DOTA Glycodendrimers as Cu(II) Complexing Agents and Their Dynamic Interaction Characteristics toward Liposomes

Copper (Cu)(II) ions, mainly an excess amount, play a negative role in the course of several diseases, like cancers, neurodegenerative diseases, and the so-called Wilson disease. On the contrary, Cu(II) ions are also capable of improving anticancer drug efficiency. For this reason, it is of great interest to study the interacting ability of Cu(II)–nanodrug and Cu(II)–nanocarrier complexes with cell membranes for their potential use as nanotherapeutics. In this study, the complex interaction between 1,4,7,10-tetraazacyclododecan-N,N′,N′′,N′′′-tetraacetic acid (DOTA)-functionalized poly(propyleneimine) (PPI) glycodendrimers and Cu(II) ions and/or neutral and anionic lipid membrane models using different liposomes is described. These interactions were investigated via dynamic light scattering (DLS), ζ-potential (ZP), electron paramagnetic resonance (EPR), fluorescence anisotropy, and cryogenic transmission electron microscopy (cryo-TEM). Structural and dynamic information about the PPI glycodendrimer and its Cu(II) complexes toward liposomes was obtained via EPR. At the binding site Cu–N2O2 coordination prevails, while at the external interface, this coordination partially weakens due to competitive dendrimer–liposome interactions, with only small liposome structural perturbation. Fluorescence anisotropy was used to evaluate the membrane fluidity of both the hydrophobic and hydrophilic parts of the lipid bilayer, while DLS and ZP allowed us to determine the distribution profile of the nanoparticle (PPI glycodendrimer and liposomes) size and surface charge, respectively. From this multitechnique approach, it is deduced that DOTA-PPI glycodendrimers selectively extract Cu(II) ions from the bioenvironment, while these complexes interact with the liposome surface, preferentially with even more negatively charged liposomes. However, these complexes are not able to cross the cell membrane model and poorly perturb the membrane structure, showing their potential for biomedical use.


Devices
NMR measurements were carried out on a Bruker DRX 500 NMR spectrometer operating at 500.13 MHz for 1 H NMR using D 2 O or DMSO-d 6 as a solvent. Sodium 3-(trimethylsilyl)-3,3,2,2-tetradeuteropropionate was added for internal calibration (δ ( 1 H) = 0 ppm). The signal assignments were performed by a combination of 1D and 2D NMR experiments using the standard pulse sequences provided by Bruker.
Time of Flight Mass Spectrometry (MALDI-TOF MS) investigations were performed on a Bruker Autoflex Speed TOF/TOF in reflector or linear modes, respectively, and positive polarity by pulsed smart beam laser (modified Nd:YAG laser). The ion acceleration voltage was set to 20 k. For the sample preparation, the substances were mixed with 2,5-dihydroxy benzoic acid as matrix, both dissolved in millipore water.
Zeta-potential (ZP) and dynamic light scattering (DLS) measurements were performed using a Malvern Zetasizer Nano ZS system (model ZEN3600, Worcestershire, UK) equipped with a standard 633 nm laser. Perkin Elmer Optima 7000 DV with a spectral range of 160-900 nm and a resolution < 9 pm were used for carrying out ICP-OES. 2% nitric acid solution was used as solvent.
Cryo-TEM images were acquired using Libra 120 microscope (Carl Zeiss Microscopy GmbH, Oberkochen, Germany) at an acceleration voltage of 120 kV.
EPR experiments were performed by using an EMX-Bruker spectrometer working at X-band (9.5 GHz).

DLS and ZP experiments
The size of particles and the polydispersity index (PDI) were measured using the DLS technique. The equilibration time was 60 sec., while the measurement angle was 173° and the number of measurements was 3 per 25 runs (run duration 5 sec). The peak size gives the zaverage. All the experiments were performed at 25°C, and at different equilibration times, t = 0, 1h, and 24 h.
The surface charge was determined by using the Zetasizer Nano-ZS with a combination of electrophoresis and laser Doppler velocimetry. Electrophoretic mobility of particles was measured by using capillary plastic cells with a copper electrode, covered with gold, in order to apply an electric field.
The data evaluation was carried out by using Malvern Software from the Helmholtz- LEC was used at a different concentration with respect to DMPC and DMPC/DMPG 3% since the lipid concentrations were initially selected in order to optimize the structural characteristics of liposomes for performing DLS measurements. However, the molar ratios between lipids, dendrimers and Cu(II) ions were maintained constant in all experiments.
The volume vs. size plots of dendrimer alone, liposomes alone, and the samples obtained by adding dendrimers to liposomes in the absence and presence of Cu(II) at the different concentrations allowed us to evaluate the particle sizes to structurally characterize the systems, together with PDI and ZP. As a reference, the dendrimer alone was also investigated in aqueous solution ( Figure S3).

EPR experiments
The suspension was placed in EPR tubes (1 mm internal diameter) and subjected to different For both the EPR studies using Cu(II) and CAT12, the EPR spectra of samples consisting of the ternary system constituted by the paramagnetic species + dendrimer + liposome were compared to the simple binary systems constituted by the paramagnetic species + liposome or dendrimer. In fact DPH can be between the chains but also flat in the core of the bilayer, with a not defined localization; (2) TMA-DPH is well-known to act as an anchor on membrane surface; the molar moiety limits the insertion but the DPH part is still in the upper part of the hydrophobic section. In the present study, DPH was used to detect surface interactions between glycodendrimer and liposomes surface.

Cryo-TEM images
Samples for acquiring Cryo-TEM images were prepared by dropping 2 µL of liposome solution (1.3 mM LEC + 0.08 mM DOTA in PBS 10 mM; 1.5 mM DMPC or DMPG + 0.09 mM DOTA in PBS 10 mM) on copper grids coated with holey carbon foil (so-called Lacey type). A piece of filter paper was used to remove the excess water; the sample was then rapidly frozen in liquid ethane at -178°C. The blotting with the filter paper and plunging into liquid ethane was done in a Leica GP device (Leica Microsystems GmbH, Wetzlar, Germany). All images were recorded in bright field at -172°C. The diameter (60-80 particles) and membrane thickness (10-20 particles) of the empty and loaded-liposome were determined from cryo-TEM images by using TEM Image Processing Software. Figure S3 shows   Figure S4 shows the pH-dependence of the zeta potential of G4-DOTA-Mal. Z-average are shown, but the samples are highly polydisperse, the D h data is not reliable.   Concerning the lecithin liposomes, the post-loading addition of G4-DOTA-Mal had no effect on the liposome size. The addition of glycodendrimers is reflected in immediate increase in PDI followed by restoration of initial conditions after 24 hours (PDI lower than 0.2, resulting in homogenous systems). Additionally, the interaction of the G4-DOTA-Mal -Cu(II) complex and LEC resulted in a further increase in the positively charged surface and the system appeared homogenous at t=0h, t=1h and t=24h.

DLS and ZP study providing size and charge of dendrimers and/or liposomes in the absence and presence of Cu(II)
The interaction between the G4-DOTA-Mal -Cu(II) complex and DMPC liposomes makes the surface charge positive. Unlike what was observed for lecithin liposomes, for DMPC systems the Cu(II) concentration was less affecting the polydispersity of the particles.
For DMPG (DMPC/DMPG 3 %) liposomes, as shown in Table S2, the addition of G4-DOTA-Mal and the further addition of Cu(II) to the dendrimer increases the positive charge of the DMPG surface. Concerning the polydispersity, the presence of G4-DOTA-Mal and the complex is reflected in the increase of PDI also after 24 h.

S13
The volume in function of the size distribution in Figure S5 shows a peak around 100 nm highlighting that no aggregation processes occurred. The hydrodynamic diameter obtained from DLS refers to particles diffusion within a fluid; surface structure, concentration as well as the type of ions in the medium affect the diffusion speed of particles and, consequently, the hydrodynamic diameter. Therefore, a well-defined liposome structure forms both in the absence and in the presence of the glycodendrimer and Cu(II). S14

Long-term stability of liposomes at different temperatures by DLS
Further DLS measurements over time were performed to verify the stability of the liposomes ( Figures S6-S8). Figure S6 shows the difference between LEC liposomes obtained using sonication or extrusion treatments. The sonication treatment ( Figure S6) leads to unstable and non-homogenous liposomes; the system is indeed characterized by a high polydispersity index (PDI greater than 0.2; results not shown). Conversely, the extrusion treatment produces homogeneous, monolamellar and long-term stable liposomes.    Furthermore, passing from 25 °C to 37 °C the average size progressively increased for liposomes stored at room temperature as well as 4°C, due to the faster dynamics at the higher temperature.
However, the size of the liposomes was completely maintained after 30 days from samples preparation. An important conclusion is that the used protocol, associated with the extrusion treatment, leads to the production of unilamellar liposomes that are stable and uniform over time, showing PDI values lower than 0.2 and having reproducible diameters. In addition to this, the liposomes show to be resistant to temperature changes.
Furthermore, the observations confirm that the size of the vesicles are not concentration and temperature dependent for the different types of lipids.  Figure 4 of the manuscript):

Further comments on the variations of the EPR spectra intensity, relative % of the interacting component, τ for the Free component, S of the interacting component, <A> for the Interacting and Free components, for the three different liposomes in the absence and presence of G4-DOTA-Mal (shown in
The intensity increases in the presence of liposomes compared to pure CAT12 solutions, confirming that CAT12 inserts into the liposomes (LEC < DMPC < DMPG) increasing its solubility.
The lower solubility found for LEC compared to other liposomes was accompanied by a

Fluorescence anisotropy study
Fluorescence anisotropy provides information on the main localization of molecules by means of specific fluorescent probes. In our case we evaluated the membrane fluidity and glycodendrimer localization in presence of the three liposomes: LEC, DMPC and DMPG, using two fluorescent probes: (1) DPH is a popular probe of membrane interiors, useful for evaluating the possible interaction between glycodendrimer and liposomes in the hydrophobic membrane region; (2) TMA-DPH is well-known to act as an anchor on membrane surface; in the present study, it was used to detect surface interactions between glycodendrimer and liposomes. The change in the fluorescence anisotropy is shown in Figure   S9 as a function of the dendrimer concentration towards the three liposomes, LEC, DMPC and DMPG. This is partially explainable, when considering the surface charge (Table S2) and perturbation properties of liposomses, induced by G4-DOTA-Mal (EPR study), only a combination of both parameters may induce hydrophilic interaction of less extent with G4-DOTA-Mal.
Using the hydrophobic probe DPH, negligible interactions are observed in all samples ( Figure   S9). In the case of DMPG at low dendrimer concentration, a small interaction occurs with the hydrophobic part of the membrane probably related to the structural deformation hypothesized on the basis of the EPR results. As the concentration increases, hydrophilic interactions prevail.