Functionalized TiO2 nanoparticles labelled with 225Ac for targeted alpha radionuclide therapy
- 316 Downloads
The 225Ac radioisotope exhibits very attractive nuclear properties for application in radionuclide therapy. Unfortunately, the major challenge for radioconjugates labelled with 225Ac is that traditional chelating moieties are unable to sequester the radioactive daughters in the bioconjugate which is critical to minimize toxicity to healthy, non-targeted tissues. In the present work, we propose to apply TiO2 nanoparticles (NPs) as carrier for 225Ac and its decay products. The surface of TiO2 nanoparticles with 25 nm diameter was modified with Substance P (5-11), a peptide fragment which targets NK1 receptors on the glioma cells, through the silan-PEG-NHS linker. Nanoparticles functionalized with Substance P (5-11) were synthesized with high yield in a two-step procedure, and the products were characterized by transmission electron microscopy (TEM), dynamic light scattering (DLS) and thermogravimetric analysis (TGA). The obtained results show that one TiO2-bioconjugate nanoparticle contains in average 80 peptide molecules on its surface. The synthesized TiO2-PEG-SP(5-11) conjugates were labelled with 225Ac by ion-exchange reaction on hydroxyl (OH) functional groups on the TiO2 surface. The labelled bioconjugates almost quantitatively retain 225Ac in phosphate-buffered saline (PBS), physiological salt and cerebrospinal fluid (CSF) for up to 10 days. The leaching of 221Fr, a first decay daughter of 225Ac, in an amount of 30% was observed only in CSF after 10 days. The synthesized 225Ac-TiO2-PEG-SP(5-11) has shown high cytotoxic effect in vitro in T98G glioma cells; therefore, it is a promising new radioconjugate for targeted radionuclide therapy of brain tumours.
KeywordsTargeted radionuclide therapy 225Ac Titanium dioxide nanoparticles Substance P Treatment cancer cells Nanomedicine
Despite of the predominant role of β − -particle emitters in radionuclide therapy trials (Iagaru et al. 2010), targeted radiotherapy based on α-particles is a promising alternative because they are highly cytotoxic agents, which deposit the whole of their energy within a few cell diameters (50–100 μm). Cell survival studies have shown that α-particle-induced killing is independent of oxygenation state (Wulbrand et al. 2013) or cell-cycle during irradiation and that as few as 1–3 tracks across the nucleus may result in cell death (Macklis et al. 1988; Sgouros et al. 2010). Additionally, the short range of α-particles corresponds to the dimensions of tumour micrometastases, allowing for localized irradiation of target cells with limited toxicity to surrounding healthy cells.
Recently, it has been found that in leukaemia, breast and brain cancer small subpopulations of tumour cells are able to self-renew and also to reconstitute the heterogeneous tumour-cell population (Al-Hajj et al. 2003; Singh et al. 2004). These, so called stem cancer cells, may be involved in the widespread metastatic dissemination of cancer (Wichal 2006). These findings have led to the suggestion that failure in tumour treatment may be associated with the failure to eradicate cancer stem cells (Al-Ejeh et al. 2011), especially that it has been found they are not sensitive to chemotherapy agents and to external and internal radiotherapy including β − -emitters (Al-Hajj et al. 2004). Given the properties outlined above, tumour stem cells are ideal targets for targeted α-particle therapy (TAT) (Sgouros and Song 2008).
There are hundreds of α-particle emitting radionuclides, but only a few have properties suitable for developing therapeutic radiopharmaceuticals: 212Bi (t1/2 = 60 min), 213Bi (t1/2 = 46 min), 211At (t1/2 = 7.2 h), 212Pb (t1/2 = 10.6 h), 227Th (t1/2 = 18.7 day), 223Ra (t1/2 = 11.4 day) and 225Ac (t1/2 = 9.9 day). However, most of the α-emitting radionuclides with optimal half-life for medical application, have one or more unstable daughter nuclides, which often emit α-particles as well. In principle such a cascade of α-particles can be very efficient in cancer treatment, but reliable delivery to the diseased site can be difficult to achieve. In the case of 225Ac, the most perspective radionuclide for TAT (Sattiraju et al. 2017; Zhu et al. 2017), the four consecutive decay daughters (i.e. 221Fr, 217At, 213Bi and 209Pb) receive recoil energies ranging from 105 to 160 keV as a direct result of the law of conservation of momentum. The recoil energy of firstly produced 221Fr is around 100 keV which is 1000 times greater than the energy of a chemical bond, implying that the recoiling daughter will always break free from the macrocyclic or acyclic chelating agents (Schwartz et al. 2011). Thus, produced daughter radionuclides are released from complexes, may accumulate in healthy organs and cause high toxicity which is a limiting factor in application of these radionuclides in targeted radiotherapy.
The new alternative strategy to sequester 225Ac and its daughters at targeted side is based on the use of nanocarriers. Such encapsulation is expected to prevent, or at least considerably reduce, the release of the recoiling daughters and by this to avoid undesired accumulation in healthy organs and tissues. Moreover, the use of nanocarriers may improve delivery of larger amounts of activity to the diseased site. There has been proposed several different types of nanostructures to retain 225Ac as well as its daughter radionuclides. Single and multi-layer liposomes have been evaluated; however, experimental studies revealed that their retention properties were poor with less than 10% of 213Bi retained (Sofou et al. 2004; Chang et al. 2008). Theoretical studies indicated that multi-walled polymerosomes could achieve around 80% of 213Bi retention, but their diameter has to be more than 800 nm (Thijssen et al. 2012). It would be very difficult for such large particles to extravasate/permeate/penetrate out of vasculature in vivo and also they can be clogged in microcapillary vessels. Woodward et al. (2011) developed 3–5 nm diameter monazite (LaPO4) nanoparticles as carriers for 225Ac, but these nanoparticles only partially retained daughters. More than ~ 50% of the 221Fr and 213Bi (from the decay of 221Fr) were released from the nanoparticle lattice. Addition of two layers of LaPO4 reduced leaching of 221Fr to 20% (Rojas et al. 2015). Similar nanocarrier based on lanthanide phosphate nanoparticles additionally coated with gadolinium and gold layers was developed by McLaughlin et al. (2013). Although multilayered nanoparticles exhibited higher retention properties, their synthesis becomes a multi-step and time-consuming procedure.
In the present work, for sequestration of 225Ac and its decay products, we propose to use titanium dioxide (TiO2) nanoparticles (NPs), which are inorganic cation exchangers exhibiting high affinity for 225Ac3+ and also 211Fr+ and 213Bi3+ cations. TiO2 NPs were conjugated with a fragment of substance P (5-11) for targeting NK1 receptors overexpressed in brain carcinomas. We examined the ability of TiO2 NPs to retain daughter products of 225Ac and investigated the cytotoxicity of the radiobioconjugate on human glioblastoma multiforme cells (T98G cells).
Materials and methods
Reagents, radionuclide and cell line
The following chemical reagents were used directly without further purification: TiO2 nanoparticles (anatase, ~ 25 nm, Sigma Aldrich), ethanol (96%, POCH), acetic acid (POCH), phosphate-buffered saline (PBS, Amresco), sodium chloride (POCH), anhydrous dimethylformamide (DMF) and triethylamine (TEA) of 99% purity were from Sigma Aldrich, substance P (5-11) (Bachem), human serum (Sigma Aldrich, stored at − 20 °C), methoxyl silane and N-hydroxysuccinimide functionalized polyethylene glycol (silane-PEG- NHS, 2000 kDa, Nanocs) and deionized water (Millipore). 225Ac was produced by radiochemical separation from 229Th source as described before (Apostolidis et al. 2005; Zielinska et al. 2007). The activity of 225Ac was quantified using a high-resolution γ-spectrometry when it was in secular equilibrium with its daughters, typically next day after sample collection.
T98G cells (human glioblastoma multiforme cells; CRL-1690; American Type Culture Collection) were cultured in DMEM/F12 K medium (Gibco) enriched with 10% foetal calf serum (FCS, Biological Industries, Israel) and streptomycin (100 μg/mL) and penicillin (100 IU/mL) (Sigma Aldrich). Cells were grown in humidified atmosphere with 5% CO2 at 37 °C. Prior to their in vitro use cells was detached using trypsin-EDTA (0.25%) (Biological Industries, Israel).
The shape and diameter of titanium dioxide nanoparticles was determined by scanning electron microscopy (SEM, Zeiss) and transmission electron microscopy (TEM, LEO 912AB). The hydrodynamic diameter and zeta potential (ζ) were measured by dynamic light scattering (DLS, Malvern). The presence of surface coating of nanoparticles by PEG and substance P was determined by thermogravimetric analysis (TGA, Q500, TA Instruments). About 5 mg of dried powder was placed in a TGA furnace and the analysis was conducted from room temperature to 900 °C at a heating rate of 10 °C per min in the presence of inert (nitrogen) atmosphere.
Gamma-ray spectra of radionuclides were measured using coaxial HPGe detector (GX 1080) for gamma spectroscopy with multichannel analyzer DSA-1000 (Canberra Packard) with Genie 2000 software; detection range 10–5000 keV. High-performance liquid chromatography (HPLC) was performed using the ELITE LaChrom (VWR-Hitachi) system with L-2310 pump coupled to L-2455 diode array detector, radiometric (GabiStar, Raytest) detector and L-2350 column oven. Hamilton column (10 μm, 10 × 250 mm) eluted at the flow rate of 2 mL/min was used for the analytical and semi-preparative chromatography. The gradient elution system consisted of deionized water (A) and acetonitrile (B), both solvents contained 0.1% (v/v) trifluoroacetic acid (TFA). The gradient conditions were as follows: 0 to 70% solvent B in 0–15 min, 70 to 95% solvent B in 15–20 min.
Conjugation of substance P (5-11) with PEG molecules
An equal molar amount of substance P (5-11) (SP(5-11)) was added to methoxyl silane and N-hydroxysuccinimide functionalized polyethylene glycol (silane-PEG-NHS) and dissolved in 250–300 μL DMF followed by 3 μL of TEA addition. Mixture was stirred for 72 h in an inert (nitrogen) gas atmosphere. After completion of the reaction (checked by HPLC), the solvent was removed under vacuum and the product (silan-PEG-SP(5-11)) was separated from unreacted substrates using above described HPLC system and conditions, and finally lyophilized.
Functionalization of TiO2 surface with silane-PEG-SP(5-11)
The TiO2 NPs were functionalized with earlier synthesized conjugate silane-PEG-SP(5-11) according to the procedure described in literature (Hermanson 2008). The acetic acid was added to 50 mL of ethanol (96%) to adjust pH to 4.5–5.5. Then about 5 mg of silan-PEG-SP(5-11) conjugate was dissolved and solution was stirred for 5 min. Next, 10 mg of TiO2 NPs was added and the mixture was sonicated for 10 min followed by 2 h stirring at room temperature (RT). At the end, obtained product was washed several times with ethanol to remove unreacted silane-PEG-SP(5-11) and dried at 100 °C for 30 min.
Labelling of TiO2-silane-PEG-SP(5-11) with 225Ac
225Ac3+ cations are adsorbed on the surface of TiO2 NPs through ion-exchange reaction on hydroxyl groups. For a typical labelling, ~ 1 mg of nanoparticles either bare TiO2 NPs or TiO2-silane-PEG-SP(5-11) NPs was suspended in 2 mL of 225AcCl3 (~ 100 kBq). The mixture was sonicated in an ultrasonic bath for 10 min and gently shaken on a circular stirrer for 1 h. After that time labelled NPs were centrifuged at 13000 rpm for 10 min, separated from supernatant and washed several times with deionized water. Finally, the product was suspended in 1 mL phosphate buffer (PB) and its activity measured on a γ-spectrometer to determine the labelling yield.
Stability studies of labelled TiO2-silane-PEG-SP(5-11) NPs
In order to study leaching of 225Ac and its decay daughters, a portion of labelled TiO2-silane-PEG-SP(5-11) NPs was dialyzed against 20 mL of 0.1 M PBS or saline (0.9% NaCl) for 10 days. Every day a 1-mL aliquot was withdrawn and the percentage of liberated activity from each daughter radionuclide was determined by its characteristic γ-ray peak. The stability study of radiolabelled nanoparticles in cerebrospinal fluid (CSF) was conducted using a centrifugal method. A 10 μL sample of labelled TiO2-silane-PEG-SP(5-11) NPs was added to 200 μL of CSF, vortexed and incubated at 37 °C for 10 days. Every day, the probes with added nanoparticles were centrifuged and 100 μL of CSF was tested for activity of 225Ac and its decay daughters.
The impact of TiO2 nanoparticles conjugated with substance P(5-11) fragment and labelled with 225Ac on metabolic activity and proliferation of T98G cells was measured with 3-(4,5-dimetyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT, Sigma Aldrich) assay. The assay was performed as described by Lankoff et al. (2012). In brief, 1 day before the experiment, T98G cells were seeded in 96-well plates (TPP) at a density of 103cells/well in 100 μL of culture medium. Cells were treated for 48, 72 and 96 h with increasing concentrations of non-radiolabelled TiO2 nanoparticles (0.0625–1 mg/mL) or the nanoparticles labelled with 225Ac (1.25–20 kBq/mL). At least three independent experiments in six replicate wells were conducted. After the described treatment, cell culture medium was removed and 100 μL of 3 mg/mL MTT solution was added to each well. After 3 h incubation at 37 °C the MTT solution was removed. Remaining insoluble formazan crystals were dissolved in 100 μL dimethylsulfoxide (DMSO, Sigma Aldrich) and absorbance of the solution was measured at 570 nm in plate reader spectrophotometer Infinite M200 (Tecan).
At least three independent experiments in six replicate wells were conducted for each toxicity point. Difference between samples and control were evaluated using GraphPad Prism 5.0 software (GraphPad Software Inc., USA). Toxicological data were evaluated by Kruskal-Wallis One Way Analysis of Variance on Ranks (ANOVA) followed by post hoc Dunnet’s method. Differences were considered statistically significant when the p value was less than < 0.05.
Results and discussion
Currently 225Ac is the most perspective radionuclide for targeted α-therapy in nuclear medicine. However, one significant factor limiting its use is the difficulty in sequestration of its decay products 221Fr and 213Bi at the targeted site. In our approach, we propose to use TiO2 nanoparticles as carriers for 225Ac and its decay daughters. Previous studies onTiO2 NPs properties indicated their high sorption affinity for 3+ metal cations (Metwally and Rizk 2014) and heavy alkali metal cations (Filipowicz et al. 2014) and facile functionalization and modification of their surface through hydroxyl groups (Bogdan et al. 2017). Therefore, these reports provided a compelling rationale to evaluate TiO2 NPs as potential carriers for stable retention of 225Ac and its decay daughters.
In order to synthesize a potential radiopharmaceutical exhibiting affinity to NK1 receptors overexpressed on glioma cells, attachment of an active biomolecule is required, in our case, an undecapeptide regulatory neuropeptide, substance P (SP). However, the native SP is metabolized in vivo due to enzymatic cleavages. The fragment SP(5-11) is the most abundant metabolite of native SP and it is still biologically active as the binding site with NK1 receptors is localized in the region of 7–11 amino acids. We have previously shown that this fragment radiolabelled with different radionuclides exhibit high affinity (IC50 = 38 nM) and interaction with NK1 receptors on T98G cells (Piotrowska et al. 2017; Lyczko et al. 2017). Additionally, fragment SP(5-11) contains l-glutamine in position 5 what allows coupling of silane-PEG-NHS via NHS-active ester to form bioconjugate that further is used to functionalize the TiO2 NPs surface.
On the basis of obtained TGA results the number of silane-PEG-SP(5-11) coating molecules per one TiO2 nanoparticle can be estimated. The calculation was performed under assumption that the nanoparticle is spherical with medium diameter of 20 nm, as measured by TEM, and that the density of material is 3.8 g cm−3. The obtained results show that in average 80 silane-PEG-SP(5-11) molecules are bound to the surface of one TiO2 nanoparticle. Taking into account that the ion-exchange capacity of TiO2 nanoparticles is about 1 meq g−1 (Filipowicz et al. 2014), the number of available OH groups on one TiO2 nanoparticle should exceed 3.8 × 104. This indicates that even after functionalization of the TiO2 nanoparticle surface with silane-PEG-SP(5-11) molecules, still the great majority of hydroxyl groups remain capable for binding of 225Ac3+ and formed in the decay process 221Fr+ and 213Bi3+ cations.
Hydrodynamic diameter and zeta (ζ) potential of bare and functionalized TiO2 NPs (pH = 7.4, PB buffer)
Zeta (ζ) potential (mV)
76.02 ± 5.54
98.78 ± 4.50
Bare TiO2 and TiO2-silane-PEG-SP(5-11) NPs were radiolabelled with 225Ac (100 kBq) at pH = 6.0 with a high yield of 99.8 ± 2.1% (n = 13) and 98.2 ± 1.3% (n = 10), respectively. The obtained results indicate that there is no difference in labelling yield between both nanoparticles and that surface functionalization did not have any negative impact on the radiolabelling process. It was expected, as previous calculations confirmed that the number of attached silane-PEG-SP(5-11) molecules is significantly much lower than the number of available OH functional groups responsible for ion-exchange approach with 225Ac. However, during our studies, it was found that due to low acidic character of the OH functional groups, the radiolabelling process is pH dependent and in acidic environment (pH = 3) the yield decreases even to only 27% for bare TiO2 NPs. Therefore, the optimal pH for radiolabelling reaction is 6.0 and it was used in further studies.
Retention of 225Ac and 221Fr on TiO2-silane-PEG-SP(5-11) NPs radiolabelled with 225Ac
221Fr retention (%)
PBS (pH = 7.4)
Physiological salt (0.9% NaCl)
Cerebrospinal fluid (CSF)
The obtained retention results might be interpreted on the base of recoil energy in the 225Ac → 221Fr reaction and subsequent α decays. Calculated, using the SRIM program (SRIM 2010), recoil range of 221Fr in water exceeds 84 nm, but in the material with density close to TiO2-anatase shows only about 44 nm. When the decay series is taken into consideration, we can assume that the radionuclides 217At, 213Bi and 209Pb (daughters of 221Fr) receive similar recoil energy. The escape of the radioactive isotopes from recoiled daughter and targeting agents allows them to freely migrate in the body, causing toxicity to healthy tissues and decreasing the therapeutic dose delivered to the diseased site. The renal toxicity induced by 213Bi is considered to be the major constraint to apply 225Ac in a large number of clinical trials (Jaggi et al. 2005; Schwartz et al. 2011). Based on the literature data, one would expect that higher amount of 221Fr and 213Bi nuclides would be ejected from a 20-nm-diameter TiO2 nanoparticle than observed in our studies. The observed smaller release of the daughter radionuclide 221Fr (< 5% in PBS and saline solution) suggests that a fraction of the recoil energy takes part in translation of the whole NP rather than being entirely dissipated through atomic displacements resulting, as it occurs in the bulk, in shorter recoil range. Additionally, in the case of TiO2 which is an effective cation exchanger, we observe subsequent reloading of escaped recoils 221Fr+ and 213Bi3+ as it contains hydroxyl surface groups with high affinity for these cations (Filipowicz et al. 2014; Bogdan et al. 2017; Perekhozheva et al. 1985; Bilewicz et al. 1991). This process can also explain slightly lower stability of radiolabelled NPs in CSF, as in this medium hydroxyl surface groups of NPs are blocked by the proteins and peptides present in CSF what prevents rebinding of released daughter radionuclides.
The cytotoxicity of 225Ac-silane-TiO2-PEG-SP(5-11) is very high in comparison to other radiopharmaceuticals including conjugates radiolabelled with α-emitters like 213Bi or 211At (Gadbois et al. 1996; Dziawer et al. 2017), however, it is comparable with other radiobioconjugates of 223Ra and 225Ac (Borchardt et al. 2003; Piotrowska et al. 2017). The decay processes of 225Ac and 223Ra include four α and two β− emissions to a stable 209Bi and 207Pb daughters. Because a very large amount of energy (~ 28 MeV) is released during these decay processes, much smaller amount of radionuclides activity is required to produce the desired effect and 225Ac and 223Ra radionuclides demonstrate extreme cytotoxicity. Comparison of obtained cytotoxicity results of 225Ac-silane-TiO2-PEG-SP(5–11) NPs with literature data for radiopharmaceuticals labelled with 213Bi (Friesen et al. 2007) demonstrate that several logs less 225Ac radioactivity is required to reach similar LD50, value, because of the longer half-life and multiple α-emissions.
We have shown that TiO2 NPs functionalized with substance P (5-11) can be used to deliver and retain 225Ac and its daughter radioisotopes at a target site; thereby reducing the absorbed dose to non-targeted organs. The TiO2 NPs successfully retain a large fraction of the 225Ac decay products without compromising the tumoricidal properties of the α radiation. Because of the reloading of the 225Ac decay products the retention efficiencies of TiO2 NPs are comparable to the radiolabelled LnPO4-Au core shell nanostructures; however, the synthesis process is much simpler and less time-consuming. Furthermore, it has been shown that 225Ac-TiO2-silane-PEG-SP(5-11) exhibits satisfactory stability in CSF what is very important as it is intended for direct intratumoral or post-resection application. The intravenous injection of the 225Ac-TiO2-silane-PEG-SP(5-11) is excluded due to the relatively large size and high hydrophilicity which prevents crossing the blood-brain barrier.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- Borchardt PE, Yuan RR, Miederer M, McDevitt MR, Scheinberg DA (2003) Targeted actinium-225 in vivo generators for therapy of ovarian cancer. Cancer Res 63:5084–5090Google Scholar
- Hermanson GT (2008) Bioconjugate techniques, 3rd. edn. Academic Press, AmsterdamGoogle Scholar
- Perekhozheva TN, Sharygin LM, Albantova GP (1985) Cation-exchange properties of an adsorbent based on hydrated titanium-dioxide. Inorg Mat 21:364–367Google Scholar
- Sofou S, Thomas JL, Lin H, McDevitt MR, Scheinberg DA (2004) Sgouros G (2004) engineered liposomes for potential alpha particle therapy of metastatic cancer. J Nucl Med 45:253Google Scholar
- SRIM. Particle Interactions with Matter. http://www.srim.org/2010
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.