Effect of surface charge of nanohybrids on the intracellular bacterial killing efficiency CURRENT

Background : Lipid polymer hybrid nanoparticles (LPHNPs) are widely investigated nanohybrid system in drug and gene delivery and also medical imaging. A knowledge of lipids-based surface engineering and its effects on the physicochemical properties of LPHNPs affect the cell – NPs interaction, consequently, influence the cytological response is in high demand. Methods and Results : Herein, we developed a cationic and zwitterionic lipids-based surface engineering approach with antibiotics (Doxycycline or Vancomycin) loaded LPHNPs and examined the surface charge influence on the physiochemical characteristics, antibiotic entrapment, and intracellular release behaviors. Importantly, we examined the intracellular antibacterial activity of drug-loaded LPHNPs against Mycobacterium smegmatis or Staphylococcus aureus infected macrophages. Furthermore, cationic or zwitterionic lipids in LPHNP formulations improved the antibiotic loading efficiency and extended the duration of antibiotic release. In vitro particle uptake studies indicated that the cationic LPHNPs and bare nanoparticles (BNPs) were more efficiently internalized into macrophages than zwitterionic LPHNPs. Conclusion : A play in surface charge in the formulation of the macrophage uptake and intracellular bacterial killing efficiency of LPHNPs loaded with clinical antibiotics on macrophages infected with bacteria, provided a basis for optimizing their use in biomedical applications. ,

engineering approach with antibiotics (Doxycycline or Vancomycin) loaded LPHNPs and examined the surface charge influence on the physiochemical characteristics, antibiotic entrapment, and intracellular release behaviors. Importantly, we examined the intracellular antibacterial activity of drug-loaded LPHNPs against Mycobacterium smegmatis or Staphylococcus aureus infected macrophages. Furthermore, cationic or zwitterionic lipids in LPHNP formulations improved the antibiotic loading efficiency and extended the duration of antibiotic release. In vitro particle uptake studies indicated that the cationic LPHNPs and bare nanoparticles (BNPs) were more efficiently internalized into macrophages than zwitterionic LPHNPs.

Background
The treatment of intracellular bacterial infections is often challenging for clinicians, and novel therapies are needed [1]. Intracellular infections can be recurrent and difficult to treat, owing to the low availability of antibiotics in infected cells and insufficient host defenses [2]. Mycobacterium tuberculosis, Salmonella typhimurium, and Staphylococcus aureus [3] are the main intracellular bacteria, found inside the macrophages, where they act like a 'Trojan horse' to cause recurrent infections at secondary sites [2]. Intracellular bacterial infections typically require long-term antibiotic intake, which may lead to poor compliance and contribute to developing resistant strain [4]. Besides, most conventional antibiotics exhibit poor intracellular penetration and retention ability, which leads to a relapse of infection and antibiotic resistance [5]. 3 Engineered nanoparticles (NPs) are helping to counteract the intracellular antibiotic delivery [6][7][8][9][10].
The intracellular delivery of antibiotics associated with engineered NPs offers significant advantages over free drugs, including an improved drug efficacy by protection from degradation, optimal therapeutic levels at the site of bacterial infection via sustained release, and a reduced dosing frequency, thereby minimizing drug-associated toxicity [11]. Several engineered NPs and liposomes have been evaluated for antibiotic delivery [12]. However, a low encapsulation efficiency, high burst release, cytotoxicity, and poor stability have limited their clinical success [2].
Lipid-polymer hybrid nanoparticles (LPHNPs) can overcome the issues associated with polymeric NPs and liposomes [13][14][15][16]. Compared with typical NPs, hybrid NPs exhibit improved cell interactions, higher drug loading, and prolonged drug release [15,[17][18][19]. We previously reported that coating of lipids of different types on the PLGA-NP core, influences the size, surface charge, therapeutics loading efficiency including gene and small molecules, and release ability [14,17,18]. Furthermore, the nature and charge of the NPs surface are crucial determinants of the macrophage recognition and phagocytosis mechanism [19]. Therefore, in the present study, we developed a cationic and zwitterionic lipids-based surface engineering approach with antibiotics (Doxycycline or Vancomycin) loaded LPHNPs and examined the intracellular antibacterial activity of drug loaded LPHNPs against Mycobacterium smegmatis (M. smegmatis) or Staphylococcus aureus (S. aureus) infected macrophages to determine the optimal formulation.

Preparation and characterization of antibiotic loaded LPHNPs
In our previous study, we have investigated the different methods of preparation and characterization of LPHNPs. This experience helped us to significantly control the size and surface of charge of LPHNPs, which further improved the credibility and repeatability of our research [17,18,20]. The Fig 1., outlines the schematic diagram of the experimental design. Fig SI1 and Table 1 displays the production of different hybrid nano-formulations of drug-loaded cationic or zwitterionic LPHNPs and non-lipid layered bare BNPs using modified emulsion solvent-evaporation process and their physicochemical characteristics [18]. The size of hybrid nanoparticles (HNPs) is considering as an 4 essential parameter, with direct effects on cellular uptake, stability, and tissue distribution [21][22][23][24].
Thus, we used dynamic light scattering (DLS) to determine the size range, size distribution (PDI), and surface charge of LPHNPs [15]. Based on DLS, the modified solvent evaporation method yielded a relatively monodispersed LPHNPs and BNPs with a size range of 150-300 nm and narrow size distribution (PDI between 0.14-0.22). As summarized in Table 1, and illustrated in Fig 3A, the incorporation of cationic or zwitterionic lipids in the LPHNP formulation, resulting in a significant reduction in size (P<0.05) compared to those of non-lipid layered formulations (BNPs). For instance, cationic (CA) or zwitterionic (ZA) LPHNPs were 203 ± 6.6 and 191 ± 5.4 nm, respectively, which were smaller than the non-lipid layered BNP formulation (NA), i.e., 226 ± 9.6 nm. Consistent with previous reports from our group and other research groups, incorporation of either cationic or zwitterionic lipids significantly reduce the size of formed NPs, which could be explained by the fact that the processing of those NPs in a single step was stabilized by the function of the lipids with an emulsifying agent, possibly reducing the coalescence of particles [14,18].
We utilized a double emulsion solvent evaporation technique and the encapsulation of drug (Doxycycline or Vancomycin), drastically affect the size of these NPs, as indicated in Table 1. For instance, NA, CA and ZA were smaller than NB, NC, CB, CC, ZC. Additionally, BNP non-lipid layered formulations encapsulating Doxycycline (NB) or Vancomycin (NC) were larger than Cationic and Zwitterionic LPHNPs encapsulating Doxy (CB and ZB) or Vanco (CC and ZC).
Surface charge is another critical factor; it dictates the NP interactions with cells, penetration, and plays an essential role in colloidal stability [23,25]. Thus, we investigated the surface charge (Zeta potential) of cationic (C) and zwitterionic (Z) LPHNPs, and non-lipid layered BNPs (N) using the Zetasizer Nano ZS (Malvern Instruments, Malvern, UK) [15,18]. As expected, the inclusion of either Cationic or Zwitterionic lipids into the LPHNP formulations resulted in a differently charged surface, as shown in Table 1 and Fig 3B. Our results confirmed that the surface charge of BNPs is not affected by the incorporation of an antibiotic, i.e., Doxy (NB) or Vanco (NCB). In contrast, the inclusion of either cationic (C) or zwitterionic (Z) lipids into the LPHNP formulation resulted in a charge reduction from -29 ± 2.1 mV towards positive or neutral values [17]. As summarized in Table 1, the addition of cationic lipid to the NP formulations (CA, CB, and CC) resulted in positively charged surfaces and the inclusion of a zwitterionic lipid to the NP formulations (ZA, ZB, and ZC) resulted in charge reductions from that of the native surface of BNPs (-29 ± 2.1 mV).
The morphologies of antibiotic-encapsulated LPHNPs and BNPs were examined to ensure that they were hybrid nanoparticles with lipid and polymeric cores, rather than a random composition of liposomes and uncovered polymeric NPs [13,15,16]. As shown in Fig 2C, D and supplementary image (SI) SI1 (FESEM), the Vanco or Doxy-encapsulated BNPs and LPHNPs exhibited a nano-spherical shape, with no distinct morphological differences among the LPHNP groups. Additionally, EF-TEM images (Fig 2D) showed the core-shell hybrid structure of LPHNPs and confirmed the presence of the lipid layer on the PLGANP surfaces. A Cryo-TEM analysis was performed to inspect further the LPHNP structures identified by FE-SEM and EF-TEM [26]. Both cationic and zwitterionic lipid layered LPHNPs exhibited a perfectly spherical shape (Fig 2), and the apparent particle sizes corresponded well to the NP sizes calculated by DLS (Table 1). Highly electron-dense structures on the LPHNPs, which were easily melted upon electron beam exposure, and extended and loose structures were present in the dispersed phase, as reported by Colombo et al. 25 To obtain direct evidence for lipid self-assembly on the PLGA NP core, we incorporated fluorescent NBD-PC lipids into cationic or zwitterionic lipids. As shown in Fig 2, the CLSM image showed green fluorescence indicating a perfect nano-sphere shell layer, demonstrating that the simple and efficient process led to the formation of lipid layered hybrid nanoparticles [27]. The stability of antibiotic-encapsulated LPHNPs and BNPs was examined to ensure that the formulations are appropriate for prolonged storage and commercial feasibility. The z-average size was used to study NP stability by DLS at specific intervals. As shown in supplementary SI6 all the LPHNP formulations stored at 4°C remained stable during the entire observation period of 2 weeks, maintaining a slightly higher z-average than that on the first day of production and showing no visible signs of instability, such as sedimentation or aggregation The drug entrapment efficiencies for Vanco-or Doxy-encapsulated BNPs and LPHNPs was shown in Fig   summarized in Table 1. The DLS results showed that antibiotic encapsulation in LPHNPs and BNPs resulted in a moderate size increase. The encapsulation efficiencies of Doxy-BNPs (NB) and Cationic 7 or Zwitterionic lipid-layered LPHNPs (CB, and ZB) were 63%, 71%, and 79%, respectively. We also observed that the Vanco encapsulation efficiencies of BNPs (NC) and Cationic or Zwitterionic lipid layered LPHNPs (CC and ZC) were 57%, 64%, and 76%, respectively. The drug incorporation efficiency of hydrophilic antibiotics in non-lipid layered formulations (BNPs) was lower than that in lipid layered formulations.

In vitro drug release from LPHNPs and BNPs
The cumulative antibiotic release from Doxy-or Vanco-encapsulated BNPs and cationic or zwitterionic lipid-layered LPHNPs were assessed for 120 h. As shown in Fig 3, the cationic (C) or zwitterionic (Z) lipid layered LPHNPs displayed the sustained release of Doxy or Vanco with minimal burst release. In contrast, non-lipid layered BNPs (N) showed an initial burst release, followed by sustained drug release [28,29]. For instance, the antibiotic release rates of Doxy-BNPs (NB) at 12, 24, and 48 h were 65%, 69%, and 75%, respectively, while Doxy release rates from cationic (CB) and zwitterionic (ZB) lipid-layered LPHNPs at 24, 48, and 72 h were 38%, 54%, and 66% and 32%, 47%, and 61%, respectively. Similarly, the cationic (CB) and zwitterionic (ZB) lipid-layered LPHNPs encapsulating Vanco had a slower release rate than the non-lipid layered BNP-Vanco formulation (ZC). As shown in Furthermore, all LPHNP formulations exhibited minimal burst release with the sustained antibiotic release, unlike the non-lipid layered BNPs. This difference could likely be explained by the lipid on the NP core, which could slow the drug release kinetics by acting as a molecular fence between the polymeric matrix and aqueous phase [16,17].   In addition, our results showed that the synergism between the antibiotics and cationic-LPHNPs enhanced the antibacterial activity against S. aureus and M. smegmatis. Our plausible explanation for the enhanced in vitro antibacterial activity is the combinatorial antibacterial effect of the antibiotic and physiochemical properties, such as the size and charge, of the lipid layer LPHNPs [30]. We After 24 h of incubation, the BNPs and cationic-LPHNPs were more efficiently taken up by J774.1 cell than the zwitterionic LPHNPs (Fig 4). The intracellular distribution of fluorescently labeled NPs was further analyzed by CLSM to confirm that NPs were present in the cells rather than absorbed into the cell membrane [32]. The Fig5 shows that the cationic LPHNPs (C) surfaces were more rapidly recognized and processed by phagocytes than the zwitterionic surfaces (Z). Furthermore, Cationic

Staphylococcus aureus or
LPHNPs (C) show higher uptake nearly 93% followed by BNPs were 80% respectively. Interestingly, Zwitterionic LPHNPs were taken by macrophages less efficiently (49%) as compared to Cationic and non-lipid layered NPs. These results depict that macrophage recognition, uptake efficiency, and subsequent intracellular trafficking of tested nanocarriers is positively correlated with the surface properties.

Intracellular antibacterial activity of LPHNPs and BNP
The intracellular antimicrobial activity of cationic or zwitterionic LPHNPs and non-lipid layered BNPs were tested in macrophages in vitro. Free Doxy or Vanco treatment decreased the intracellular bacterial load (2-log-CFU compared with the initial infection) for M. smegmatis and S. aureus (Fig 5).

Treatment with cationic LPHNP-Doxy (CB) and LPHNP-Vanco (CC) resulted in a reduction of 4 or 3 log
CFU compared with the untreated infected cells as shown in (Fig 6) and (Fig 7). Similarly, treatment  Doxycycline is a broad-spectrum tetracycline-based antibiotic, which exerts bactericidal action by inhibiting protein synthesis [29,33]. Vancomycin is the most commonly used antibiotic for S. aureus infections; it inhibits the biosynthesis of peptidoglycan and the assembly of NAM-NAG-polypeptide in the peptidoglycan chain [5,28]. Overall, our experimental data indicated that the encapsulation of Doxy or Vanco into the LPHNPs or BNPs significantly reduced the antibacterial efficacy, but may have prolonged activity by extending the time of action of a single dose as compared to that for free antibiotics [29,33] [5,34]. This improved antibiotic entrapment ability of LPHNPs can be explained by the ability of lipids to act as a molecular fence, which prevents drug leakage in the external phase

Bacterial strain and macrophages
The bacteria and macrophages were obtained from our laboratory resources. The bacterial cultures

Preparation of antibiotic loaded LPHNPs and BNPs
The antibiotic (Doxy or Vanco)-loaded LPHNPs were prepared as described previously, with slight changes in the modified emulsion solvent-evaporation method [14,18]. lipid) nanoparticles were also prepared using the same conditions, with few modifications [17,18].

Characterization of LPHNPs and BNPs
The particle size (z-average) and size distribution (polydispersity index) of blank, antibiotic-loaded was removed, resulting in the formation of a thin (10-500 nm) sample film [26,36]. Additionally, the presence of the assembled lipid layer on the PLGA NP core was further confirmed by confocal laser scanning microscopy (CLSM; Carl Zeiss LSM 510 Meta) with fluorescent NBD-PC [17,18,27].

Antibiotic(s) encapsulation efficiency and release behavior
The loading efficiency of Doxy or Vanco in BNPs or LPHNPs was quantified by spectrophotometry (UV-160; Shimadzu, Tokyo, Japan) by measuring the amount of non-entrapped Doxy or Vanco in the external aqueous solution [37,38]. The external aqueous solution was obtained after centrifugation of the colloidal suspension for 30 min at 18,000 × g. In vitro drug release was performed for 120 h at 37°C. Antibiotic-encapsulated BNPs and LPHNPs were suspended in 3 mL of PBS with continuous shaking. The eluted Doxy or Vanco was withdrawn at predetermined time points, and an equal volume of fresh buffer was replaced at every sampling point. The samples were analyzed by UV spectrophotometry, as described above.

Determination of minimum inhibitory concentrations
The minimum inhibitory concentrations (MICs) of drug-free and Doxy-or Vanco-loaded BNP and LPHNP formulations were determined against M. smegmatis or S. aureus [5,33].

Intracellular uptake of BNPs and LPHNPs by macrophages
The J774.

Statistical analysis
The Bonferroni post-hoc test was used for comparisons between groups after at least three independent sets of experiments performed in triplicate. Differences were considered significant at P < 0.05. The Prism software package (version 5.02; GraphPad Software Inc., La Jolla, CA, USA) was used to perform the statistical tests.

Availability of Data and Materials
Data and materials are available for any research. Funding.