeting moiety [2][3][4][5], today's focus lies on radiometals, particularly on bismuth and radium radionuclides, as well as on thorium-227 and actinium-225 [6,7]. Each radiometal has its advantages and obstacles. Bismuth-212 (t 1/2 = 61 min) or the lead-212/bismuth-212 in vivo generator system and bismuth-213 (t 1/2 = 46 min) showed promising results in clinical and preclinical studies [8,9]. Still, there is a lack of applications, most likely induced by the high amount of administered activities resulting in adverse reactions and a loss of flexibility due to the short half-lives. Radium-223 (t 1/2 = 11 d) and radium-224 (t 1/2 = 3.6 d) possess suitable physical properties by emitting multiple alpha particles in their decay chains with total decay energies of about 30 MeV resulting in a high linear energy transfer. Xofigo ® is the only clinically relevant application with EMA-and FDA-approval for radium-223 to date and is used in the form of [ 223 Ra]RaCl 2 to treat metastatic prostate cancer [10]. The major limitation of radium is the lack of an organic or inorganic framework to bind this radionuclide stably [11]. Despite several approaches to coordinate Ra 2+ , (e.g., using macrocyclic ligands [12][13][14][15], inorganic nanomatrices such as nanoparticles based on BaSO 4 [16,17], barium ferrite [18], TiO 2 and hydroxyapatite [19,20] or polyoxometalates [21]), an ideal carrier was not found, yet.
Our recent work focuses on new macropa-based ligands for the development of actinium-225-conjugates, combining both the extremely high complex stability and a clickable functionalization site for attaching biological targeting molecules. We aimed to synthesize a chelator, which can be functionalized for conjugations using click chemistry by the [3+2] Huisgen cycloaddition approach (CuAAC). For this purpose, we designed two propargyl-functionalized mcp-chelators based on the diaza-18-crown-6-skeleton. In a preclinical proof-of-concept-study, this chelator was conjugated to an azide-functionalized carboxylic acid as well as an azide-functionalized PSMA-binding motif. We compared the Cancers 2021, 13,1974 3 of 19 radiolabeling efficiency of our new chelators and their PSMA conjugates, performed cellbinding affinity and colony survival assays on PSMA-positive human prostate adenocarcinoma (LNCaP) cells. Finally, both new tracers were characterized regarding biodistribution in healthy and LNCaP tumor-bearing mice.

Chelator Development
Our new targeting conjugates for an improved radiolabeling with actinium-225 are based on a macropa-derived chelator and the well-characterized targeting biomolecule PSMA-617. We designed two chelators functionalized either with one or two alkyne moieties for an easy and quick conjugation with biomolecules of interest using the coppercatalyzed azide-alkyne-cycloaddition reaction [28]. For comparison, we synthesized the chelator macropa according to previously published procedures [29]. The synthesis of the novel mono-and di-alkyne-modified chelators mcp-M-click and mcp-D-click is presented in Scheme 1. carboxylic acid as well as an azide-functionalized PSMA-binding motif. We compared the radiolabeling efficiency of our new chelators and their PSMA conjugates, performed cellbinding affinity and colony survival assays on PSMA-positive human prostate adenocarcinoma (LNCaP) cells. Finally, both new tracers were characterized regarding biodistribution in healthy and LNCaP tumor-bearing mice.

Radiolabeling and Complex Stability
To compare the efficiency of the synthesized chelators and their conjugates, 100 kBq of actinium-225 were incubated with each ligand at different concentrations (10 −5 M; 10 −6 M; 10 −7 M). The radiochemical yield was determined after one hour of reaction time using normal phase and reversed-phase TLC. Whereas one TLC-system was designed to move free actinium-225 with the solvent front, the other system was developed only to move the actinium-225 labeled complexes. Concentration-dependent radiochemical yields for all ligands and 7-day-longterm yields, which were calculated using both TLC systems, are presented in Table 1.
All investigated ligands show excellent complex formation abilities with [ 225 Ac]Ac 3+ , even at the low ligand concentration of 10 −7 M. No activity-release was detected over seven days. Furthermore, the improved labeling yields obtained at the lowest concentrations after seven days showing that complexation might not be finished after one hour. Comparing mcp-M-click and mcp-D-click with the non-functionalized reference ligand macropa, the slight difference in the affinity might result from the steric hindrance caused by the propargyl moiety. A further enhancement even in yields at low concentrations and long-time stabilities was found using the model conjugate [ 225 Ac]Ac-mcp-M-COOH containing an additional triazole moiety, presumably due to the additional nitrogen donor

Radiolabeling and Complex Stability
To compare the efficiency of the synthesized chelators and their conjugates, 100 kBq of actinium-225 were incubated with each ligand at different concentrations (10 −5 M; 10 −6 M; 10 −7 M). The radiochemical yield was determined after one hour of reaction time using normal phase and reversed-phase TLC. Whereas one TLC-system was designed to move free actinium-225 with the solvent front, the other system was developed only to move the actinium-225 labeled complexes. Concentration-dependent radiochemical yields for all ligands and 7-day-longterm yields, which were calculated using both TLC systems, are presented in Table 1.
All investigated ligands show excellent complex formation abilities with [ 225 Ac]Ac 3+ , even at the low ligand concentration of 10 −7 M. No activity-release was detected over seven days. Furthermore, the improved labeling yields obtained at the lowest concentrations after seven days showing that complexation might not be finished after one hour. Comparing mcp-M-click and mcp-D-click with the non-functionalized reference ligand macropa, the slight difference in the affinity might result from the steric hindrance caused by the propargyl moiety. A further enhancement even in yields at low concentrations and longtime stabilities was found using the model conjugate [ 225 Ac]Ac-mcp-M-COOH containing an additional triazole moiety, presumably due to the additional nitrogen donor atom.
[ 225 Ac]Ac-mcp-M-COOH thereby predicts excellent labeling properties for the PSMA conjugates as well. Higher yields were obtained as well using the PSMA-functionalized conjugates mcp-M-PSMA and mcp-D-PSMA. Unfortunately, a suitable reversed-phase iTLC system for the ligands mcp-M-PSMA and mcp-D-PSMA was not found so far to determine the radiochemical yields. The resolution of our standard system was not efficient, most likely due to the bigger size of these molecules compared to the basic chelators, different protonation steps and possible conformational structures. Additionally, logD 7.4 values were determined to be −2.8 ± 0.1 for the complex [ 225 Ac]Ac-mcp-M-PSMA and −2.5 ± 0.1 for the complex [ 225 Ac]Ac-mcp-D-PSMA, respectively. Additionally, the stability of the 225 Ac-labeled conjugates in acetate buffer was investigated by HPLC analyses over a period of ten days. Representative chromatograms of [ 225 Ac]Ac-mcp-M-PSMA and [ 225 Ac]Ac-mcp-D-PSMA with and without the addition of the antioxidant excipients ascorbic acid (Asc) or 2,5-dihydroxybenzoic acid (DHB), respectively, are shown in Figure 1. Unfortunately, a suitable reversed-phase iTLC system for the ligands mcp-M-PSMA and mcp-D-PSMA was not found so far to determine the radiochemical yields. The resolution of our standard system was not efficient, most likely due to the bigger size of these molecules compared to the basic chelators, different protonation steps and possible conformational structures. Additionally, logD7.4 values were determined to be −2.8 ± 0.1 for the complex [ 225 Ac]Ac-mcp-M-PSMA and −2.5 ± 0.1 for the complex [ 225 Ac]Ac-mcp-D-PSMA, respectively. Additionally, the stability of the 225 Ac-labeled conjugates in acetate buffer was investigated by HPLC analyses over a period of ten days. Representative chromatograms of [ 225 Ac]Ac-mcp-M-PSMA and [ 225 Ac]Ac-mcp-D-PSMA with and without the addition of the antioxidant excipients ascorbic acid (Asc) or 2,5-dihydroxybenzoic acid (DHB), respectively, are shown in Figure 1. The radio-HPLC-chromatograms in Figure 1 show two to three signals originating from radionuclides of the actinium-225 decay chain. Consequently, analyses of the respective HPLC fractions were performed using a high resolution gamma detector. The peak with a retention time of ~1.8 min was identified to be uncomplexed francium-221. The The radio-HPLC-chromatograms in Figure 1 show two to three signals originating from radionuclides of the actinium-225 decay chain. Consequently, analyses of the respective HPLC fractions were performed using a high resolution gamma detector. The peak with a retention time of~1.8 min was identified to be uncomplexed francium-221. The other two signals were identified as a mixture of [ 213 Bi]Bi-and [ 225 Ac]Ac-labeled PSMA conjugates.
Regarding the intensively discussed side-effects that are often caused by daughter nuclides, this could be an additional advantage when the chelators form stable complexes with both metal ions actinium-225 and bismuth-213. Nevertheless, recoiling daughter nuclides like bismuth-213 remains an aspect to be considered [30,31], since breaking of chemical bonds represents an inevitable result of alpha decay no matter which chelator is used.
Most importantly, radiolabeled mcp-M-PSMA as well as mcp-D-PSMA are stable in acetate buffer at room temperature for at least 48 h without antioxidants as demonstrated in Figure 1. However, the addition of 0.1 M DHB increases the stability of both PSMA derivatives by preventing radiolytic degradation to a certain extent. The presence of 0.1 M ascorbic acid did not stabilize the 225 Ac-labeled conjugates and led to less complex stability and transchelation (verified by TLC). Although DHB appears to be more effective than ascorbic acid in retarding the dissociation by ionizing radiation, that might not apply for other 225 Ac-based radiopharmaceuticals. Therefore, we conclude that the stability evaluation and stabilization needs to be assessed for each individual radioconjugate. In contrast to other previously published results on crown-based actinium-chelators [32], we do not conclude that actinium and alpha particle emitting radiopharmaceuticals in general, are only applicable in the first hours after production. To the best of our knowledge, future pharmaceuticals using our radioconjugate would not have to be produced in-house solely and possess adequate stability at an appropriate molar activity of 5 MBq/nmol (regarding the total ligand amount).

Characterization of the PSMA-Binding Properties
Saturation binding studies using PSMA-positive LNCaP cells (see SI, Figures S14 and S15 for PSMA expression analysis) in combination with radiolabeled mcp-M-PSMA or mcp-D-PSMA, respectively, were performed to evaluate the binding specificity and affinity of the PSMA targeting variants. Incubation of [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA (30 nM) with LNCaP cells revealed a specific binding to PSMA, which was blocked significantly by the addition of an excess of PSMA-617 ( Figure 2). Regarding the intensively discussed side-effects that are often caused by daughter nuclides, this could be an additional advantage when the chelators form stable complexes with both metal ions actinium-225 and bismuth-213. Nevertheless, recoiling daughter nuclides like bismuth-213 remains an aspect to be considered [30,31], since breaking of chemical bonds represents an inevitable result of alpha decay no matter which chelator is used.
Most importantly, radiolabeled mcp-M-PSMA as well as mcp-D-PSMA are stable in acetate buffer at room temperature for at least 48 h without antioxidants as demonstrated in Figure 1. However, the addition of 0.1 M DHB increases the stability of both PSMA derivatives by preventing radiolytic degradation to a certain extent. The presence of 0.1 M ascorbic acid did not stabilize the 225 Ac-labeled conjugates and led to less complex stability and transchelation (verified by TLC). Although DHB appears to be more effective than ascorbic acid in retarding the dissociation by ionizing radiation, that might not apply for other 225 Ac-based radiopharmaceuticals. Therefore, we conclude that the stability evaluation and stabilization needs to be assessed for each individual radioconjugate. In contrast to other previously published results on crown-based actinium-chelators [32], we do not conclude that actinium and alpha particle emitting radiopharmaceuticals in general, are only applicable in the first hours after production. To the best of our knowledge, future pharmaceuticals using our radioconjugate would not have to be produced in-house solely and possess adequate stability at an appropriate molar activity of 5 MBq/nmol (regarding the total ligand amount).

Characterization of the PSMA-Binding Properties
Saturation binding studies using PSMA-positive LNCaP cells (see SI, Figures    Ac]Ac-mcp-D-PSMA, respectively, the equilibrium dissociation constants K d were calculated by the GraphPad Prism software using a "One site-total and nonspecific binding" analysis ( Figure 3). A K d of (38.44 ± 8.22) nM and a B max of (0.163 ± 0.011) pmol/mg protein were determined for [ 225 Ac]Ac-mcp-M-PSMA, whereas a K d of (9.92 ± 0.62) nM and a B max of (1.528 ± 0.023) pmol/mg protein were calculated for [ 225 Ac]Ac-mcp-M-PSMA. The obtained data indicate an improved binding affinity of the dimeric derivative compared to the monomer. These results are in line with findings published by Schäfer et al. [33] and Frei et al. [34] revealing the outperforming cell binding and uptake properties of dimeric PSMA samples. The presence of two independent PSMA-binding motifs within the dimeric [ 225 Ac]Ac-mcp-D-PSMA allows the compound to bind first with one motif to the respective antigen. The remaining (unoccupied) binding motif is now in close spatial proximity to other target sites on the cell surface and can bind to a second PSMA receptor. This divalent binding behavior stabilizes the [ 225 Ac]Ac-mcp-D-PSMA interaction as transient unbinding of a single site does not necessarily result in the release of the captured compound unless the dissociation occurs at both sites [35]. Accordingly, the equilibrium dissociation constant of the monomeric [ 225 Ac]Ac-mcp-M-PSMA is approximately three times higher than that of the dimeric compound, suggesting that monovalent binding is less stable than bivalent binding. [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA, respectively, the equilibrium dissociation constants Kd were calculated by the GraphPad Prism software using a "One sitetotal and nonspecific binding" analysis ( Figure 3). A Kd of (38.44 ± 8.22) nM and a Bmax of (0.163 ± 0.011) pmol/mg protein were determined for [ 225 Ac]Ac-mcp-M-PSMA, whereas a Kd of (9.92 ± 0.62) nM and a Bmax of (1.528 ± 0.023) pmol/mg protein were calculated for [ 225 Ac]Ac-mcp-M-PSMA. The obtained data indicate an improved binding affinity of the dimeric derivative compared to the monomer. These results are in line with findings published by Schäfer et al. [33] and Frei et al. [34] revealing the outperforming cell binding and uptake properties of dimeric PSMA samples. The presence of two independent PSMA-binding motifs within the dimeric [ 225 Ac]Ac-mcp-D-PSMA allows the compound to bind first with one motif to the respective antigen. The remaining (unoccupied) binding motif is now in close spatial proximity to other target sites on the cell surface and can bind to a second PSMA receptor. This divalent binding behavior stabilizes the [ 225 Ac]Ac-mcp-D-PSMA interaction as transient unbinding of a single site does not necessarily result in the release of the captured compound unless the dissociation occurs at both sites [35]. Accordingly, the equilibrium dissociation constant of the monomeric [ 225 Ac]Ac-mcp-M-PSMA is approximately three times higher than that of the dimeric compound, suggesting that monovalent binding is less stable than bivalent binding.

Cytotoxicity of 225 Ac-Labeled PSMA Derivatives
To evaluate the cytotoxicity in vitro induced by [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA, LNCaP cells were exposed once for 1 or 4 h, respectively, and the cell number was determined 24 h after the treatment by DNA staining (Figure 4).

Cytotoxicity of 225 Ac-Labeled PSMA Derivatives
To evaluate the cytotoxicity in vitro induced by [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Acmcp-D-PSMA, LNCaP cells were exposed once for 1 or 4 h, respectively, and the cell number was determined 24 h after the treatment by DNA staining (Figure 4). [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA, respectively, the equilibrium dissociation constants Kd were calculated by the GraphPad Prism software using a "One sitetotal and nonspecific binding" analysis ( Figure 3). A Kd of (38.44 ± 8.22) nM and a Bmax of (0.163 ± 0.011) pmol/mg protein were determined for [ 225 Ac]Ac-mcp-M-PSMA, whereas a Kd of (9.92 ± 0.62) nM and a Bmax of (1.528 ± 0.023) pmol/mg protein were calculated for [ 225 Ac]Ac-mcp-M-PSMA. The obtained data indicate an improved binding affinity of the dimeric derivative compared to the monomer. These results are in line with findings published by Schäfer et al. [33] and Frei et al. [34] revealing the outperforming cell binding and uptake properties of dimeric PSMA samples. The presence of two independent PSMA-binding motifs within the dimeric [ 225 Ac]Ac-mcp-D-PSMA allows the compound to bind first with one motif to the respective antigen. The remaining (unoccupied) binding motif is now in close spatial proximity to other target sites on the cell surface and can bind to a second PSMA receptor. This divalent binding behavior stabilizes the [ 225 Ac]Ac-mcp-D-PSMA interaction as transient unbinding of a single site does not necessarily result in the release of the captured compound unless the dissociation occurs at both sites [35]. Accordingly, the equilibrium dissociation constant of the monomeric [ 225 Ac]Ac-mcp-M-PSMA is approximately three times higher than that of the dimeric compound, suggesting that monovalent binding is less stable than bivalent binding.

Cytotoxicity of 225 Ac-Labeled PSMA Derivatives
To evaluate the cytotoxicity in vitro induced by [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA, LNCaP cells were exposed once for 1 or 4 h, respectively, and the cell number was determined 24 h after the treatment by DNA staining (Figure 4).  The relative number of LNCaP cells decreased with increasing radioactive doses and with extended incubation time. Already after 1 h of exposure, the proliferation rate was reduced by more than 10% at the second highest dose of 50 kBq/mL. The relative cell number is further decreased by approximately one third at the highest applied dose after 4 h of exposure to [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA. However, the incubation with [ 225 Ac]Ac 3+ as non-targeted activity control reduced the proliferation to a similar extent.
To determine whether the antiproliferative activity of the 225 Ac-labeled PSMA derivatives is associated with triggering programmed cell death pathways, the induction of apoptosis was analyzed by measuring the level of phosphatidylserine exposure on the outer leaflet of the cell membrane using an annexin V based assay ( Figure 5). The incubation of human prostate adenocarcinoma cells with the 225 Ac-labeled PSMA derivatives or with non-targeted [ 225 Ac]Ac 3+ , respectively, resulted in increased staining for exposed phosphatidylserine beginning at 4 h, indicating the induction of apoptosis. The magnitude of annexin V binding grew steadily until 8 h owing to more phosphatidylserine exposure by the progression of the cellular apoptotic program. Noteworthy, no appreciable loss of membrane integrity due to secondary necrosis was detected during this period (data not shown). Phosphatidylserine exposure appears dose-and time-dependent, whereas no substantial difference was observed between the treatment with [ 225 Ac]Ac-mcp-M-PSMA, [ 225 Ac]Ac-mcp-D-PSMA or non-targeted [ 225 Ac]Ac 3+ . This is in agreement with the data obtained by the proliferation assay and suggests that 225 Ac-induced cell damage is independent of PSMA targeting. Obviously, [ 225 Ac]Ac 3+ added directly to the cell culture media appears to be as effective as the specifically binding 225 Ac-labeled ligands. However, these observations originate from the experimental design. Under the static (flow-free) conditions of a two-dimensional cell culture in a simple petri dish, molecular targeting and cell binding via one or two PSMA-binding motifs plays no-or just a negligible-role for the efficacy of

Clonogenicity in Response to the Incubation with 225 Ac-Labeled PSMA Derivatives
A colony formation assay, performed to assess the clonogenic survival of LNCaP cells after treatment with increasing activity (0.05 kBq-50 kBq/mL) of [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA, respectively, revealed a time-and dose-dependent decrease in colony-forming cells (see SI, Table S1). Furthermore, notable differences in the for-

Clonogenicity in Response to the Incubation with 225 Ac-Labeled PSMA Derivatives
A colony formation assay, performed to assess the clonogenic survival of LNCaP cells after treatment with increasing activity (0.05 kBq-50 kBq/mL) of [ 225 Ac]Ac-mcp-M-PSMA or [ 225 Ac]Ac-mcp-D-PSMA, respectively, revealed a time-and dose-dependent decrease in colony-forming cells (see SI, Table S1). Furthermore, notable differences in the formation of colonies upon incubation with PSMA-targeted ([ 225 Ac]Ac-mcp-M-PSMA and [ 225 Ac]Ac-mcp-D-PSMA) and [ 225 Ac]Ac 3+ as non-targeted activity control were observed. The one-hour treatment with 5 kBq/mL resulted in 19.4% of the reference amount of colonies for the monomeric PSMA derivative and in 3.1% for the dimeric counterpart. In contrast, the clonogenic cell survival was higher (32.9%) after the treatment with the identical activity concentration of non-targeted [ 225 Ac]Ac 3+ . Noteworthy, in terms of the cytotoxic capacity, the dimeric [ 225 Ac]Ac-mcp-D-PSMA is superior to the monomeric [ 225 Ac]Ac-mcp-M-PSMA at all time points and even at low activity levels ( Figure 6).  Microscopic images of the crystal violet stained well plates underline the cytotoxic potency of both 225 Ac-radioconjugates and highlight the increased cell-killing capacity of the dimeric complex compared to the monomeric ligand (Figure 7). Nearly no colonies were detected after a one-hour exposure of the human prostate adenocarcinoma cells with 50 kBq/mL [ 225 Ac]Ac-mcp-D-PSMA. The treatment of LNCaP cells with 5 and 50 kBq/mL [ 225 Ac]Ac-mcp-D-PSMA, respectively, for 4 h also resulted in almost 100% inhibition of colony formation.

Small Animal Biodistribution Experiments
The obtained biodistribution data showed similar in vivo behavior for both studied radiotracers [ 225 Ac]Ac-mcp-M-PSMA and [ 225 Ac]Ac-mcp-D-PSMA in LNCaP tumorbearing mice (Figure 8; see Tables S2 and S3 in the SI for quantitative data). The biodistribution data using non-tumor-bearing SCID mice is presented in the SI ( Figure S16, Tables S4 and S5).
The organ distribution profile indicates a PSMA-mediated uptake in tissues with documented PSMA expression including kidneys and spleen besides the pathologic radiotracer accumulation in the tumor [36][37][38][39]. The highest activity accumulation was found in kidneys at 10 min (117.81 ± 19.86%ID/g vs. 66.01 ± 7.20%ID/g) and 60 min (67.84 ± 23.92%ID/g vs. 103.03 ± 24.68%ID/g) p.i., rapidly decreasing in time (1.35 ± 0.26%ID/g vs. 7.17 ± 1.96%ID/g at 24 h p.i.) for both compounds. [ 225 Ac]Ac-mcp-D-PSMA showed longer renal retention during the first 60 min p.i. compared to [ 225 Ac]Ac-mcp-M-PSMA, what was expected due to the difference in molar masses, but also through the binding potential in the proximal tubuli. The renal clearance of both radiotracers from the blood was very fast and the majority of the activity was eliminated from the blood circulation within 60 min post injection. The other excised organs revealed negligible radioactivity uptake except the spleen at early time points. The radioactivity uptake in the spleen 10 min p.i. differs between the monomeric (25.05 ± 10.42 %ID/g) and the dimeric (13.42 ± 6.11 %ID/g) radioligand. For both 225 Ac-labeled PSMA derivatives, the spleen uptake is substantially lower after 1 h p.i. with 2.90 ± 1.69 %ID/g and 4.57 ± 1.71 %ID/g, respectively. The observed difference in the spleen uptake 10 min p.i. originates from the individual blood clearance kinetics of both ligands as underlined by related differences in the kidney accumulation.
The tumor accumulation of [ 225 Ac]Ac-mcp-M-PSMA was higher at 10 and 60 min p.i. compared to [ 225 Ac]Ac-mcp-D-PSMA. The opposite results were observed at 24 h p.i., where [ 225 Ac]Ac-mcp-M-PSMA showed a lower accumulation compared to [ 225 Ac]Ac-mcp-D-PSMA (6.78 ± 0.45%ID/g versus 12.21 ± 4.31%ID/g). Tumor-to-background ratios were favorably high for both tested compounds and increased rapidly with time after administration. Tumor-to-blood ratios at 24 [40,41] as well as with recent data for other small molecules targeting PSMA [27,39,42,43]. differs between the monomeric (25.05 ± 10.42 %ID/g) and the dimeric (13.42 ± 6.11 %ID/g) radioligand. For both 225 Ac-labeled PSMA derivatives, the spleen uptake is substantially lower after 1 h p.i. with 2.90 ± 1.69 %ID/g and 4.57 ± 1.71 %ID/g, respectively. The observed difference in the spleen uptake 10 min p.i. originates from the individual blood clearance kinetics of both ligands as underlined by related differences in the kidney accumulation.

Chemistry
All chemicals were purchased from commercial suppliers and used without further purification. NMR spectra of all synthesized compounds were carried out on an Agilent DD2-400 MHz NMR or Agilent DD2-600 MHz NMR spectrometer with ProbeOne. All chemical shifts of 1 H and 13 C signals were reported in parts per million using TMS as internal standard at 25 • C and all spectra were calibrated using the respective solvent signal. Mass spectra were recorded on an Advion ExpressIon CMS or Xevo T-QS (Waters) using electrospray ionization. TLC analyses for reaction control were performed on Merck Silica Gel 60 F 254 TLC plates and visualized using UV light. Analytical HPLC was performed on VWR Hitachi with an Agilent C18 column (Agilent Zorbax 300SB-C18, 100 mm × 4.6 mm) and acetonitrile/water (0.1% TFA each) as mobile phase. Chromatographic separations were performed using automated flash column chromatography on Isolera Four (Biotage) using silica gel cartridges (SNAP KP-Sil; 10 g or 25 g) and reversed phase HPLC system Knauer Azura with Zorbax 300SB-C18 semi-preparative column and acetonitrile/water (0.1% TFA each) as mobile phase. Starting materials such as anhydrous solvents, dimethyl 4-hydroxypyridine-2,6-dicarboxylate (TCI Europe) and 4,13-diaza-18-crown-6 (Merck) were purchased and used as received. Methyl 6-(chloromethyl)picolinate was prepared according to the literature [44].

Cell Culture
The human prostate adenocarcinoma cell line LNCaP was obtained from the CLS Cell Lines Service (#300265, CLS Cell Lines Service GmbH, Eppelheim, Germany) and cultured in RPMI-1640 medium (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal calf serum (FCS, Merck KGaA, Darmstadt, Germany). Cells were grown in a humidified atmosphere of 95% air/5% CO 2 at 37 • C. For harvesting, counting and sub-cultivation, culture media was removed and cells were washed three times with PBS (Merck KGaA, Darmstadt, Germany). After the addition of 1% trypsin-EDTA (Merck KGaA, Darmstadt, Germany), cells were incubated for 5 min at 37 • C for detachment from the bottom of the culture flask. Then, culture media was added and cell suspension was transferred into conical centrifuge tubes and spun down for 5 min at 200 × g and room temperature. The cell pellet was resuspended in culture media and the cell number, as well as viability, was determined using a CASY cell counter (Roche Diagnostics GmbH, Penzberg, Germany) according to the manufacturer's protocol.

Protein Determination
Protein concentration was determined with the DC Protein Assay (Bio-Rad Laboratories GmbH, Feldkirchen, Germany) according to the manufacturer's microplate assay protocol using bovine serum albumin as a protein standard.

Gel Electrophoresis and Western Blot Analysis
Denaturing sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to a standard protocol [45]. For each gel, the PageRuler Plus Prestained Protein Ladder (Thermo Fisher Scientific, Waltham, MA, USA) was used as size standard.
LNCaP cell lysates were prepared by addition of 1 mL of radioimmunoprecipitation assay (RIPA) cell lysis buffer (G-Biosciences, St. Louis, MO, USA) with EDTA-free protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific, Waltham, MA, USA) and 25 U endonuclease (Thermo Fisher Scientific, Waltham, MA, USA). After resuspension and incubation on ice for at least 15 min, cell lysates were centrifuged at 14,000× g and 4 • C for 20 min. Equal protein amounts (10 µg) of the supernatants were separated on a 12% SDS polyacrylamide gel and transferred onto a polyvinylidene difluoride (PVDF) membrane (Merck KGaA, Darmstadt, Germany). Membranes were probed with primary anti-PSMA (D4S1F) and anti-β-actin (13E5) monoclonal rabbit antibodies (Cell Signaling Technology Europe B.V., Frankfurt am Main, Germany), respectively. As secondary antibodies, goat anti-rabbit horseradish peroxidase (HRP) conjugates (Merck KGaA, Darmstadt, Germany) were used. For detecting HRP on immunoblots, the WesternSure PREMIUM Chemiluminescent Substrate (LI-COR Biosciences-GmbH, Bad Homburg, Germany) in combination with the C-DiGit Blot Scanner and the Image Studio software (LI-COR Biosciences-GmbH, Bad Homburg, Germany) was used.

Saturation Binding Studies
Saturation binding studies were performed according to a published procedure [46]. LNCaP cells were plated in 48 well culture microplates (Greiner Bio-One GmbH, Frickenhausen, Germany) at a density of 3 × After the incubation time, the cells were washed three times with ice-cold PBS. Finally, cells were lysed by adding 1% SDS/0.1 M NaOH and shaking for 30 min at room temperature. To quantify the radioactivity in the cell extracts, an automatic gamma counter (Hidex Deutschland Vertrieb GmbH, Mainz, Germany) was used.

Cell Viability, Apoptosis and Cell Proliferation Assays
The treatment response on cellular proliferation was assessed using the CyQUANT direct cell proliferation assay (Thermo Fisher Scientific, Waltham, MA, USA). For this purpose, LNCaP cells were plated in black 96 well cell culture microplates (Greiner Bio-One GmbH, Frickenhausen, Germany) at a density of 2 × 10 4 cells/0.05 mL/well. After 24 h, five different activity concentrations (0.05, 0.5, 5, 50 and 50 kBq/mL) of the 225 Ac-labeled ligands [ 225 Ac]Ac-mcp-M-PSMA and [ 225 Ac]Ac-mcp-D-PSMA or the non-targeted activity control [ 225 Ac]Ac 3+ were diluted and added in triplicate so that the final volume was 0.1 mL/well. After 1 or 4 h of incubation, the cell culture media was removed and fresh media was added to each well. Cell numbers were subsequently determined 24 h later as per the manufacturer's protocol. The fluorescence intensity was measured with the Varioskan Flash microplate reader (Thermo Fisher Scientific, Waltham, MA, USA).
The exposure of phosphatidylserine on the extracellular side of cell membranes during the apoptotic process was measured using the RealTime-Glo™ Annexin V Apoptosis and Necrosis assay (Promega GmbH, Walldorf, Germany). To this end, LNCaP cells were plated in white 96 well cell culture microplates (Greiner Bio-One GmbH, Frickenhausen, Germany) at a density of 2 × 10 4 cells/0.05 mL/well. After 24 h, five different activity concentrations (0.05, 0.5, 5, 50 and 50 kBq/mL) of the 225 Ac-labeled ligands or the non-targeted activity control were diluted and added in triplicate so that the final volume was 0.1 mL/well. After 1 or 4 h of incubation, the cell culture media was removed and the assay reagent diluted in fresh media was added to each well. The fluorescence intensity and luminescence signals were measured for up to 24 h using the Varioskan Flash microplate reader (Thermo Fisher Scientific, Waltham, MA, USA).

Colony Formation Assay
LNCaP cells were plated in 6-well cell culture microplates (Greiner Bio-One GmbH, Frickenhausen, Germany) at a density of 6 × 10 3 cells/2 mL/well. After 24 h, four different activity concentrations (0.05, 0.5, 5 and 50 kBq/mL) of the 225 Ac-labeled ligands [ 225 Ac]Acmcp-M-PSMA and [ 225 Ac]Ac-mcp-D-PSMA were added in triplicate. After 1 or 4 h of incubation, the cell culture media was removed and fresh media was added to each well. Plates were then incubated at 37 • C for 8 days and re-fed every 3 to 4 days. Finally, the cell media was removed and colonies were stained with 1 mL of 0.5% crystal violet in 50% methanol for 30 min, after which the plates were rinsed three times with deionized water and subsequently air-dried. The plates were scanned with an Amersham Typhoon 5 Scanner (Cytiva Europe GmbH, Freiburg, Germany) and the colonies were counted using the Image-Quant TL software (Version 8.1, Cytiva Europe GmbH, Freiburg, Germany).

Animal Studies
The biodistribution studies with For biodistribution studies, the radiotracers were diluted with saline. The prepared radiotracers were applied retro-orbitally (r.o.) to the experimental animals [47] at a dose of 50 kBq per mouse corresponding to 10 pmol of the peptide. The tracer injection was carried out under 2% isoflurane anesthesia (FORANE, Abbott Laboratories, Abbott Park, IL, USA) to minimize animal suffering and to prevent animal motion. The mice were sacrificed by cervical dislocation 10 min, 60 min and 24 h post-injection and organs of interest (blood, spleen, pancreas, stomach, intestine, kidneys, liver, heart, lungs, muscle, bone and tumor) were collected. The samples were weighed and the activity was measured on an automatic gamma counter. The radiotracer accumulation was expressed as a percentage of injected dose per gram (%ID/g) of the corresponding organ.

Conclusions
Two chelators for mild condition radiolabeling with 225 Ac were developed, which enable a future use not only for the application of 225 Ac-radiolabeled peptides, but also 225 Ac-radiolabeled antibodies in targeted alpha therapy. In contrast to DOTA, both chelators revealed excellent labeling yields at low chelator concentrations after short labeling times at room temperature and appropriate long-term stabilities, as well. Additionally, the therapeutic efficiency in vitro and the applicability of two PSMA-targeting model radioconjugates on LNCaP tumor-bearing mice were shown in a proof of concept study on these triazole-containing click conjugates. The presented results provide bright prospects for various future applications for 225 Ac in targeted alpha therapy, such as site-specific modification of bio(macro)molecules of the treatment of various types of cancer and a general improvement of DOTA-containing conjugates as well.