Hyaluronic Acid-Based Nanomotors: Crossing Mucosal Barriers to Tackle Antimicrobial Resistance (2025)

2.1. Synthesis and Characterization of HA-NPs

We explored the preparation of covalently cross-linked HA-based NGs, as the core for the NMs (Figure 1A). For that, we modified HA with methacrylic anhydride (MA) to introduce vinyl groups into the polymeric chains (Figure S1) for their posterior use in the formation of the NGs. We confirmed the methacrylation by ¹H NMR spectrum, which evidences the structural changes of the molecule with peaks emerging at 6.1, 5.7, and 1.9 ppm, corresponding to the protons of the methylene (a, b in the NMR spectra) and methyl (c in the NMR spectra) groups, respectively, in contrast to the spectrum of unmodified HA (Figure 1B). Subsequently, we prepared HA-NGs by copolymerizing the methacrylated HA with the cross-linker diethylene glycol dimethacrylate (DEGDA), in an aqueous solution at 70 °C for 2 h. This process was initiated by potassium persulfate (KPS), and the reaction took place under an argon atmosphere. We purified the final material with a dialysis membrane of 50 kDa against ultrapure water with water change twice per day for a week. The Fourier transform infrared attenuated total reflectance (FTIR-ATR) spectra of methacrylated hyaluronic acid (MAHA), DEGDA, and the HA-NGs (Figure 1C) exhibited a transmission band at 3200 cm⁻¹ for HA-NGs, associated with the stretching of hydroxyl groups of MAHA. Additionally, a peak at 1700 cm⁻¹ indicates the stretching of the carbonyl group from DEGDA, confirming that the NGs were formed through the cross-linking reaction between MAHA and DEGDA. We further confirmed the shape and morphology of the NGs by transmission electron microscopy (TEM) (Figure S2) and dispersity by scanning electron microscopy (SEM) (Figure 1D), ratifying a narrow size distribution with a diameter ca. 54.1 ± 5.1 nm (N = 500) (Figure 1E). As an alternative strategy for HA-NPs synthesis, we prepared HA-based NPs via self-assembly HA (20 kDa) with p-l-arginine (5800 Da), with few modifications from the reported literature (Supporting Information, Figure S3 and Tables S1–S3). (51) We successfully obtained highly monodisperse NPs with a 485.0 ± 27.01 nm diameter using an injection speed rate of 0.67 mL min⁻¹.

2.2. HA-NMs Fabrication

To fabricate the HA-based NMs, we chemically linked urease onto the surface of HA-NGs and HA-NPs via click-chemistry of the 1-ethyl-3-(3-(dimethylamino)propyl)carbodiimide and N-hydroxysuccinimide (EDC/NHS) (Figure 2A). EDC/NHS allowed the formation of chemical junctions between the amino groups of the lysine from the urease with the carboxylic groups of the HA. For HA-NPs, we observed a lack of stability after their covalent surface functionalization with urease to form NMs, i.e., aggregation and loss of monodispersity. We attributed this undesired effect to the strong chemical bonds used for anchoring the enzyme, which affected the physical forces essential to maintaining the structural integrity of the assembled NPs. Consequently, this strategy was considered to be unsuitable for further experiments. In contrast, for HA-NGs we proved the successful preparation of the NMs by dynamic light scattering (DLS), observing an increase in the overall diameter after enzyme functionalization (Figure 2B). The results indicated a highly monodisperse population, as made evident by the narrow peak width in both samples. The formation of HA-NMs resulted in a noticeable increase in the z-average size with a narrow polydispersity [Figure S4A, (i,ii)], which increased from 217 ± 2 nm to 541 ± 24 nm, confirming the successful enzyme conjugation on the HA-NMs. It is worth noting that this striking change of size can be attributed to enzyme clustering since the size of a single enzyme is 12.1 nm according to crystallography calculations (Figure S4B). Therefore, the NMs would not simply have double the size of the NGs. In addition, we hypothesize that clustering formation could play a key role in the asymmetry formation of NMs. (52) It is important to mention that while the HA-NGs diameter determined by DLS was around 217 ± 2 nm, the particle size measured by TEM and SEM was approximately 55 nm (Figures 1D and S2). This discrepancy is attributed to dehydration of the NGs in the dry state.

Regarding the surface charge of the NGs, the z-potential of HA-NGs was −51.6 ± 0.3 mV in ultrapure water, attributed to the presence of carboxylic groups on their surface. After the covalent attachment of urease, the surface charge shifted to −18.6 ± 0.5 mV (Figure 2C). This change toward more positive values indicates the successful attachment of the enzyme to the NGs’ surface, as the surface charge of urease, determined computationally, is on average −1.5 ± 0.03 kcal mol⁻¹ × 10⁻¹ (Figure S4B). The enzyme attachment was also confirmed using a BCA kit assay. The amount of urease bounded was 98.2 ± 1.27 μg mL⁻¹ (Figure S4C).

Moreover, we determined the enzymatic activity of free urease and that on the surface of the NMs, at different urea concentrations, and adjusted them to the Michaelis–Menten model (Figure 2D). The enzymatic kinetic parameters of HA-NG NMs displayed a slightly lower k_cat^app and higher K_m^app compared to the free enzyme resulting in a reduced specificity constant, i.e., 9.3 ± 2.7 and 7.9 ± 1.0 mM⁻¹ s⁻¹, which diminishes the enzyme’s efficiency in converting substrates into products (Table S4). This slight reduction is a common occurrence when enzymes are immobilized due mainly to the restriction of overall protein structure and substrate access to the active center. (53)

Finally, we performed FTIR analysis of the resulting HA-NMs (Figure S4D). Although no new bands were observed due to the overlapping signals from amine, hydroxyl, and carboxyl groups in both urease and the HA polymer, the spectra confirmed the presence of these functional groups. We confirm that after urease anchoring the NGs remain stable, as all functional groups from the initial HA-NGs are still present in the spectrum, displaying a similar pattern to that observed in Figure 1C.

2.3. Motion Analysis of HA-NMs

We evaluated the motion of HA-NMs at the single and collective level in the presence of different urea concentrations. For the single particle tracking, we evaluated the self-propelled capabilities of HA-NMs and recorded their trajectories in the presence of 0 and 100 mM of urea (Figure 3A and Video S1). The mean squared displacement (MSD) (Figure 3B) was derived from the particle trajectories for single particle tracking. HA-NMs with and without urea exhibited a linear MSD, indicative of diffusive motion, typically observed for nanoscale particles. (54) The covalent bonds between the enzymes (urease) and HA-NGs surface via EDC/NHS led to the asymmetric enzyme distribution on the particle’s surface, which produces a propulsive force that drives NMs motion, as previously reported. (55−57)

We observed an increase in the MSD slope in the case of NMs in the presence of urea, indicating enhanced diffusion. Notably, the MSD slope values for HA-NMs were higher, reaching approximately 6.7 μm² after 1.5 s, compared to other urease-powered NMs of similar diameters constructed from inorganic materials such as mesoporous silica-based NMs (MSNP-NMs), which exhibited MSD values in the range of 3–3.5 μm² at same time point. (13,17,47,58) We attributed these differences to several factors, such as surface interactions between the HA polymer and the aqueous surrounding media, propulsion efficiency, and hydrophilicity of the materials, highlighting the benefit of using organic-based NMs. Typically, MSNP-based NMs, being more rigid and hydrophobic than HA, tend to exhibit lower diffusion rates, whereas the flexibility of polymers such as HA enhances the propulsion efficiency, resulting in a steeper MSD slope. Besides, we observed a significant increase in the diffusion coefficient of HA-NMs in the presence of urea (Figure 3C), rising from 1.11 ± 0.039 to 1.51 ± 0.079 μm s⁻¹. This 36% diffusion increment provides clear evidence of HA-NMs’ self-propulsion capabilities, driven by the catalytic reaction of urease, which is in accordance with previous reports of urease-powered NMs. (13,17,21,47,58)

After proving the motion of HA-NMs at a single particle level, we explored their collective dynamic motion in the presence of 0, 50, 100, and 200 mM of urea (Figures 3D and S5 and Video S2). We observed an increase in the normalized covered area explored by the HA-NMs, consistent with the reported literature. (20,27) In the absence of urea, most of the HA-NMs were set onto the bottom of the Petri dish after slight diffusion. We quantified the area covered by the NMs (Figure 3E). Although the covered areas were similar in the presence of 0 mM (35.1 ± 2.0 mm²) and 50 mM urea (35.8 ± 6.2 mm²), we observed the formation of plumes at 50 mM urea (Figure S5). This behavior suggests that the NMs were active, consistent with previously reported characteristics of urease-based NMs. (20,59) In the presence of 100 mM urea, the area covered by the HA-NMs increased to 39.1 ± 3.5 mm² after 30 s (1.25-fold increase compared to 0 mM urea) and to 41.7 ± 4.3 mm² after 60 s, representing a 1.18-fold increase. When the urea concentration was raised to 200 mM, the HA-NMs spread more rapidly, covering 44.5 ± 1.6 mm² at 30 s (1.42-fold increase) and 56.7 ± 7.2 mm² at 60 s (1.6-fold increase). These results suggest that the active propulsion of HA-NMs, enhanced by fluid mixing induced by urea, significantly promotes their expansion and collective displacement in a concentration-dependent manner.

2.4. Evaluation of the Antimicrobial Effect of CHL@HA-NMs

The higher encapsulation of CHL in HA-NMs compared to that in MSNP-NMs can be attributed to the structural and physicochemical differences between these two materials. HA-NGs exhibit a highly hydrated, flexible, and expandable polymeric structure that allows them to absorb significant amounts of water, which facilitates the solubilization and retention of drugs. Moreover, the drug-loading capacity in NGs can be chemically tuned to optimize the matrix or surface interaction with the drug, as well as network structure permeability, thereby maximizing the encapsulation efficiency of CHL. (62) We observed around 30% (65.01 ± 11.43 μg) of CHL released from the HA-NMs after 24 h. This value is in line with reported literature for similar HA-NGs systems that load hydrophobic drugs, such as doxorubicin. (63) The antimicrobial efficacy of CHL@HA-NMs and CHL@MSNP-NMs was assessed against a nonpathogenic strain of E. coli. Minimum inhibitory concentration (MIC) values were determined by measuring the optical density (OD) at 600 nm in the absence or presence of urea (100 mM urea in PBS) and exploring a range of NMs (CHL@HA-NMs and CHL@MSNP-NMs) concentrations of 0, 5, 10, 20, and 50 μg mL⁻¹ (Figure 4B). The measurements were conducted at different time points, specifically at 4, 18, and 24 h. As observed after 24 h (Figure 4C and Table S5), bacterial proliferation increased significantly over time for the lowest concentration of CHL@HA-NMs, 5 and 10 μg mL⁻¹, but not much for the higher concentrations, 20 and 50 μg mL⁻¹. The most effective CHL@HA-NMs concentration to inhibit the bacteria proliferation was 50 μg mL⁻¹ (equivalent to 0.87 μg of CHL), although significant inhibition was also observed at 20 μg mL⁻¹, which effectively controlled proliferation up to 24 h. It is worth mentioning that the catalytic products of urea hydrolysis resulted in a synergic effect with the drug, reducing the proliferation even at the lowest NMs concentration, i.e., 1−1.5-fold proliferation reduction for all tested concentrations of HA-NMs at 24 h (Figure 4D). We compared the capacity of MSNP-NMs (already tested in our group) (48) and HA-NMs as drug delivery systems for antimicrobial effect. Conversely, as displayed in Figure 4E and Table S6, we found that CHL@MSNP-NMs were ineffective at any concentration at 24 h, likely due to their lower drug-loading capacity. Noticeably, the synergic effect between ammonium and the drug displayed the same trend for MSNP-NMs as it was observed for HA-NMs (Figure 4E). Moreover, we also observed no significant difference in the fold change synergistic effect of ammonium for the case of MSNP-NMs (Figure 4F). Data corresponding to the earlier time points (4 and 18 h) for CHL@HA-NMs are provided in Figure S7 and Tables S7, S8 and for CHL@MSNP-NM in Figure S8 and Tables S9, S10, which are in line with the results at 24 h. We observed a similar antimicrobial profile for the free CHL (Figure S9 and Tables S11–S13).

Considering the potential capacities of CHL@HA-NMs for inhibiting bacteria proliferation, we aim to explore the HA-NMs’ interaction with bacteria. Therefore, we assessed the internalization of the HA-NMs inside the bacteria by confocal microscopy (Figure 5). For that, we labeled the HA-NMs with fluorescein isothiocyanate (FITC), reacting the FITC with the amine groups of ureases prior to attachment in the NGs (Figure 5A) and staining the bacteria with 4′,6-diamidino-2-phenylindole, dihydrochloride (DAPI) (Figure 5B) and Nile red (Figure 5C), which bind to the DNA and membrane, respectively.

In Figure 5D, we show a colocalization of fluorophores, suggesting that HA-NMs bind and internalize into bacteria cells. Furthermore, we observed that a significant amount was taken up in less than 2 h of incubation. We hypothesized that the polysaccharide-like structure of HA-NMs might foster their internalization inside the cells. We also evaluated the stability of the HA-NMs over time, following 24 and 72 h of incubation in the presence or absence of urea. The results showed a reduction in the intracellular presence of HA-NMs, likely due to dilution upon cell division (Figure S10).

2.5. NMs for Crossing Biological Barriers

Hydrophobic drugs often become trapped within viscous BBs, limiting their therapeutic efficacy. (64) Here, we assessed the potential of CHL@HA-NMs for crossing mucin, as a BB model, enhancing the CHL antibacterial potential. To test this, we employed an in vitro transwell system designed to mimic a mucin layer, enabling the evolution of HA-NMs to inhibit bacterial growth across such a barrier. In this setup, we cultured E. coli bacteria into the lower compartment of the transwell system, while the upper chamber contained a mucin layer (Figure 6A). Later, we added in the upper chamber a solution containing free CHL [Figure 6A, (i)] or CHL@HA-NMs [Figure 6A, (ii)] combined with urea (100 mM). The bacterial growth was quantified by measuring the absorbance of the liquid in the lower chamber, which directly correlates with bacterial proliferation (higher absorbance values indicate greater bacterial growth). Absorbance was monitored at 4, 8, and 24 h, and the results were plotted as a percentage of bacterial proliferation (Figure 6B). The results demonstrated that treatment with CHL@HA-NMs significantly inhibits E. coli proliferation compared to the control and the free CHL across different time points. At 4 h, all conditions showed minimal bacterial proliferation (∼5−10%) due to lag-time of bacterial growth. Although a clear trend can be seen in the samples with CHL (both free CHL and CHL@HA-NMs), no significant differences are observed. After 8 h, E. coli from the positive control group proliferated to 46.5 ± 1.5%, while proliferation in the free CHL group increased to 31.7 ± 8.2%, still remaining lower than the control, indicating partial antibiotic efficacy despite the mucin barrier. Notably, CHL@HA-NMs treatment further suppressed bacterial proliferation to 5.7 ± 1.7%, indicating that a huge percentage of HA-NMs crossed the mucin and delivered the CHL to the bacteria cells, enhancing therapeutic efficacy over the free drug. This response was more pronounced after 24 h, where positive control peaked at 68.2 ± 28.1% proliferation. In contrast, free CHL allowed bacterial growth to progress to 30.9 ± 12.9%. Notably, the presence of CHL@HA-NMs treatment resulted in negligible bacterial growth, with bacterial proliferation reduced to 2.4 ± 1.7%, suggesting that the NMs provide more sustained and effective delivery of the antibiotic, with an 11.6 ± 2.6-fold greater reduction in proliferation compared to that achieved with the free antibiotic (Figure 6C). The self-propelling CHL@HA-NMs effectively inhibited bacterial growth by enabling targeted, continuous CHL release and enhancing penetration through the mucin barrier, demonstrating their potential as an efficient method for delivering antibacterial agents across mucosal barriers.

Altogether, the HA-NM system presented in this work offers multiple strategic advantages over previously reported systems, particularly those based on inorganic materials like MSNP-NMs. These urease-powered HA-NMs exhibit superior diffusion and propulsion efficiency in the presence of fuel, facilitated by their hydrated, flexible, and lightweight structure, which allows them to navigate more easily through media compared to heavier nanostructures. Additionally, they exhibit enhanced drug-loading capacity compared with MSNP-NMs due to their amphiphilic matrix and their capability to swell. Furthermore, HA-based nanocarriers possess mucopenetrating properties, which are advantageous for developing drug delivery systems in viscous environments, such as mucus. Their autonomous propulsion removes the need for external activation (e.g., magnetic guidance or light), simplifying clinical translation and positioning HA-NMs as powerful and versatile platforms for overcoming BBs and advancing therapeutic delivery.

4.1. Synthesis of Methacrylated HA Polymer

MAHA was synthesized as a precursor for the preparation of HA-NGs. Briefly, sodium hyaluronate (200 mg, 0.01 mmol) was dissolved in ultrapure water (20 mL). Then, MA (200 μL, 1.26 mmol) was added to the solution. The pH was adjusted to 8 using 0.1 M NaOH and the mixture reacted for 24 h at 4 °C. Then, the product was purified by recrystallization in cold EtOH. The final product, obtained with a 65% yield, was collected and stored at −20 °C until use. The product was then characterized by proton nuclear magnetic resonance (¹H NMR) spectroscopy using a Bruker 400 MHz equipment and Fourier Transform Infrared with attenuated total reflectance (FTIR-ATR) spectroscopy performed on a PerkinElmer Frontier equipment.

4.2. Synthesis of HA Nanogels

HA-NGs were synthesized via the chemical cross-linking of MAHA with DEGDA, as reported in the literature, with minimum modifications. (63) Briefly, MAHA (100 mg, 0.005 mmol) and DEGDA (126 mg, 0.58 mmol) were dissolved in ultrapure water (20 mL). The mixture was bubbled with argon for 1 h before starting the polymerization to avoid the formation of peroxides’ undesired side products. Then, an aqueous solution of KPS previously degassed (3 mg, 0.011 mmol, 4.54 mL) was added to initiate the polymerization. The reaction was performed at 70 °C for 2 h. The final material was purified through dialysis (MWCO 50 kD) against ultrapure water. After freeze-drying the material (Christ alpha 1−4 LSC basic, working operation at −60 °C and 0.010 mbar), the resulting product was obtained as a white solid with an 80% yield.

4.3. Functionalization of HA-NGs with Urease to Fabricate HA Nanomotors

Surface modification of HA-NGs was achieved by functionalizing them with urease via click chemistry of EDC/NHS. First, the carboxylic groups on the HA-NGs (3 mg, 3 mg mL⁻¹ in PB) were activated by adding EDC (1 μL, 5.70 μmol), forming an active ester intermediate. Next, NHS (2 mg, 0.017 mmol) reacted with the activated ester, generating a more stable NHS-ester intermediate, which reacts toward nucleophiles. The mixture was stirred for 30 min to ensure carboxylic activation. Finally, purified urease dispersed in PB (500 μg, 0.925 nmol) was added to the mixture, which was stirred for 2 h and 30 min at room temperature (see Figure S11 for more details about urease purification). This allowed the covalent immobilization of urease onto the HA-NGs surface via amino groups of lysine with the carboxylic groups of HA.

After the reaction, the resulting NMs were purified to remove the unbound enzyme using a dialysis membrane (MWCO 1000 kD) against PB. The amount of protein attached was quantified using a BCA Protein Assay Kit by measuring the absorbance at 562 nm of the NMs solution with a Synergy HTX Absorbance microplate reader.

4.4. Purification of Urease from Canavalia ensiformis

Protein purification was carried out with a fast protein liquid chromatography, NGC Quest 10 Chromatography system, Biorad, as previously reported. (65) Briefly, 90 mg (0.167 μmol) of jack bean urease from C. ensiformis (type IX, Sigma-Aldrich catalog no. U4002) was dissolved in PB 1× (300 μL, pH = 7.4). Sample was previously centrifugated at 9400 g for 5 min to remove aggregates and insoluble particles. The protein solution was loaded onto an ENrich SEC 650 10 × 300 size-exclusion column (Biorad cat. no. 780-1650) using a 100 μL loop. The purified urease was collected in fractions of 0.5 mL performing three successive runs, thus collecting 1.5 mL of each fraction.

4.5. Synthesis of MSNPs and MSNP-NMs

MSNPs with an average diameter of 450 nm were produced via a sol−gel process, modified from the Stöber method. (66) First, CTAB (570 mg, 1.56 mmol) and TEOA (35 g, 0.23 mol) were dissolved in ultrapure water (20 mL) and heated to 95 °C in a silicon oil bath. The reaction took place under reflux and continuous stirring for 30 min. Then, TEOS (1.5 mL, 6.71 mmol) was added slowly with a Pasteur pipet, and the reaction continued for 2 h under the same conditions. The resulting MSNPs were collected by centrifugation at 1350g for 5 min. To remove CTAB, MSNPs were treated with an acidic MeOH solution under reflux. The MSNPs were suspended in MeOH (30 mL), and HCl (1.8 mL, 37%) was added. This mixture was heated in a silicon oil bath at 80 °C for 15 h. Finally, the MSNPs were collected by centrifugation at 1350g for 5 min, followed by three EtOH washes, with 10 min of ultrabath sonication between each centrifugation. The MSNPs were functionalized with APTES to introduce amine groups on their surface. Briefly, APTES (6 μL, 25 μmol) was added to a MSNP suspension (1 mg mL⁻¹) in 70% EtOH. The mixture was heated to 70 °C and stirred for 1 h. Afterward, the resulting MSNP-NH₂ were collected by centrifugation and washed three times in EtOH and ultrapure water, with vortexing and sonication between each step. Next, MSNP−NH₂ particles (900 μL, 1 mg mL⁻¹ in PBS 1×) were activated with GA (100 μL, 0.264 mmol) as a linker for covalent protein binding for 2 h at room temperature on a rotary shaker. After that time, particles were collected, washed in PBS, and resuspended in a solution containing urease (3 mg mL⁻¹) to fabricate the MSNP-NMs. The reaction was left on a rotary shaker overnight at room temperature. The final MSNP-NMs were collected by centrifugation and washed with PBS.

4.6. Hydrodynamic Diameter and Surface Charge

The hydrodynamic diameter, size distribution, and surface charge of HA-NGs, HA-NMs, and MSNP-NMs were measured by DLS using a Zetasizer Nano ZS Malvern Instrument. All measurements were conducted at 25 °C with light scattering detected at 173°, and three replicates were acquired for each condition.

The surface charge of urease was analyzed using UCSF ChimeraX software (UCSF Resource for Biocomputing) employing the Coulombic potential method.

4.7. HA-NMs Enzymatic Activity

4.8. Transmission and Scanning Electron Microscopy

TEM images were obtained using a JEOL JEM-2100 microscope at 200 kV, equipped with a LaB6 thermionic electron gun, and operated at an accelerating voltage of 200 kV. Images were recorded using a Gatan Orius CCD camera. Data analysis was carried out using the software ImageJ (v.1.53). SEM images were captured by an FEI NOVA NanoSEM 230 instrument at 5 kV. Before SEM observation, the samples were sputter-coated with a 10 nm layer of gold atoms to enhance conductivity, resolution, and contrast. The coating was applied using a Leica EM ACE600 sputter coater with a current strength of 30 mA, resulting in a 10 nm thick coating after 2 min of exposure.

4.9. Single-Particle Tracking and Motion Analysis

The motion of HA-NMs at a single particle level was recorded by using an inverted Leica DMi8 microscope equipped with a Hamamatsu high-speed camera and a 63× water-immersive objective.

The diffusion coefficient (D_e) was determined by fitting the data to eq 2, which is applicable for small particles with low rotational diffusion at short time intervals. In both cases, 20 particles were analyzed per condition, and the standard error of the mean (SE) was used to calculate the error.

4.10. Collective Displacement and Motion Analysis

The collective displacement of HA-NMs was recorded using optical videos obtained with a Hamamatsu camera attached to a Leica DMi8 inverted fluorescence microscope equipped with a 12.5× objective. A Petri dish was filled with an aqueous urea solution at different concentrations (0, 50, 100 and 200 mM). Then, a drop of HA-NMs (5 μL, 1.5 mg mL⁻¹) was then added to the dish, containing aqueous media with urea. Videos were recorded for 1 min at a frame rate of 20 FPS and posteriorly analyzed with the software ImageJ (v.1.53) to quantify the areal coverage of HA-NMs over time.

4.11. Bacterial Strains and Media

Nonpathogenic strainE. coli BL21 DE3 was grown in Luria–Bertani (LB) medium for 24 h at 37 °C (220 rpm). On the next day, bacterial overnight was diluted 1:500 in LB broth (5 mL) and they were grown in LB at 37 °C incubating until the logarithmic phase (OD = 0.6–0.8) was reached before inoculation.

4.12. Confocal Microscopy

HA-NMs were labeled with FITC to enable visualization by confocal microscopy. First, a stock solution of FITC (6 mg mL⁻¹) was prepared by dissolving FITC (6 mg) in DMSO (1 mL). From this stock solution, 50 μL was added to HA-NMs (1 mL, 1.5 mg mL⁻¹) in a light-protected environment. The mixture was continuously stirred overnight to ensure complete labeling. To remove nonencapsulated FITC, the solution was dialyzed against PB (MWCO 14 kD). All steps were protected from light by covering the materials with aluminum foil.

E. coli cells grown in LB liquid medium were incubated with urea (0 or 100 mM) at 37 °C for 2, 24, and 72 h with FITC-labeled HA-NMs at a concentration of 12.5 μg mL⁻¹. After this incubation, bacteria were rinsed twice with PBS 1x to wash away the excess of FITC-labeled HA-NMs. Bacteria were stained with Nile red, a lipophilic cell membrane stain (10 μg mL⁻¹) and DAPI, a DNA-specific fluorescent stain (1 μg mL⁻¹) for 30 min at 37 °C. After that, bacteria were centrifuged at 5000g for 5 min, washed once with PBS, and recentrifuged. Finally, the bacterial pellet was resuspended in a small volume of PBS (50 μL). Bacteria was fixed with 1% paraformaldehyde in ultrapure water for 30 min at room temperature, centrifuged and resuspended in PBS (50 μL). Images were obtained with a confocal microscope (ZEISS LSM 800) by using a 100× oil-immersive objective.

4.13. Chloramphenicol Encapsulation in HA-NMs and MSNP-NMs

MSNP-NMs (0.75 mg) were resuspended in a CHL solution (1 mL, 1 mg mL⁻¹) for drug loading. In contrast, for HA-NMs, an aliquot of HA-NMs (500 μL, 1.5 mg mL⁻¹) was mixed with a CHL solution (500 μL, 2 mg mL⁻¹) in a 1:1 ratio (v/v). The mixtures were incubated for 24 h at room temperature with continuous end-to-end mixing on a rotary shaker. After incubation, the CHL-loaded MSNP-NMs (CHL@MSNP-NMs) were collected and washed three times by centrifugation to remove the nonloaded drug. For CHL-loaded HA-NMs (CHL@HA-NMs) the procedure was similar, but the nonloaded drug was removed by filtration using Amicon Eppendorf filters (MWCO 30 kD). Supernatants from both types of NMs were kept and analyzed by measuring the absorbance at 277 nm using a Synergy HTX Absorbance microplate reader to determine the drug-loading capacity and entrapment efficiency of both types of NMs.

4.14. Antibacterial Assays

MICs of NMs were determined using the broth microdilution technique in LB with an initial inoculum of 5 × 10⁵ cells mL⁻¹ in sterile 96-well polystyrene microplates. Briefly, NMs were added to the plate as solutions in LB broth at concentrations ranging from 0 to 50 μg mL⁻¹. In the case of free antibiotic, the concentration added was determined by a prior release assay to enable a more accurate comparation. The MIC value was considered the lowest concentration of the antimicrobial system that inhibited the visible growth of bacteria. After 2, 4, 18, and 24 h of incubation at room temperature, the plates were read in a spectrophotometer (Synergy HTX Absorbance microplate reader) at 600 nm. All assays were done in three independent replicates.

4.15. Mucin Crossing Studies

To evaluate the ability of the NMs to penetrate viscous media, a transwell model with a mucin layer was employed to test the capacity of the HA-NMs to cross the mucin barrier. In this setup, the lower chamber was inoculated with E. coli at a concentration of 5 × 10⁵ cells mL⁻¹ (LB medium, final volume of 600 μL). The upper chamber was coated with a mucin layer (100 μL, 2 wt % in PBS), and urea was added to achieve an equilibrium concentration of 100 mM. Then, a suspension (200 μL) containing free CHL drug (0.867 μg) or NMs loaded with CHL (10 μg) combined with urea (100 mM) was introduced into the upper compartment. Absorbance in the lower chamber was measured at 600 nm using a spectrophotometer (Synergy HTX Absorbance microplate reader) for 4, 8, and 24 h.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.5c03636.

• Structural modification of HA; structural characterization of HA-based NGs (HA-NGs); optimization of the HA-NPs by self-assembly; structural characterization of HA NG NMs (HA-NMs); characterization of mesoporous silica-based NMs (MSNP-NMs); evaluation of the antimicrobial effect of the chloramphenicol (CHL) loaded into the HA-NMs or MSNP-NMs; internalization assay by confocal microscopy; and urease purification (PDF)

• Video S1: Single particle tracking of HA-NMs in the presence or absence of 100 mM urea in aqueous media (MP4)

• Video S2: Collective motion of HA-NMs in the presence or absence of 50, 100, and 200 mM of urea in PB, in aqueous media (AVI)

• Corresponding Author

  • Samuel Sánchez - TheBarcelona Institute of Science and Technology (BIST), Institute for Bioengineering of Catalonia (IBEC), Baldiri i Reixac 10-12, 08028 Barcelona, Spain; CatalanInstitution for Research and Advanced Studies (ICREA), Passeig Lluís Companys 23, 08010 Barcelona, Spain; https://orcid.org/0000-0001-9713-9997; Email: [emailprotected]

• Authors

  • Noelia Ruiz-González - TheBarcelona Institute of Science and Technology (BIST), Institute for Bioengineering of Catalonia (IBEC), Baldiri i Reixac 10-12, 08028 Barcelona, Spain; Facultatde Física, Universitat de Barcelona(UB). C. Martí i Franquès, 1-11, 08028 Barcelona, Spain; https://orcid.org/0000-0002-2467-331X

  • Daniel Sánchez-deAlcázar - TheBarcelona Institute of Science and Technology (BIST), Institute for Bioengineering of Catalonia (IBEC), Baldiri i Reixac 10-12, 08028 Barcelona, Spain; Present Address: Nanobots Therapeutics SL, Baldiri i Reixac 4, 08028 Barcelona, Spain; https://orcid.org/0000-0001-6034-8525

  • David Esporrín-Ubieto - TheBarcelona Institute of Science and Technology (BIST), Institute for Bioengineering of Catalonia (IBEC), Baldiri i Reixac 10-12, 08028 Barcelona, Spain; https://orcid.org/0000-0002-9858-415X

  • Valerio Di Carlo - TheBarcelona Institute of Science and Technology (BIST), Institute for Bioengineering of Catalonia (IBEC), Baldiri i Reixac 10-12, 08028 Barcelona, Spain; https://orcid.org/0000-0002-5099-0685

• Author Contributions

  N.R.G. and D.S.dA. contributed equally to this work. N.R.G. optimized the synthesis of HA self-assembly NPs, HA-NGs, and NMs. She also conducted DLS, z-potential, single-particle motion, and collective motion measurements. D.S.dA. conducted bacterial experiments, protein purification, and enzymatic activity assays. D.E.U. provided scientific guidance and supervision for the preparation of NMs and performed SEM, NMR, FTIR, and collective motion characterizations. V.D.C. carried out confocal microscopy experiments. N.R.G. wrote the first draft of the manuscript. D.S.dA., D.E.U., V.D.C., and S.S. improved the manuscript. S.S. supervised and administered the work. All authors have reviewed and approved the final version of the manuscript.

• Notes

  The authors declare no competing financial interest.
  

The research leading to these results has received funding from the grant PID2021-128417OB-I00 funded by MCIN/AEI/10.13039/501100011033 and “ERDF A way of making Europe” and European Union Next Generation EU/PRTR, (Bots4BB project), the CERCA program by the Generalitat de Catalunya, Secretaria d’Universitats i Recerca del Departament d’Empresa i Coneixement de la Generalitat de Catalunya through the project 2021 SGR 01606, and the Centro de Excelencia Severo Ochoa CEX2023-001282 S, funded by MICIU/AEI/10.13039/501100011033. This project has also received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (iNANOSWARMS, 866348), from the projects MucOncoBots (101138723) and OrthoBots (101189426), and from “la Caixa” Foundation under the grant agreement LCF/PR/HR21/52410022 (BLADDEBOTS). This project was supported via the Leonardo Fellowship “Beca Leonardo a Investigadores y Creadores Culturales 2024” from BBVA Foundation.N.R.G. acknowledges the MCIN/AEI for her predoctoral fellowship (PRE2019-088801). D.S.dA acknowledges the MCIN/AEI for his postdoctoral Juan de la Cierva fellowship (FJC2021-048024-I) and Torres Quevedo grant (PTQ2023-013272). S.S. also acknowledges the “Constantes y Vitales” 2023 prize. Some of the figures and the Table of Contents were created in BioRender.

Download PDF