Targeting of radioactive platinum-bisphosphonate anticancer drugs to bone of high metabolic activity

Platinum-based chemotherapeutics exhibit excellent antitumor properties. However, these drugs cause severe side effects including toxicity, drug resistance, and lack of tumor selectivity. Tumor-targeted drug delivery has demonstrated great potential to overcome these drawbacks. Herein, we aimed to design radioactive bisphosphonate-functionalized platinum (195mPt-BP) complexes to confirm preferential accumulation of these Pt-based drugs in metabolically active bone. In vitro NMR studies revealed that release of Pt from Pt BP complexes increased with decreasing pH. Upon systemic administration to mice, Pt-BP exhibited a 4.5-fold higher affinity to bone compared to platinum complexes lacking the bone-seeking bisphosphonate moiety. These Pt-BP complexes formed less Pt-DNA adducts compared to bisphosphonate-free platinum complexes, indicating that in vivo release of Pt from Pt-BP complexes proceeded relatively slow. Subsequently, radioactive 195mPt-BP complexes were synthesized using 195mPt(NO3)2(en) as precursor and injected intravenously into mice. Specific accumulation of 195mPt-BP was observed at skeletal sites with high metabolic activity using micro-SPECT/CT imaging. Furthermore, laser ablation-ICP-MS imaging of proximal tibia sections confirmed that 195mPt BP co-localized with calcium in the trabeculae of mice tibia.


Results
Pt-BP favors bone-specific delivery of Pt with reduced Pt-DNA adduct formation. We comparatively evaluated the bone-seeking properties of Pt-BP and its bisphosphonate-free precursor Pt(NO 3 ) 2 (en) by quantitatively analyzing Pt biodistribution using Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) in different tissues (e.g. tibia, femur, humerus, spine, blood, heart, lung, kidney, liver, and spleen) as shown in Fig. 1A. To this end, we intravenously injected sterile saline solutions containing Pt-BP or Pt(NO 3 ) 2 (en) with identical Pt concentrations into the tail vein of mice, which were euthanized 24 h after injection. The dissected tissues were weighed and digested in 65% nitric acid for subsequent ICP-MS analysis of Pt content. Figure 1B shows the Pt concentrations in different tissues 24 h after injection, normalized for specific tissue weight. Pt-BP clearly showed bone-seeking properties as evidenced by higher amounts of Pt accumulation (~3 ng Pt/mg tissue) in hard tissue (bone) compared to BP-free control Pt(NO 3 ) 2 (en) complex (~1 ng Pt/mg tissue).
The efficacy of Pt-based drugs for cancer treatment relates to the formation of Pt-DNA adducts resulting in DNA damage, which hinder mitotic processes and halt cell division. Drawbacks of the formation of Pt-DNA adducts include the high rate of DNA damage in non-target cells or healthy tissue 36 . We determined the extent of Pt-DNA adduct formation by extracting genomic DNA from different tissues and quantifying the amount of Pt using high-resolution ICP-MS. Figure 1D shows the relative Pt uptake within specific tissues leading to Pt-DNA adduct formation 24 h after injection. Pt-BP formed a low amount of Pt-DNA adducts in all tissues (<0.5%) except for the kidney (2.8%) and spleen (1.4%), which confirms the relatively low extent of DNA damage in non-targeted tissue. In contrast, the bisphosphonate-free Pt complexes showed a much higher extent of Pt-DNA adduct formation, especially in the kidneys (4.8%) and spleen (9.8%).
195m Pt-BP accumulates specifically in metabolically active bone. Radioactive 195m Pt-BP was synthesized using 195m Pt(NO 3 ) 2 (en) as the precursor for 195m Pt. To compare the biodistribution of 195m Pt-BP with the precursor 195m Pt(NO 3 ) 2 (en), a dose of 11.2 ± 0.4 MBq 195m Pt was administered intravenously via the tail vein in C57BL/6N mice. Although the administered Pt dose was below the Pt toxicity dose of 6 mg/kg for mice 37 , injection of 195m Pt(NO 3 ) 2 (en) evoked a significant body weight loss within the first 3 days post injection (p.i.), leading to euthanization on reaching a humane end point. On the other hand, no behavioral or body changes were observed for 195m Pt-BP treated mice.
Micro-SPECT/CT images were acquired 1 h -7 days post intravenous administration of 195m Pt-BP ( Fig. 2A). The micro-SPECT/CT scans clearly demonstrated effective targeting of 195m Pt-BP to growth plates in long bones, whereas 195m Pt(NO 3 ) 2 (en) showed specific accumulation in soft tissues (Fig. 2B). Generally, growth plates are metabolically active regions in bones of young mice since they are responsible for longitudinal bone growth 38 . After excision of tissues, we quantified the uptake of radioactive 195m Pt-BP complexes by gamma counting (Fig. 3A). Bone-specific uptake of 195m Pt-BP was highest in the femur (3.1 ± 0.37%ID/g) followed by the tibia (2.86 ± 0.46%ID/g) and humerus (2.53 ± 0.33%ID/g), whereas 19m Pt(NO 3 ) 2 (en) showed limited uptake in study in C57Bl/6N mice. Pt-BP or Pt(NO 3 ) 2 (en) compounds were administered intravenously in mice (2.5 mM Pt concentration) followed by sacrifice after 24 h. Hard tissues (tibia, femur, humerus, spine), soft tissues (heart, lungs, kidney, liver, spleen) and blood were collected. Collected tissues were subjected to nitric acid digestion and genomic DNA extraction for Pt quantification using ICP-MS and high-resolution ICP-MS, respectively. (B) Pt concentration in different tissues was normalized for specific tissue weight 24 h after injection. Data from 5 mice per group are represented as ng Pt/mg tissue ± SD. (C) Percentage of Pt uptake in different tissues 24 h after injection. Data from 5 mice per group are represented as % I.D./g ± SD of specific tissue type. (D) Percentage of Pt-DNA adduct formation relative to total Pt uptake in specific tissue as quantified by High Resolution-ICP-MS. Data from 4-5 mice per group are presented. **P < 0.01; ***P < 0.001; ****P < 0.0001 as determined by two-way ANOVA with a Bonferroni (multiple comparisons) post-hoc test.
Next, we analyzed the micro-SPECT/CT images quantitatively to determine the uptake of radioactive 195m Pt-BP complexes. 195m Pt-BP showed rapid and strong uptake in bone (1.8 ± 0.25%I.D/g) at 1 h p.i., which remained constant (1.0 ± 0.12%I.D/g) from 24 h until day 7. 195m Pt-BP uptake was significantly lower in soft tissue than in hard tissue (0.93 ± 0.27%I.D/g at 1 h and 0.18 ± 0.03%I.D/g at 24 h). In contrast, 195m Pt(NO 3 ) 2 (en) showed equal uptake in both hard (1.5 ± 0.1%I.D/g) and soft tissue (1.43 ± 0.21%I.D/g) at 1 h p.i., and the biodistribution remained unchanged until day 3 as shown in Fig. 3B. Subsequently, the quantification of the hot spots in bone represents a twofold increased uptake of 195m Pt for 195m Pt-BP compared to 195m Pt(NO 3 ) 2 (en) for all time points (Fig. 3C). The hard-to-soft tissue uptake ratio of 195m Pt-BP increased periodically from 3.3 at 1 h, 6 at 24 h and 6.7 after 72 h. On the other hand, the hard-to-soft tissue uptake ratio of 195m Pt(NO 3 ) 2 (en) remained almost constant at 1 at 1 h, 1.2 at 24 h and 1.2 at 72 h (Fig. 3D).
Spatial co-localization of pt with ca in bone. The spatial distribution of Pt as determined using Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) is depicted in Fig. 4. The calcium and phosphorus distribution (Supplement Fig. S1) clearly conformed to the structure of trabecular bone as observed in the dark-colored features in the microscopic image (Fig. 4A). The overlay of Ca and Pt distinctly differentiates co-localization of Pt with Ca (yellow) from Pt surrounding the trabecular structure (green). 195m Pt-BP showed predominant Pt uptake along the trabecular bone and inner region of the cortical bone, as reflected by an www.nature.com/scientificreports www.nature.com/scientificreports/ increased density of Pt co-localization with calcium. Conversely, 195m Pt(NO 3 ) 2 (en) showed the highest Pt density in the soft tissue surrounding the cortical bone.
To quantitatively examine the co-localization of Pt with Ca, we calculated the percentage of Pt co-localized with calcium. This calculation (Fig. 4B) shows a co-localization of Pt with calcium of 73.5% for 195m Pt-BP and only 37.3% for 195m Pt(NO 3 ) 2 (en). Relative to the extent of co-localization of Pt with Ca as observed for 195m Pt-BP, co-localization of 195m Pt(NO 3 ) 2 (en) was only 27% (Fig. 4C). The total amount of Pt not co-localized with Ca refers to Pt in soft tissue of the proximal tibia where 10% and 20% of Pt was detected in soft tissue for 195m Pt(NO 3 ) 2 (en) and 195m Pt-BP, respectively. 195m Pt-BP showed almost a fourfold increased accumulation of Pt in bone compared to 195m Pt(NO 3 ) 2 (en) as shown in Fig. 4C. These results confirm again that 195m Pt-BP binds specifically to the bone and effectively penetrates the trabecular structure of mouse tibiae.

Discussion
Recent advances in imaging technologies allow early detection of malignant tissues directly in bone marrow 39,40 or detection of indirect tumor activity related to bone remodeling 41,42 . Such early assessment of developing bone metastases would enable patient-specific bone-targeted treatment. However, treatment options for patients with bone metastases are scarce and palliative rather than curative 4,5 . Consequently, the design of novel theranostic compounds which detect and treat bone metastases in cancer patients would overcome this unmet need.
Our present work focuses on the design of radioactive and bone-seeking chemotherapeutics based on 195m Pt. To this end, we synthesized a Pt-BP compound comprising two 195m Pt moieties and a bone-seeking bisphosphonate group to target the compound specifically to metabolically active bone. This synthesis of Pt-BP was modified compared to previously reported synthesis strategies to facilitate introduction of radioactive 195m Pt atoms by using Pt(NO 3 ) 2 (en) as Pt precursor 29,31 . 195m Pt(NO 3 ) 2 (en) was obtained from Na 2 [ 195m PtCl 4 ], which is the Pt precursor used for synthesis of radioactive cisplatin, thereby allowing for future up-scaling and clinical translation of 195m Pt-BP 23 . Pt hard-to-soft tissue uptake ratio excluding bladder uptake at 1 h. Data from 4-5 mice per group are presented. **P < 0.01; ****P < 0.0001 as determined by two-way ANOVA with a Bonferroni (multiple comparisons) post-hoc test. For the ratios, paired t-test was used to determine the differences among the two groups where **P < 0.01 was considered as significantly different. (2020) 10:5889 | https://doi.org/10.1038/s41598-020-62039-2 www.nature.com/scientificreports www.nature.com/scientificreports/ The release of therapeutically active Pt species from Pt-BP is vital to impart the desired chemotherapeutic effects on cancer cells. Previous in vitro studies on Pt-BP dissociation upon release from solid hydroxyapatite nanoparticles surface confirmed that Pt-BP does not simply split into BP and Pt 31 . Pt-BP most likely releases one Pt moiety in the form of PtCl 2 (en) or related species, whereas the other Pt is sequestered with the BP molecule due to the coordination with the free amino group as present in Pt-BP. This observation is further supported by similar results obtained upon releasing Pt-BP from silica xerogels 29 .
To understand Pt release behavior in vivo, we investigated Pt-BP dissociation in near-physiological (pH 7.4) and acidic conditions (pH 5.25) (Supplement Fig. S2). Pt-BP underwent hydrolysis into a mononuclear Pt-n-BP derivative at both pH values, characterized by faster release of Pt at pH 5.25 than at pH 7.4. At physiological pH, Pt-BP forms new species with asymmetric bisphosphonate groups, which results most likely from displacement of one coordinated oxygen atom of the bridging bisphosphonate ligand by either a chloride (120 mM concentration) or bisphosphonate-bound amino group 43 . Besides, a signal belonging to an as yet unknown species appeared at 15.32 ppm at pH 5.25. As expected, the formation of unsymmetrical species having the amino group coordinated to the Pt atom is not favored at acidic pH, since this moiety is protonated and hence not available for coordination to the metal.
To investigate the bone-seeking ability of Pt-BP, the bisphosphonate-free compound Pt(NO 3 ) 2 (en) was selected as control, since this compound was also used as precursor for the synthesis of Pt-BP. The primary Pt-BP and 195m Pt(NO 3 ) 2 (en) treated mice. ****P < 0.0001, two-tailed student's test. (C) Total amount of Pt co-localized with Ca (in hard tissue, Pt ≈ Ca) and Pt not co-localized with Ca (in soft tissue, Pt ≠ Ca). Data are presented from 10 sections from three tibia per group. ****P < 0.0001 as determined by two-way ANOVA with a Bonferroni (multiple comparisons) post-hoc test. (2020) 10:5889 | https://doi.org/10.1038/s41598-020-62039-2 www.nature.com/scientificreports www.nature.com/scientificreports/ reason to select Pt(en)(NO 3 ) 2 as control was that, as demonstrated by NMR on the cold Pt-BP complex, 195m Pt-BP releases 195m Pt(en)(OH 2 ) 2 ] 2+ , that undergoes anation reactions in physiological-like conditions. The same fate occurs also for 195m Pt(en)(NO 3 ) 2 . In order to understand if the 195m Pt-species released by hydrolysis of 195m Pt-BP were distributed in tissues different from bone, we used the Pt-species released. Moreover, since 195m Pt(en) (NO 3 ) 2 was also the precursor used by the supplier (NRG) of radioactive 195m Pt in the synthesis of 195m Pt-BP, our choice allowed to reduce costs and time for the synthesis of radioactive 195m Pt-BP compounds. Generally, we hypothesized that the Pt(NO 3 ) 2 (en) control compound lacked bone-seeking properties due to the absence of a mineral-binding moiety. Our in vivo biodistribution studies showed that Pt-BP compounds delivered Pt specifically to the bone. Although Pt uptake was specifically observed in hard tissues for Pt-BP compounds, this Pt did not form many Pt-DNA adducts after 24 h. These results corroborate the slow release of Pt from Pt-BP compounds observed in vitro at physiological pH. In contrast, Pt(NO 3 ) 2 (en) showed increased Pt-DNA adduct formation in kidneys and the spleen, which are well-known off-target organs for Pt-based drugs 44 . On the contrary, the passive process of Pt-DNA adduct formation upon slow release of Pt from Pt-BP might be beneficial by diminishing undesired off-target effects.
The biodistribution of radioactive 195m Pt-BP and 195m Pt(NO 3 ) 2 (en) compounds revealed a pattern similar to our previous observations using cold Pt-based therapeutics. The extended follow-up period of 7 days confirmed specific accumulation of 195m Pt-BP within bone only. In contrast, 195m Pt(NO 3 ) 2 (en) specifically accumulated in kidney, liver, and spleen, thereby affecting the excretory organs and resulting in weight losses of more than 20%, which was set as humane end point. Previously, several studies have shown that cisplatin-treated mice (maximum tolerated dose of 6 mg/Kg) undergo rapid weight loss (~10%) followed by recovery within a week 37 . S3), the half-life of 195m Pt (t 1/2 = 4 days) 23,24 enabled quantitative, high resolution SPECT imaging. We further demonstrated 195m Pt-BP specificity towards skeletal sites by an increasing hard-to-soft tissue contrast from 1 h until day 3. Moreover, LA-ICP-MS imaging confirmed Pt co-localization with calcium within the trabecular region of the tibia of mice. In contrast, LA-ICP-MS imaging of the Pt distribution in 195m Pt(NO 3 ) 2 (en)-treated mice showed preferential uptake in soft tissues (bone marrow and blood vessels) surrounding the trabecular bone. Thus, LA-ICP-MS imaging indicates that 195m Pt(NO 3 ) 2 (en) uptake in the skeletal system is mediated by a passive mechanism via blood perfusion within the vasculature of the skeletal system.
Based on our preclinical results, 195m Pt-BP offers several advantages as a theranostic agent for treatment of bone metastases. First, BP moieties effectively target 195m Pt to bone sites with high metabolic activity. This opens up new opportunities for bone tumor-target delivery of Pt-BP compounds, since bone metastases are typically characterized by very high metabolic activity. Second, off-target effects caused by 195m Pt are decreased by this bone-specific targeting, rapid excretion within 24 h, and the slow release of Pt from Pt-BP. Third, 195m Pt-BP compounds remain stable for at least 24 h after reaching the targeted site, thereby improving the imaging quality for qualitative and quantitative diagnosis. Indeed, our results show slow release of Pt from Pt-BP in vivo upon attachment to hydroxyapatite mineral, which is in agreement with the in vitro stability of Pt-BP as observed previously using NMR. However, the main goal of systemic delivery of 195m Pt-BP is to deliver this 195m Pt (a rich source of Auger electrons) specifically to bone metastases. Considering the vigorous bone remodeling and increased bone metabolic activity in bone metastatic lesions, we assume that release of 195m Pt at such sites will be accelerated due to enhanced bone remodeling involving osteoclastic resorption at reduced pH. Generally, Pt-BP compounds dissociate at acidic pH, which means that release of Pt can be accelerated upon tumor-induced acidification. This hypothesis should be investigated in future studies in tumor-bearing mice. Nevertheless, further investigations are required for the translation of 195m Pt-BP towards clinical application.
The main goal of the study was successfully achieved, i.e. synthesis of radioactive 195m Pt-BP compounds which accumulated preferentially in metabolically active bone. Nevertheless, several shortcomings of the current study should be highlighted. First, the 195m Pt precursor 195m Pt(NO 3 ) 2 (en) was used as the control compound since other radioactive 195m Pt derivatives of platinum-based drugs were not available. Second, additional therapeutic efficacy beyond the cytostatic effect of Pt may arise from Auger electrons and gamma rays emitted by 195m Pt, which we will investigate in a preclinical bone metastasis model in tumor-bearing mice by comparing therapeutic effects of radioactive and non-radioactive Pt-BP compounds 46 . Finally, Pt toxicity is a major concern for clinical applications of Pt-based drugs 47 . Nevertheless, we did not observe any Pt-based toxicity in 195m Pt-BP-treated mice over a period of 7 days. However, further long-term investigation is required to confirm adverse effect on bone biomarkers. Moreover, we addressed potential toxicity of Pt released from Pt-BP compounds in a zebrafish model, since this model provides a 500-fold higher sensitivity for early assessment of toxic effects caused by Pt-based chemotherapeutics 48,49 . As such, the zebrafish model is a highly valuable model for initial ototoxicity screening. Visual inspection of kidneys based on size and shape revealed apparent shrinkage for cisplatin controls but not for Pt-BP. In addition, no visible phenotypic effects of Pt-BP were observed in zebrafish embryos up to a concentration of 100 µM (Supplement Fig. S4). Pt-BP revealed a dose-dependent ototoxic effect on zebrafish embryos 48 h post treatment, as confirmed by staining for and quantification of lateral line neuromast hair cells (Supplement Fig. S5).

Scientific RepoRtS |
(2020) 10:5889 | https://doi.org/10.1038/s41598-020-62039-2 www.nature.com/scientificreports www.nature.com/scientificreports/ In summary, we described the synthesis and biodistribution of a bone-seeking 195m Pt-BP compound for active targeting of 195m Pt to bone sites with high metabolic activity. It should be stressed that radioactive 195m Pt-BP compounds with bone-seeking properties have not been reported before. Their synthesis is highly complex from a logistical perspective since access to radioactive 195m Pt of high specific activity is very limited worldwide. In addition, biodistribution of radioactive or inactive Pt-BP compounds has never been confirmed in vivo. Our results show that for the first time using micro-SPECT imaging that radioactive 195m Pt-BP compounds accumulate in metabolically active bone. Finally, this accumulation was confirmed for the first time in hard tissues using elemental mapping by means of Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). We envision that 195m Pt-BP as a bone-seeking theranostic agent could meet the current clinical need to "see and treat" bone metastases in cancer patients and reduce off-target side effects of Pt-based drugs.

Methods
Study design. The objective of the study was to evaluate Pt-BP as potential therapeutic (Pt) or theranostic agent ( 195m Pt) to selectively deliver Pt or 195m Pt to metabolic active region of bone. All experiments were performed with five animals per group to reach statistical significance. Mice were randomized into different groups before administration of Pt/ 195m Pt compounds. Blinding was applied in housing. In addition, biotechnicians responsible for Pt compounds administration and monitoring of animal welfare was blinded. Image acquisition in vivo and ex vivo was performed in a non-blinded manner. Data analysis was also performed in a non-blinded manner.
Characterization via elemental analysis, Electrospray Ionisation-Mass Spectrometry (ESI-MS), and spectroscopic features (Supplement Figs. S6 and 7) shows consistency of the Pt-BP product with the one obtained using Pt(OSO 3 )(OH 2 )(en) as a Pt precursor 29,30 . The 1 H-NMR spectrum of Pt-BP (Supplement Fig. S7) shows two broad singlets at 5.93 and 5.43 ppm assigned to the aminic protons of coordinated ethylenediamine 29 . The acidic conditions decelerated the exchange of the aminic protons with the deuterium of the solvent, allowing detection of the aminic protons in aqueous solution. The doublet centered at 3.34 ppm is assigned to the protons of the methylene group of the bisphosphonate. The 31 P-NMR spectrum (Supplement Fig. S7B) of the Pt-BP in D 2 O shows a singlet with unresolved Pt satellites positioned at 37.60 ppm. This singlet can be assigned to the two phosphorus atoms of the phosphonic groups, which are magnetically equivalent and shifted at lower field (Δδ = 21 ppm) with respect to the free 2-amino-1-hydroxyethane-1,1-diyl-bisphosphonic acid (AHBP) ligand at the same pH 29 . preparation of 195m pt(no 3 ) 2 (en). The 195m Pt was produced as previously reported 24 . The specific activity of 195m Pt(NO 3 ) 2 (en) was 48,5 MBq/mg Pt. The radionuclide purity of 195m Pt(NO 3 ) 2 (en) as received from NRG (Petten, The Netherlands) is reported in Supplement Table S1. pt-Bp pH stability assessed by nMR. 31 P NMR spectra were recorded on a Bruker Avance III 700 MHz instrument 29 . Standard pulse sequences were used for 31 P{1H} (121.5 MHz) spectra. Chemical shifts ( 31 P) were referenced to external H 3 PO 4 (85% w/w; 0 ppm). A Crison Micro-pH meter Model 2002, equipped with Crison microcombination electrodes (5-and 3-mm diameter) and calibrated with Crison standard buffer solution at pH 4.01, 7.02, and 10.00, was used for pH measurements. The pH readings from the pH meter for D 2 O solutions are indicated as pD values and are uncorrected for the effect of deuterium on glass electrodes. The stability of Pt-BP in buffered solutions at 37 °C was assessed by 31 P NMR spectroscopy as previously reported 53 . Pt-BP (∼4 mg) was dissolved in 0.8 mL of D 2 O containing (i) 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) (HEPES) buffer (pD = 7.4) and 120 mM NaCl or (ii) 50 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer (pD = 5.25) and 120 mM NaCl. The resulting two solutions were transferred into NMR tubes and maintained at 37 °C. 31 P NMR spectra were recorded over a period of 15 days. The relative concentrations of the individual species in solution were deduced from integration of the 31 P signals.
preparation of radioactive pt-Bp complex. The synthesis procedure was followed as described for the cold platinum-bisphosphonate complex provided in the supplementary material. The 195m Pt(NO 3 ) 2 (en) solution was received from NRG (Petten, the Netherlands). The pH of 195m Pt(NO 3 ) 2 (en) solution was neutralized to pH 7 using 1M NaOH and reconstituted in sterile saline solution. Characterization via elemental analysis and Electrospray Ionisation-Mass Spectrometry (ESI-MS) was not possible due to insufficient amount of residual 195m Pt-BP. The amount of 195m Pt (and thus 195m Pt-BP) was limited due to limited supply by NRG (Petten, the Netherlands). www.nature.com/scientificreports www.nature.com/scientificreports/ Netherlands. For both the experiment, 10 male C57Bl/6N mice (Charles River), with an average weight of ~25 g and an age of approximately 6-8 weeks were housed in filter-topped cages (5 mice per cage) under non-sterile standard conditions provided with standard animal food and water ad libitum. The mice were allowed to adapt to laboratory conditions for 1 week before experimental use. The humane end point were defined as follows, body weight gain of> 20% within 10 days, no water or food intake, weight loss of >15% within 1-2 days or> 20% of normal body weight of that animal, serious circulatory or respiratory problems, behavioural changes (hyperactivity, passive, auto-mutilation), and morbidities such as impaired mobility or lethargy 52 .

In vivo biodistribution of cold Pt-species and quantification of Pt accumulation. The Pt-BP and
Pt(NO 3 ) 2 (en) complexes were administered intravenously in the tail vein of C57BL/6N mice (n = 5 per each platinum complex). Sterile saline solution (0.9% NaCl) was used to dissolve platinum complexes. The concentration of injected Pt-BP or Pt(NO 3 ) 2 (en) solutions was 2.5 mM platinum, calculated based on the maximum tolerated dose for cisplatin of 6 mg per kg body weight of the mice 37 . The mice were euthanized with CO 2 24 h after injection, after which blood (approximately 600 mg per mouse), liver, spleen, kidneys, heart, lungs, and bones (femur, humerus, tibia, and spine) were harvested. Approximately half of the tissues were prepared for inductively coupled plasma-mass spectrometry (ICP-MS) analysis by digestion in 65% (v/v) nitric acid at 75 °C for approximately three days until the tissues were digested completely. Each tissue was cut into three samples and the samples were weighed prior to digestion in nitric acid. The digested solutions were diluted with up to 6 ml of Milli-Q water to obtain 2% (v/v) nitric acid in order to measure platinum concentrations by ICP-MS (X series I, Thermo Electron Corporation). The detection limit for determining platinum concentration with ICP-MS was 1 ppb. The standard solutions were prepared from 1000 mg·1 −1 platinum ICP standard Certipur (1.70341.0100, Merck) ranging from 1 ppb to 2500 ppb. The measured isotopes for platinum were 194 Pt,195 Pt,196 Pt,and 198 Pt. Furthermore, scandium ( 45 Sc) was added as an internal standard, which was prepared with 1000 mg·1 −1 scandium standard Certipur (1.19513.0100, Merck) to correct for matrix effects and long-term fluctuations of the measurement signal. The platinum concentrations in the various tissues were calculated relative to 1 mg of the specific tissue and represented as ng Pt/mg tissue . Three replicates of Pt-BP and Pt(NO 3  The DNA was digested with DNase I Solution (Thermofisher, USA) and filtered using 0.2 µm acrodisc GHP before analysis for Pt content using high resolution-ICP-MS (Element2, Thermo Finnigan). The detection limit for determining the platinum concentration was 2 ng/l. The standard solutions were prepared from 1000 mg·1 −1 platinum ICP standard Certipur (1.70341.0100, Merck) ranging from 5 ppt to 10 ppb. The selected isotopes for platinum were 194 Pt and 195 Pt. Furthermore, thallium (Tl) was added as an internal standard and determined at a molecular mass of 205 Da, which was prepared using a 1000 mg·1 −1 thallium standard (CGTL-1, Inorganic Ventures) to correct for matrix effects and long-term fluctuations of the analytical signal. The Pt-DNA adduct concentration is represented as percentage of Pt involved in adduct formation to the total amount of Pt accumulated in the specific tissue 52 . In vivo biodistribution of radioactive Pt-species and micro-SPECT quantification of 195m pt biodistribution. The 195m Pt-BP and 195m Pt(NO 3 ) 2 (en) complexes were reconstituted in sterile saline solution (0.9% NaCl) and administered intravenously in the tail vein of C57BL/6N mice (n = 5 per each platinum complex). All mice received an intravenous injection with a dose of 11.2 ± 0.4 MBq 195m Pt. Immediately after injection and 1, 3, and 7 days after injection, images were acquired with U-SPECT-II/CT(MILabs) as reported previously 54,55 . Mice were scanned under general anesthesia (isoflurane/O 2 ) for 15 to 60 minutes using the 1.0-mm diameter pinhole mouse high sensitivity collimator tube, followed by a CT scan (spatial resolution 160 mm, 65 kV, 615 mA) for anatomical reference. Scans were reconstructed with MILabs reconstruction software using an ordered-subset expectation maximization algorithm, with a voxel size of 0.4 mm. SPECT/CT scans were analyzed and maximum intensity projections were created using the Inveon Research Workplace software (IRW, version 4.1). A 3D volume of interest was drawn using CT threshold to differentiate soft tissue from skeletal tissue and uptake was quantified as the percentage injected dose per gram (%ID/g), assuming a tissue density of 1 g/cm 3 . The hot spot in the skeletal tissue region of interest (ROI) was chosen with the location of the edge of the ROI contour representing 75% of maximum intensity. The mice from 195m Pt(NO 3 ) 2 (en) group were euthanized with CO 2 after 3 days due to excessive loss of body weight, whereas mice from treated with 195m Pt-BP were sacrificed at the end of experiments. After euthanizing the animals, blood (approximately 600 mg per mouse), liver, spleen, kidneys, heart, lungs, and bones (femur, humerus, tibia, and spine) were harvested and analyzed with a gamma counter (1470 Wizard, Perkin Elmer).
Laser ablation icp MS imaging of 195m pt biodistribution in mice tibia. The experimental procedure was carried out as reported previously 35 . The mice tibias were incubated in neutral buffered formaldehyde for 36 h. Later, tibias were dehydrated in ascending grades of 70% to 100% ethanol and embedded in poly(methyl methacrylate) (pMMA) resin, freshly prepared by mixing 600 mL of methyl methacrylate monomer (Acros Organics BVBA, Geel, Belgium), 60 ml dibutyl phthalate (Merck KGaA, Darmstadt, Germany) and 1.25 g perkadox ® 16 (Aldrich, Netherlands). The polymerization was followed by cutting serial horizontal sections (perpendicular (2020) 10:5889 | https://doi.org/10.1038/s41598-020-62039-2 www.nature.com/scientificreports www.nature.com/scientificreports/ to the tibia sample) of 5 µm thickness within the trabecular region of interest using an RM2155 microtome with a TC 65 blade (Leica Microsystems GmbH, Wetzlar, Germany). Microscopic images were obtained using an inverted fluorescence/bright field microscope (BZ-9000, Keyence Deutschland GmbH, Neu-Isenburg, Germany). For image recording and processing, the software BZ-II Viewer and BZ-II Analyzer were used, respectively. To calculate the percentage of platinum co-localized with calcium, the background of the image and all pixels without hard bone tissue were excluded from the calculations. A pixel is considered to be background if its calcium intensity is below 15%. The platinum concentrations of the remaining pixels were then added up and divided by the sum of the entire platinum image to yield the percentage of platinum co-localized with calcium 35 . Zebrafish care and handling. The zebrafish (Danio rerio) strains were kept under standard conditions (28 °C in E3 buffer) until 48 hours post fertilization (hpf) as described previously 56 . All animal experiments were conducted at larval stages before the point of independent feeding and were in agreement with the animal protection law (Tierschutzgesetz). phenotypic assessment of drug-treated embryos. 48 hpf wildtype embryos were separated into 6 groups (n = 20 embryos/group), i.e., control (untreated) and embryos treated with 5 µM, 10 µM, 50 µM, and 100 µM Pt-BP as well as 30 µM cisplatin. Both types of compounds were added externally to the medium and the embryos were incubated at 33 °C for the duration of the experiment. The embryos were assessed each day for phenotypic toxicity until 2 days post treatment (48 hpt) and imaged live in an Olympus MVX10 microscope. ototoxicity assay. The lateral line neuromast hair cells in embryos were stained with a vital dye to analyse the loss of hair cells after Pt-BP treatment. Cisplatin was used as a control, as it was known to affect hair cells in humans. The fluorescent dye 2-[4-(dimethylamino)styryl]-N-ethylpyridinium iodide (DASPEI) [Molecular Probes, Eugene, OR] was used to stain hair cells in the neuromast as described previously 57 . Larvae were incubated in embryo medium containing 0.005% DASPEI for 15 min, anesthetized in Tricaine MS222 (10 lg/ml) for 5 min, rinsed once in fresh embryo medium, and imaged using Vertebrate automated screening technology (VAST, Union Biometrica), combined with high-speed and super-resolution Zeiss cell observer spinning disk confocal system. The quantification of number of neuromast hair cells in the treated embryos was done manually with confocal microscopy (Zeiss LSM780).

Statistical analysis.
All results except ratios are depicted as mean ± standard deviation. The statistical analyses were performed using GraphPad Prism (version 6.0) software. Two-way analysis of variance (ANOVA) with a Bonferroni (multiple comparisons) post-hoc test was used to determine the differences among the two groups. For ratios, paired t-test was used to determine the differences among the two groups. For the ototoxicity assay, one-way analysis of variance (ANOVA) was performed followed by the Dunnett's method for multiple comparisons.