*2.4. General Methods*

The apparent hydrodynamic diameter and ζ-potential of nanoparticles in water were estimated at room temperature (approximately 25 ◦C) using a Photocor Compact-Z instrument (Russia) at λ = 659 nm and θ = 90◦(10 scans, each one for 15 s).

IR spectroscopy was recorded on a Shimadzu IRSpirit at 4700 to 350 cm−<sup>1</sup> (10 mg of sample without any specified sample preparation).

UV spectra were recorded using a Mettler UV5 spectrophotometer.

Differential thermal analysis (DTA) and thermogravimetric analysis (TGA) were performed on the SDT Q600 using a heating rate of 5 ◦C/min in the temperature range from 40 ◦C to 600 ◦C.

X-ray diffraction analysis was carried out on a Dron-7 X-ray diffractometer, using a 2θ angle interval from 7◦ to 40◦ with scanning step Δ2θ = 0.02◦ and exposure of 7 s per point. Cu K<sup>α</sup> radiation (Ni filter) was used, which was subsequently decomposed into Kα<sup>1</sup> and Kα<sup>2</sup> components during the processing of the spectra.

Loading efficiency (LE) was calculated using the following equation:

LE = [(m(deferoxamine total) − m(c deferoxamine in supernatant))/m(deferoxamine total)] × 100

The mass of deferoxamine in the supernatant was determined by UV spectroscopy at a wavelength of 252 nm (calibration curve method).

X-ray fluorescence analysis of the samples was performed on a Clever C-31 X-ray fluorescence spectrometer. The relative measurement error was ±7%. A rhodium tube with a voltage of 50 kV and a current of 100 μA acted as a generator of γ-rays. The samples were taken without filters for 2000 s.

High-resolution electrospray ionization mass spectrometry (positive ion mode) was carried out on a APEX-Qe ESI FT-ICR instrument (Bruker, Billerica, MA, USA) with CH3CN as a solvent.

Antibacterial activity (in vitro and in vivo) and toxicity were evaluated completely as previously described by some of our group [32–35].

## **3. Results and Discussion**

*3.1. Preparation of Nanoparticles POX-1, POX-2 and POX-3*

Treatment of the chitosan solution with iron(III) chloride immediately results in the generation of yellow-colored nanoparticles POX-1 of unimodal size distribution with hydrodynamic diameter ca. 285 nm and a high positive zeta potential (ca. 32 mV, see Table 1). The formed nanoparticles do not change their characteristic size and zeta potential values in a water nanosuspension for at least 10 days. Remarkably, when immersed in water after lyophilization, POX-1 is almost instantly redispersed, with the complete restoration of the starting values of the hydrodynamic diameter and zeta potential.

**Table 1.** Characteristics of the obtained nanoparticles.


The addition of deferoxamine to the POX-1 nanosuspension leads to the rapid formation of novel POX-2 nanoparticles of smaller size (hydrodynamic diameter ca. 254 nm) and the same zeta potential value as for POX-1 (ca. +32 mV, see Table 1). In water, POX-2, after 24 h, converts into POX-3 with hydrodynamic diameter ca. 260 nm, while the zeta potential value remains unchanged. It is likely that POX-2 and POX-3 are the same system, but this will be discussed in the following sections. It should be noted that both POX-2 and POX-3 are completely redispersible after lyophilization.

To confirm that deferoxamine is part of POX-2 and POX-3 and is not present in the solution in free form, we separated the resulting POX-2 and POX-3 from the supernatant after mixing the POX-1 and deferoxamine solutions. High-resolution mass spectrometry with electrospray ionization of the supernatant did not reveal any deferoxamine signals. Therefore, deferoxamine is included in the POX-2 and POX-3 nanoparticles. The loading efficiency (LE) of deferoxamine was ca. 100%.

X-ray fluorescence analysis of POX-1, POX-2 and POX-3 confirmed the presence of iron in the samples.

POX-2 also was characterized by scanning electron microscopy. The SEM image of POX-2 is presented in Figure 1.

**Figure 1.** SEM image of POX-2.
