Design and Synthesis of Piperazine-Based Compounds Conjugated to Humanized Ferritin as Delivery System of siRNA in Cancer Cells

Gene expression regulation by small interfering RNA (siRNA) holds promise in treating a wide range of diseases through selective gene silencing. However, successful clinical application of nucleic acid-based therapy requires novel delivery options. Herein, to achieve efficient delivery of negatively charged siRNA duplexes, the internal cavity of “humanized” chimeric Archaeal ferritin (HumAfFt) was specifically decorated with novel cationic piperazine-based compounds (PAs). By coupling these rigid-rod-like amines with thiol-reactive reagents, chemoselective conjugation was efficiently afforded on topologically selected cysteine residues properly located inside HumAfFt. The capability of PAs-HumAfFt to host and deliver siRNA molecules through human transferrin receptor (TfR1), overexpressed in many cancer cells, was explored. These systems allowed siRNA delivery into HeLa, HepG2, and MCF-7 cancer cells with improved silencing effect on glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene expression with respect to traditional transfection methodologies and provided a promising TfR1-targeting system for multifunctional siRNA delivery to therapeutic applications.


Crystal data, data collection and refinement parameters
The crystal and molecular structures of polyamines 5 and 9 have been established by single crystal X-ray diffraction experiments at low temperature [200(2) K (5) and 170 K (9)] on a Bruker D8 Venture PhotonII diffractometer (Chiesi Farmaceutici SpA is acknowledged for the support with the D8 Venture X-ray equipment) equipped with a CCD detector and using a CuKα (λ = 1.54178 Å) radiation. Collected frames were processed using the SAINT and SADABS software to obtain the intensities data file. 1 All reagents were purchased from Sigma-Aldrich and were used without further purification unless stated otherwise. Crystal data and experimental details for the X ray measurements and structure refinement are summarized in table S1. The two structures were solved by Direct Methods using the sir97 package 2 and refined by least-squares methods on the F o 2 by the SHELXL-2014 program 3 . All the atoms were treated with anisotropic thermal parameters with the exception of the hydrogen atoms placed at their calculated positions and refined "riding" with isotropic thermal parameters on their attached atoms [(C-H 0.98-0.99 Å, Uiso(H) = 1.2Ueq(C) and 1.5Ueq(C met )]. When possible, the water hydrogen atoms were located in electron density map. In the crystal lattice of compound 9, 5 water molecules were found; two of them (O1W e O4W) are disordered on two different positions with site occupancy factors of 0.3 and 0.7, respectively. In the case of O5W, the molecule is disordered over two positions each with occupancy of 0.5. One of the two positions could not be refined, and its contribution was taken into account using the program SQUEEZE [4]. The program calculated a solvent accessible volume of 32 Å 3 and 15 electron per unit cell. The weighting scheme used in the last cycle of the refinement was w = 1/ σ

Protein mass spectrometry
For 0.5 mL of each sample, 0.5 mL of acetonitrile and 1 µL of formic acid were added, and each solution obtained was injected into the Synapt G2 Si mass spectrometer (Waters) equipped with an electrospray ionization source, using these parameters: infusion flow rate 10 µL/min, capillary 3 kV, source temperature 120°C, sampling cone 80 V, source offset 80 V, desolvation temperature 150°C, nitrogen desolvation gas flow 600 L/h, nitrogen nebuliser gas flow 7 bar; the multicharged ions signals were acquired in positive polarity mode in the range 50-2000 m/z using leucine enkephalin as mass reference compound; deconvoluted data were obtained using MaxEnt1 software (Waters).

Dynamic Light Scattering
Dynamic Light Scattering (DLS) experiments were carried out using a Zetasizer Nano S (Malvern Instruments, Malvern, UK) equipped with a 4 mW He-Ne laser (633 nm). Briefly, the measurements were performed at 25°C, at an angle of 173° with respect to the incident beam. The average hydrodynamic diameters (Z-average diameter) of the scattering particles were calculated using peak intensity analyses. Samples were prepared at 1 mg/mL in 20 mM Hepes, 50 mM MgCl 2 , pH 7,4. All the traces for DLS experiments were analyzed with the software Origin 8.0 (Originlab Corporation, Northampton, MA, USA). Figure S29: Dynamic light scattering profiles of Pas-HumAfFt before (red) and after encapsulation with siRNA (black).

siRNA sequences
All siRNAs and FITC-siRNAs targeting GAPDH were synthesized by Sigma Aldrich and purified by desalting chromatography. FITC-siRNAs were designed with a fluorescein group at 5' in the sense sequence. The sequences are shown below:

TfR1 expression by Immunoblotting
Cells were lysed and analyzed by electrophoresis as already described in the Experimental Section, immunoblotting was performed with the following antibodies: anti-CD71 (VMA00037) from BioRad; anti-GAPDH (MAB-10578) from Immunological Sciences. Figure S30: a) Total protein lysates were extracted from HeLa, MCF-7 and HepG2 and analyzed by immunoblotting using specific antibodies as indicated. b) Histogram showing immunoblot's densitometric analysis (ImageJ software), bars indicate SD. Figure S31: Flow Cytometry analysis of FITC-siRNA-PAs-HumAfFt delivery. Cells were treated and data were acquired as described in the Material and Methods section. As shown in the first line of the plot, the gate for the final detection was set in the control sample (CTRL), naked siRNA and FITC-siRNA show no detectable uptake (NEG) and FITC-plain-HumAfFt (POS) confirm a 99% of internalization under these experimental conditions. FACS analysis of HeLa cells confirm the uptake of FITC-siRNA-PAs-HumAfFt. In the second line, one of two independent experiments is shown as representative and the percentage of FITC positive cells for each sample is indicated in each plot. For each sample 30,000 events gated on live cells were acquired.