Design, synthesis, and biological evaluation of a multifunctional neuropeptide-Y conjugate for selective nuclear delivery of radiolanthanides

Background Targeting G protein-coupled receptors on the surface of cancer cells with peptide ligands is a promising concept for the selective tumor delivery of therapeutically active cargos, including radiometals for targeted radionuclide therapy (TRT). Recently, the radiolanthanide terbium-161 (161Tb) gained significant interest for TRT application, since it decays with medium-energy β-radiation but also emits a significant amount of conversion and Auger electrons with short tissue penetration range. The therapeutic efficiency of radiometals emitting Auger electrons, like 161Tb, can therefore be highly boosted by an additional subcellular delivery into the nucleus, in order to facilitate maximum dose deposition to the DNA. In this study, we describe the design of a multifunctional, radiolabeled neuropeptide-Y (NPY) conjugate, to address radiolanthanides to the nucleus of cells naturally overexpressing the human Y1 receptor (hY1R). By using solid-phase peptide synthesis, the hY1R-preferring [F7,P34]-NPY was modified with a fatty acid, a cathepsin B-cleavable linker, followed by a nuclear localization sequence (NLS), and a DOTA chelator (compound pb12). In this proof-of-concept study, labeling was performed with either native terbium-159 (natTb), as surrogate for 161Tb, or with indium-111 (111In). Results [natTb]Tb-pb12 showed a preserved high binding affinity to endogenous hY1R on MCF-7 cells and was able to induce receptor activation and internalization similar to the hY1R-preferring [F7,P34]-NPY. Specific internalization of the 111In-labeled conjugate into MCF-7 cells was observed, and importantly, time-dependent nuclear uptake of 111In was demonstrated. Study of metabolic stability showed that the peptide is insufficiently stable in human plasma. This was confirmed by injection of [111In]In-pb12 in nude mice bearing MCF-7 xenograft which showed specific uptake only at very early time point. Conclusion The multifunctional NPY conjugate with a releasable DOTA-NLS unit represents a promising concept for enhanced TRT with Auger electron-emitting radiolanthanides. Our research is now focusing on improving the reported concept with respect to the poor plasmatic stability of this promising radiopeptide.


Background
Targeted radionuclide therapy (TRT) uses radioisotopes linked to a vector (peptides, small molecules, antibodies, and other targeting agents) in order to target tumor cells or their natural environment and deliver cytotoxic radiations. Encouraging results have been obtained in metastatic neuro-endocrine tumors with [ 177 Lu]Lu-DOTATATE, a somatostatin analog radiolabeled with lutetium-177 ( 177 Lu), that targets SST 2 R, the somatostatin receptor subtype 2 [1]. Similar success is envisioned with 177 Lu-labeled prostate specific membrane antigen (PSMA) inhibitors in patients with castration-resistant metastatic prostate cancer [2]. However, about 20% of these patients will not respond to therapy or rapidly relapse with regrowth of micrometastases [2]. It is therefore necessary to increase toxicity of TRT to tumor cells. Modeling studies have identified terbium-161 ( 161 Tb) as a potential alternative to 177 Lu to deliver higher cytotoxic radiation doses without kidney damage [3]. 161 Tb is a radiolanthanide that is chemically similar to 177 Lu but is expected to outperform 177 Lu due to its emission spectra rich in Auger electrons [4,5]. Recent preclinical findings confirm the higher tumor control of [ 161 Tb]Tb-PSMA compared with [ 177 Lu]Lu-PSMA [6,7]. Due to the short range of Auger electrons (few micrometers), the efficacy of such Augerelectron emitters can be maximized if located inside or close to the nucleus. A nuclear delivery of 161 Tb after cellular internalization could therefore significantly increase the dose deposited to the genomic DNA and enhance the cancer cell-killing effect of this radiolanthanide [8].
Our team works on improving the efficacy of TRT and extending applications to other targetable receptors that are overexpressed in tumors [9,10]. One promising target is the human Y 1 receptor (hY 1 R) which was found to be overexpressed in some tumors, such as estrogen receptor (ER)-positive breast cancer [11].
Here, we describe the combination of a full-length hY 1 R-targeting peptide with a releasable NLS-DOTA unit in order to deliver radiolanthanides for better efficacy.

Materials and methods
Synthesis of the novel pb12 NPY conjugate; its labeling with natural terbium ( 159 Tb), or its radio-labeling with indium-111 ( 111 In) The synthesis of the new NPY-conjugates is summarized below. Additional details are available in supporting information.
For the DOTA-NLS-NPY conjugate pb12, newly developed in our laboratories, the backbone sequence of [F 7 , P 34 ]-NPY was synthesized by robot-assisted SPPS using the Fmoc/tBu strategy [12]. As modification point, the natural Lys 4 in [F 7 ,P 34 ]-NPY was chosen. In previous studies, modification of Lys 4 with toxic agents or carboranes did not change the activity and selectivity profile of the peptide [13,14]. To facilitate side-specific modification in pb12, the N ε -amino group of Lys 4 was orthogonally protected by Dde. After the removal of the Dde group with hydrazine, the amino acid Dde-Lys(Fmoc)-OH was coupled as branching unit to the N ε -amino group of Lys 4 (Fig. 1). Subsequently, manual Fmoc deprotection was performed and the free N ε -amino group of the Lys branching unit could be extended with the amino acid sequence βAla-PAAKRVKLDGLFG by automated SPPS. GLFG thereby acts as cleavage site for the lysosomal protease cathepsin B [15], while PAAKRVKLD represents the NLS from the c-Myc protein [16]. β-Ala was introduced as short spacer. Next, the bifunctional chelating agent DOTA was manually coupled to the N-terminus of the β-alanine spacer with DIC/HOBt by using the protected building block DOTAtris(tBu)ester. Finally, the Dde protecting group at the N αamino group of the Lys branching unit was removed, and palmitic acid was manually coupled with DIC/HOBt to yield conjugate pb12. For the control peptide pb13, the peptide backbone [F 7 ,A 33 ,P 34 ,A 35 ]-NPY was used and modification at Lys 4 was performed as described for pb12. Replacement of the Arg 33 and Arg 35 residues in [F 7 ,P 34 ]-NPY with alanine was previously shown to delete the binding affinity of the peptide and is therefore suitable for the generation of a negative control peptide [17].
For first biological in vitro experiments, the DOTAcontaining NPY conjugates pb12 and pb13 were labeled with nat Tb or 111 In as surrogate for radiolanthanides as described below: The pb12 and pb13 conjugates (0.34 μmol) were dissolved in 500 μL of 0.4 M aqueous ammonium acetate solution (pH 5), and 100 μL of 0.01 M aqueous nat TbCl 3 solution were added. The mixture was incubated for 30 min at 37°C and 500 rpm in a dry block heater and subsequently cooled to room temperature. Removal of salts was accomplished by using Amicon® Ultra-4 MWCO 3000 centrifugal filter units (Merck) or PD MidiTrap G-25 desalting columns (GE Healthcare, Chicago, IL, USA) to obtain conjugates [ nat Tb]Tb-pb12 and [ nat Tb]Tb-pb13.
For radiolabeling with 111 In, 111 InCl 3 (Curium Netherlands B.V,~250 MBq) was incubated at room temperature with 50 μg of DOTA-peptide (pb12 or pb13) in 500 μL of 0.1 M acetate buffer (pH 5) for 1.5 h under gentle agitation. The raw product was diluted to 3 mL with acetate buffer and purified on a PD-10 size exclusion column (GE Healthcare) according to the manufacturer's instructions. Elution was performed with 3.5 mL of acetate buffer, and fractions of 0.5 mL were collected. The two fractions with the highest activities were pooled. Radiochemical purity was monitored by UV-radio RP-HPLC using a Phenomenex Luna C18 column (150 mm × 4.6 mm, 5 μm, 4 mL/min) with a linear gradient of 20-70% eluent B (ACN) in eluent A (0.1% (v/v) TFA in water) over 10 min and detection at λ = 220 nm.

Determination of lipophilicity
The lipophilicity of 111 In-labeled peptides was assessed by the water-octanol partition/distribution coefficient method as described in detail in supporting information.
Cell culture, hY 1 R, and cathepsin B expressions assessed using western blot and immunofluorescence on MCF-7 cells and cathepsin B activity of MCF-7 cells determined using ELISA assay are described in supporting information.
Receptor activation, receptor internalization, membrane integrity assay, PIXE experiment, and metabolic stability are also described in supporting information.

Saturation radioligand binding assay
Saturation binding assays were conducted on hY 1 R-expressing MCF-7 breast cancer cells following the methodology of Ginj et al. used for [ 111 In]In-DOTA-NLS-TOC [18]. The blocking agent used was the hY 1 R antagonist BIBP3226 (1 μM).

Determination of specific internalization of radiolabeled NPY conjugates
The cellular accumulation of the 111 In-labeled NPY conjugates was performed as described by Maschauer et al. [19], with only minor modifications (MCF-7 cells were seeded at a density of 1 million cells per well in 6-well plates and incubated overnight; 50 kBq of the respective 111 In-labeled NPY conjugates was used, and membranebound radioligand was removed with 20 mM acetate buffer (pH 5)). Each peptide was tested in three independent experiments.

Determination of cellular efflux of [ 111 In]In-pb12
For efflux experiments, MCF-7 cells were seeded at a density of 1 million cells per well in 6-well plates and cultured overnight. A total of 50 kBq of [ 111 In]In-pb12 (with or without pre-incubation of the efflux inhibitor probenecid at a concentration of 10 μM) was added to the medium and the cells were incubated (in triplicates) for 1 h at 37°C. Three minutes before the end of the incubation time, internalization was stopped on ice, and the supernatant was removed. Each well was washed with 1 mL of ice-cold PBS. The membrane-bound fraction was retrieved in 2 mL sodium acetate buffer (20 mM, pH 5) for 2 min; each well was rinsed a second time with 1 mL ice-cold PBS, and fresh culture medium was added. At each time point (0.5, 1, 2, 4, and 24 h), the efflux was stopped by collecting the medium and washing cells twice with ice-cold PBS. Finally, cells were treated with NaOH (1 M). The radioactivity of the collected culture medium supernatant, the PBS wash fractions, and the total internalized fraction were measured in a gamma counter. The experiment was performed three times independently.

Determination of nuclear uptake of 111 In
To determine the amount of radioactivity in the nucleus, we proceeded similar to the radioligand internalization experiments. As additional control, the cathepsin-B inhibitor CA-074 Me was pre-incubated at a concentration of 10 μM for 1 h in selected wells. Instead of adding NaOH at the last step, 500 μL of trypsin was added to each well and incubated for 3 min at 37°C. The content of each well was then centrifuged (5 min, 1800×g, 4°C). The supernatant was removed; 1 mL of hypotonic buffer was added and incubated for 15 min on ice. Subsequently, the samples were again centrifuged (5 min, 1800×g, 4°C), and the resulting supernatant corresponds to the cellular fraction, and the pellet corresponds to the nuclear fraction. Both fractions were separated and measured in a gamma counter. Finally, the percentage of radioactivity in the nucleus relative to the overall specifically internalized radioactivity or the total amount of used radioactivity was calculated. Experiments were independently performed at least two times in triplicates.

Results
Chemistry Peptide conjugates were prepared by a combination of automated and manual SPPS (Fig. 1). The sequences and the names of the generated compounds are listed in Table 1. All compounds were highly pure (> 95%) (see supplemental Table 1 and supplemental Figure 1).
Characterization of pb12 and [ nat Tb]Tb-pb12 compared with the control compounds pb13 and [ nat Tb]Tb-pb13 Cleavage of pb12 by cathepsin-B The pb12 peptide cleavage mediated by recombinant cathepsin B was investigated in vitro and monitored using UV-HPLC as first proof of concept. Cathepsin B metabolite formation increases over time and was significantly inhibited by the cathepsin B inhibitor CA074 (Fig. 2)

Affinity, receptor activation, and internalization
The specific receptor binding of the non-radioactive conjugates pb12 and [ nat Tb]Tb-pb12 (Fig. 3) and the control peptides pb13 and [ nat Tb]Tb-pb13 (Supplemental Figure  5) were investigated on MCF-7 cells along with the reference compound [F 7 ,P 34 ]-NPY. First, saturation binding curves were measured by plasmon-waveguide resonance (PWR) spectroscopy. The pb12 conjugate displayed only a slightly decreased affinity compared with [F 7 ,P 34 ]-NPY (Fig. 3a) with a K D of 3.1 ± 0.8 nM (Fig. 3b and Table 2). Labeling with nat Tb, however, led to a ten-fold decrease in binding affinity (K D 30 ± 3.7 nM for [ nat Tb]Tb-pb12, Fig.  3c) compared with the unlabeled pb12 molecule.
For the non-affine control peptides pb13 and [ nat Tb]Tb-pb13, no saturation binding up to a concentration of 1 μM could be observed (Supplemental Figure  5A and 5B).
As additional experiment, binding of pb12 and [ nat Tb]Tb-pb12 to non-hY 1 R-expressing HEK293 cells was investigated and some non-specific membrane interaction was measured (Supplemental Figure 5E and 5F).
Receptor activation of the hY 1 R and hY 2 R was investigated for the unlabeled DOTA-NLS-[F 7 ,P 34 ]-NPY The cathepsin B-cleavable linker is marked in red and the c-Myc nuclear localization sequence in blue. Palm palmitic acid, DOTA 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid conjugates pb12 and [ nat Tb]Tb-pb12 on COS-7 cells stably co-expressing the hY 1 R or hY 2 R and the chimeric G protein Gα Δ6qi4myr which switches the endogenous Gα icoupled signaling pathway of the activated NPY receptors to the Gα q -coupled pathway, which results in the intracellular generation of inositol phosphates (Fig. 4a). Unlabeled conjugate pb12 is a full agonist at the hY 1 R and displayed only a slightly lower potency compared with NPY and [F 7 ,P 34 ]-NPY with an EC 50 of 1.6 nM. Labeling with nat Tb in pb12 did not change the nanomolar activity of the conjugate at the hY 1 R (EC 50 1.2 nM). Considering hY 2 R activation, pb12 and [ nat Tb]Tb-pb12 were able to activate the hY 2 R only at very high concentration (10 μM). As regards the non-binding control [ nat Tb]Tb-pb13, it was nearly inactive for the hY 1 R and completely inactive at the hY 2 R.
[ nat Tb]Tb-pb12 and the unlabeled control peptide pb13 were also investigated in live-cell fluorescence microscopy studies for their capability to induce internalization of the hY 1 R and hY 2 R (Fig. 4b). In the unstimulated state, the fluorescence-tagged hY 1 R and hY 2 R (green) were predominantly localized in the plasma membrane. [ nat Tb]Tb-pb12 was able to stimulate internalization of the hY 1 R comparable to the endogenous ligand NPY, whereas no internalization of the hY 2 R was observed (Fig. 4b), thus confirming the retained hY 1 R selectivity of [ nat Tb]Tb-pb12 as determined in the receptor activation assay. The nonbinding control pb13 was not able to induce internalization of either hY 1 R or hY 2 R.
Membrane integrity after application of pb12, [ nat Tb]Tb-pb12 compared with the control peptide pb13 and nonexposed cells Given the identification of some non-specific association of pb12 and [ nat Tb]Tb-pb12 in experiments with non-hY 1 R-expressing HEK293 and potential impact in membrane organization and integrity, we also tested MCF-7 membrane integrity using a LDH release assay after application of the respective peptides (pb13 peptide and   Figure 5C and  Figure 5D) For easier comparison, Table 2 summarizes affinity values of all compounds evaluated in this work. Next, the hY 1 R-mediated internalization of the radiolabeled conjugates [ 111 In]In-pb12 and [ 111 In]In-pb13 into MCF-7 cells was analyzed (Fig. 4a). For conjugate [ 111 In]In-pb12, specific, time-dependent internalization into MCF-7 cells was observed with a maximum of 32.0 ± 9.7% of the total amount of used radiopeptide being internalized after 4 h (vs 11.7 ± 2.9% for the control peptide [ 111 In]In-pb13 at 4 h, differences were significant at 1 h, 2 h and 4 h).
In another set of experiments, the nuclear delivery of 111 In after internalization of [ 111 In]In-pb12 into MCF-7 cells was investigated (Fig. 5b). Incubation of MCF-7 cells with the multifunctional conjugate [ 111 In]In-pb12 resulted in a time-dependent nuclear uptake of 111 In. After incubation for 2 h, a maximum of 7.7 ± 2.4% of the total radioactivity was found to be located in the nucleus (Fig. 5b). This value would yield to 24.1% of the internalized fraction addressed to the nucleus. A significant reduction of nuclear uptake of 111 In at 3.8 ± 0.1% (p < 0.01) was observed when cells were pre-incubated with a cathepsin B inhibitor (Fig. 5b).
[ 111 In]In-pb12 was further evaluated regarding cellular efflux. A high and fast efflux of radioactivity was found for [ 111 In]In-pb12 (Fig. 6). Already after 30 min post internalization, 77 ± 5% of the total internalized radioactivity was detected in the medium outside the cells.  This further increased to 88 ± 3% extracellular radioactivity after 4 h post internalization. However, the observed efflux was not inhibited by the efflux inhibitor probenecid (Fig. 6) and the control conjugate [ 111 In]In-pb13 displayed a similar efflux profile (not shown).

Plasma stability of [ 111 In]In-pb12
Metabolic stability of [ 111 In]In-pb12 was evaluated in vitro in human plasma. Unfortunately, fast degradation of conjugate [ 111 In]In-pb12 was observed, with~20% of the initial radiopeptide left intact 1 h after incubation at 37°C.

SPECT/CT imaging of [ 111 In]In-pb12
Given the low plasmatic stability of [ 111 In]In-pb12, we focused on the first imaging time point at 12 minutes postinjection. [ 111 In]In-pb12 showed preferential elimination via the liver (Fig. 7a, b; 0.205 ± 0.07%IA/mm 3 ). Nonspecific signal was also quantified in the heart (0.121 ± Interestingly, low but specific uptake was noted in MCF-7 tumor (vs the blocked group, p = 0.001). Also, the tumor/ muscle ratio was highly significantly reduced in the cold pb12-blocked mice (Fig. 7c, p = 0.01).

Discussion
Targeted radionuclide therapy holds promise to deliver radiation to tumor cells over-expressing peptide receptors. In this work, we aimed to explore the feasibility of enhanced-TRT by using a subcellular active (radio) pharmaceutical able to deliver Auger-electron emitter like 161 Tb, in tumors over-expressing the hY 1 R. Several strategies have been used to develop NPY ligands for imaging and therapy (non-peptide ligands, antagonistictruncated peptides, or agonistic full-length peptides) with variable success [19,21,22]. The full-length hY 1 Rpreferring peptide analog [F 7 ,P 34 ]-NPY allows selective delivery of cargos such as radioisotopes [12,17,23]. We described here the multivalent construct pb12 based on the hY 1 R-preferring [F 7 ,P 34 ]-NPY sequence, which rapidly internalizes into acidic endosomes after agonistic stimulation of the hY 1 R, modified with a cathepsin Bcleavable linker able to release a nuclear localization sequence (NLS) linked to a DOTA chelator to convey lanthanides into the nucleus using importins [24]. For therapy perspective, a profound in vitro characterization is needed to fully understand the biological mechanisms which might impact the efficacy. Receptor binding studies demonstrated that the advanced modification of [F 7 , P 34 ]-NPY with a DOTA-NLS-GLFG unit and a fatty acid in pb12 did not influence the high hY 1 R affinity of the peptide. In addition, labeling with nat Tb results in a decrease in affinity from 3 to 30 nM but the activity and biological behavior of the nat Tb-labeled conjugate was hopefully maintained as demonstrated by PIXE experiments. pb12 was also radiolabeled with 111 In as "surrogate" of lanthanides, and we showed that, in this case, the high hY 1 R affinity was preserved. The difference in affinity between the 111 In-labeled version of pb12 and its nat Tb counterpart illustrates the necessity of carefully determining affinity of metallated compounds for imaging and therapy. Importantly, these nanomolar affinity values compete well with other full-length peptides [17] or non-peptide ligands [21] and are better than those reported for truncated peptides [22].
We then looked for the hydrophilicity of [ 111 In]In-pb12, and results showed that this compound is fairly hydrophilic. Addition of the lipophilic palmitic acid moiety might be responsible for the increased lipophilicity of our compound and lead to some non-specific binding of the peptide to the cell membrane. However, this interaction does not compromise cell membrane integrity as demonstrated by the LDH assay. Previous studies have shown that palmitoylation is an effective approach to enhance association of peptides with cell membrane and therefore increasing their proximity to the target, which is an important parameter for TRT with 161 Tb [25]. Exploring the additional radiobiological impact of 161 Tb on tumor cell membrane would be interesting [26].
To envision the use of Auger-electron emitters for enhanced TRT, the "shuttle" pb12 should be internalized. In living cells, the high affinity [ nat Tb]Tb-pb12 displayed similar internalization to the endogenous peptide NPY. When radiolabeled with 111 In, pb12 displayed a time dependent, specific internalization reaching 30% of total binding corresponding to the maximum of hY 1 R internalization capacity, reflecting the full agonist behavior of [ 111 In]In-pb12 [27]. Moreover, hY 1 R internalization is rapid and leads to membrane recycling of the receptor [28] illustrated by a significant rapid increase of [ 111 In]In-pb12 bound to the cell membrane which stayed constant over the course of 4 h. (Fig. 5).
After hY 1 R-mediated internalization followed by endocytosis [29], our compound was designed to be cleaved by cathepsin B to release the 111 In-DOTA-NLS unit, which is next addressed to the nucleus after binding to importins. We demonstrated that the recombinant cathepsin-B enzyme effectively cleaved the pb12 sequence (Fig. 2) and this cleavage is necessary to maximize the addressing of 111 In to the nucleus. In general, cytosol-located molecules with a molecular weight < 45 kDa are able to enter the nucleus via passive diffusion, while bigger molecules require a NLS to be transported by the nuclear pore complex [30]. Therefore, conjugation with NLS moieties was mostly performed for targeted antibodies such as trastuzumab [31]. The NLS conjugation to smaller molecules, as in this study, was also reported to be successful. A trifunctional [ 111 In]In-DOTA-NLS-TOC conjugate facilitated nuclear uptake of 111 In compared with [ 111 In]In-DOTA-TOC [18]. In our experiments, we cannot exclude a passive transport of 111 In (via interactions with membranes as demonstrated above for pb12 and [ nat Tb]Tb-pb12) into the nucleus, as the addition of the cathepsin B inhibitor did not fully inhibit the nuclear uptake of 111 In.
Another requirement for TRT radiopeptides is their low efflux rate. Indeed, a high efflux would result in suboptimal irradiation of tumor cells, particularly with shortrange particles, and higher off-target dose delivery.
[ 111 In]In-pb12 demonstrated high efflux which was not inhibited by the efflux inhibitor probenecid. However, using PWR, we demonstrate that compounds pb12 and [ nat Tb]Tb-pb12 interact with membrane lipids in a nonspecific manner (Supplemental Figure 5E and 5F). Taken together, the efflux of [ 111 In]In-pb12 we measured probably reflects the association of the peptide with the cell membrane rather than a high efflux, representing a significant improvement compared with truncated peptides [19].
To envision clinical application of our concept, the [ 111 In]In-pb12 conjugate needs to remain stable in vivo although full length peptides might be highly sensitive proteases. Despite its palmitic acid modification, known to increase resistance against proteolysis [32], [ 111 In]In-pb12 is rapidly metabolized potentially limiting its availability for tumor uptake. In a previous study, [F 7 , P 34 ]-NPY conjugated to methotrexate by the GFLGlinker exhibited much higher plasma stability [13]. Therefore, the reason for the fast degradation of [ 111 In]In-pb12 is most likely high susceptibility of the arginine/lysine-rich c-Myc NLS (PAAKRVKLD) to proteolysis and stabilization of this sequence is required for further application. We would like to point out that the metabolic stability regarding nucleus targeting studies was never reported in the literature, and this data, although disappointing, is of high interest for laboratories working in this research field. To give more insight in plasmatic stability, we intravenously injected [ 111 In]In-pb12 in mice bearing a MCF-7 xenograft. At an early time point, before major radiopeptide degradation, we noted a specific uptake in MCF-7 tumor, confirming the interest of [ 111 In]In-pb12 to target the hY 1 R. High accumulation was noted in the liver, consistent with the lipophilicity of the radiopeptide. Another limitation of this work was the low availability of 161 Tb. Therefore, we performed preliminary experiments with nat Tb which may be only partially representative of the behavior of the 161 Tb-counterpart as the radiolabeled fraction might be different. In the future, we plan to improve stability of the conjugate and to perform cellular studies comparing the radiotoxicity of different products, including 111 In, 90 Y, 177 Lu, and 161 Tb-labeled molecules (we are seeking European collaboration to receive 161 Tb).

Conclusion
In summary, we have developed a first multifunctional NPY analog allowing nuclear delivery of radiolanthanides. Our results suggest that the sequential mode of action of the construct works, i.e., hY 1 R binding with high affinity and selectivity, hY 1 R-mediated internalization, low efflux, cathepsin B-mediated cleavage of the multifunctional NPY analog, and finally nuclear delivery of the radiolanthanide. However, increasing metabolic stabilization is now needed for future application.