Recent developments in the use of nanoparticles for treatment of biofilms

2.1 Biofilm developmental cycle

The formation of the three-dimensional complex structure of the biofilm starts with the attachment of free-floating bacteria cells onto a living or non-living surface in a moist environment, followed by the secretion of extracellular polymeric substance (EPS) and the growth of the three-dimensional complex structures of biofilms [48]. EPS is the main component of all biofilms accounting for almost 90% of the entire biomass, and it is composed mainly of polysaccharides, proteins, nucleic acids, and lipids [49]. The EPS matrix physically holds the biofilm bacteria together, facilitating cell-cell communication and gene transfer and protecting bacteria from antibiotics and the host’s immune system. Moreover, the EPS matrix provides nutrients to the bacteria such as carbon-, nitrogen-, and phosphorus-containing compounds [49]. After the biofilm is mature, biofilm bacteria disperse from the biofilm community to form new colonies and gain better access to nutrients [48].

Different treatment strategies have been proposed for each stage of biofilm development. Anti-adhesion agents, such as mannosides, pilicides, and curlicides as well as anti-biofilm polysaccharides, can be used to prevent bacteria from adhering onto the surface [50], [51], [52], [53]. During microcolony formation and biofilm maturation, silver NPs, lytic phages, and EPS-degrading enzymes may be useful [32], [54], [55]. By manipulating dispersing signals, biofilm bacteria can be dispersed into surrounding environments before the natural dispersal occurs [56].

2.2 Biofilm antibiotic resistance

Within just 2 years of the introduction of penicillin, resistance to the antibiotic was noted [57]. The rise of antibiotic resistance has posed a tremendous burden on healthcare system [58], and biofilms play a potentially large role in contributing to this health crisis.

Treatment of biofilm-related diseases requires an underlying understanding of the mechanisms of biofilm antibiotic resistance. In general, antibiotics attack bacteria by disrupting bacterial cell wall biosynthesis, protein synthesis, as well as DNA replication and repair. It has been recognized that bacterial cells withstand these attacks through efflux pumps, deactivating enzymes, and reprogramming antibiotic targets [59]. Efflux pumps on bacterial cell membranes pump out the antibiotics to keep the antibiotic concentration low inside bacterial cells. Enzymes deactivate antibiotics by modifying their active components. Antibiotics targeted structures are modified to avoid recognition [57].

However, these mechanisms are understood in planktonic bacteria, and there exists drastic difference between antibiotic resistance in the biofilm bacterial mode of growth and that of planktonic cells, indicating that the mechanisms behind them are dissimilar. It is unwise to confer the understanding about planktonic bacteria antibiotic resistance to biofilm bacteria without discretion. Although the mechanisms mentioned above may be present in biofilms, they likely work in synergy with multiple mechanisms that are distinct to the biofilm mode of growth. Although the exact mechanisms are still under investigation, several mechanisms have been hypothesized to explain the antibiotic resistance of biofilms. A few of the most prominent mechanisms are mentioned here, but a more thorough review written by Stewart [60] is available. It was originally suggested that the matrix structure of the biofilm may serve as a physical barrier to prevent antibiotics from diffusing inside; however, more recent studies have proved that the diffusion of antibiotics is not hindered by the biofilm matrix [61], [62]. An additional hypothesis is that after binding to polysaccharides, proteins, and DNA present in the biofilm EPS, the antibiotics reaching the target site may no longer be bioactive or may not achieve an adequate concentration to effectively kill bacteria [63]. Genetic components from lysed bacterial cells, such as plasmids, are retained inside the matrix, increasing the frequency of gene transfer between bacterial cells [53]. These plasmids may contain the genes beneficial to bacteria, such as genes for antibiotic resistance [64]. Umadevi et al. [65] have reported that most β-lactam antibiotics can be hydrolyzed by plasmid-encoded carbapenemases, furthering the evidence for gene transfer-related resistance mechanisms. Furthermore, at the interior of the biofilm, the bacteria are anaerobic and have very limited access to oxygen and nutrients, making antibiotics whose action relies on disrupting metabolic processes much less effective in bacterial killing [66]. For instance, penicillin interrupts bacterial cell wall synthesis, but if bacterial cells in the interior of the biofilm are not actively growing or synthesizing cell wall, then penicillin has no effect. In addition, bacterial cells inside biofilms may be at different growth phases; therefore, a subset of the bacterial population can survive from cell division-directed antibiotic attacks. According to persister theory, a small portion of bacterial cells become dormant cells that are tolerant to antibiotics, which then contribute to incomplete bacterial killing by interventions [2], [67], [68]. Quorum-sensing, a mechanism that bacteria use to control gene expression by sensing the local cell density [69], plays a role in biofilm formation [70], [71], [72], [73]. It is still unclear to what degree quorum-sensing is involved in biofilm antibiotic resistance, but the involvement of quorum-sensing in efflux pump genes has been reported by Maseda et al. [74]. A more recent review summarized the evidence that may indicate the relationship between quorum-sensing and antibiotic resistance [75]. Although only a few of the numerous hypotheses are mentioned here, it is readily apparent that biofilm antibiotic resistance is a complex, formidable challenge.

2.3 Nanoparticles in the treatment of biofilms

To combat bacterial infections, NPs made of polymers, silica, and lipids can be used as drug delivery carriers to potentiate current antibiotics, or metallic NPs can act as antimicrobial agents. When used as drug delivery carriers, NPs are expected to protect antimicrobial agents from deactivating enzymes such as β-lactamases, prevent electrostatic binding to EPS components such as DNA or polysaccharides, and release antimicrobial agents locally in a sustained and controlled manner to lower the incidence of systemic side effects as well as enhance antimicrobial efficacy. Consequently, NPs can improve the bioavailability and targeted delivery of antibiotics [76]. The treatment strategies of various NP formulations for biofilm-associated infections are summarized in Figure 1 [23], [33], [37], [42], [43], [44], [77], [78], [79], [80], [81], [82], [83].

Figure 1:

Applications of various nanoparticle treatments to either prevent, disrupt and inhibit, or disperse bacteria from biofilm infections.

2.4 In vitro biofilm quantification

In order to study the anti-biofilm efficacy of the NPs designed, it is necessary to understand the methods that can be used to quantify biofilms. Common methods used to assess the anti-biofilm effect of NPs are summarized in this section. The simplest method is the use of the bacterial growth rate, which can be measured by reading the optical density (OD) at 590 nm based on the fact that higher bacterial cell density causes more lightscattering.

Live/dead staining is a well-established method to evaluate bacterial cell viability. Bacterial cells with intact or damaged cytoplasmic membranes are stained with SYTO 9 (fluorescent green) and propidium iodide (fluorescent red) [84]. This method is time consuming, and in general, only a small portion of biofilm cultures can be examined using confocal laser microscopy. In order to investigate the overall biofilm bacterial cell viability after treatments of NPs with or without an alternating magnetic field, metabolic assays, such as the XTT assay and Alamar Blue assay, can be used to estimate bacterial viability as an indirect measurement [85].

The effect of NPs on biofilm formation can be assessed by staining biofilms with crystal violet and measuring the absorbance. Crystal violet stains the entire biomass, including extracellular matrixes as well as live and dead biofilm bacterial cells [86]. As EPS accounts for almost 90% of the biomass, the bacterial cell viability cannot be measured using crystal violet. In this case, a combination of live/dead staining and crystal violet staining is used to more thoroughly characterize in vitro biofilms. Additionally, because crystal violet sticks to tissue culture plastic, a blank control is critical. An alternate method for measurement of biomass quantifies the thickness of fluorescently-labeled biofilms using confocal microscopy.

3.1 Silver nanoparticles

The disinfectant effect of silver compounds has been recognized since antiquity [87]. However, with the invention of penicillin, the era of antibiotics began and silver was largely forgotten. Now, with increasing concern about antibiotic resistance of bacterial infections and biofilms, silver’s anti-microbial capacity has again become a topic of interest to researchers, primarily in the form of silver NPs. Moreover, silver NPs are a broad spectrum antiseptic that are effective against Gram-negative and Gram-positive bacteria as well as fungi and viruses [88]. Silver NPs typically range from 1 to 100 nm in diameter, and they are composed of 20 to 15,000 silver atoms [89].

Silver NPs can be prepared through a number of methods. The most common chemical method reduces Ag⁺ (e.g. AgNO₃) to Ag⁰ using a reducing agent such as sodium borohydride in the presence of a stabilizer like poly(N-vinyl-2pyrrolidone) to keep silver NPs from aggregating. Reducing agents and stabilizers must be selected carefully to limit reactivity and environmental impact while increasing particle stability [32]. More environmentally friendly ways to synthesize silver NPs include the use of living organisms like bacteria to produce the particles. For example, AgNO₃ solution can be added to wet Bacillus licheniformis in a flask, and the mixture can be left on a shaker at 37°C for 24 h to produce particles. Cells are removed, and the silver NPs are purified [32]. Another method to prepare silver NPs with low environmental impact makes use of a biopolymer such as a starch as the stabilizer [79]. A solution of silver NPs in corn starch maintained colloidal stability (no particle aggregation) up to 90 days, as analyzed by absorption. Eighty-eight percent and 85% of Pseudomonas aeruginosa biofilm reduction were observed after 48 and 24 h of treatment, respectively, indicating an effective impedance of biofilm formation with these starch-stabilized particles [79].

Habash and colleagues [18] assessed the response of P. aeruginosa biofilms to treatments with citrate-capped silver NPs of various sizes (10 nm–100 nm), the antibiotic aztreonam, and the combination of silver NPs and antibiotics. Particle size was determined to be critical after treating preformed biofilms with silver NPs with various sizes at different concentrations. Habash et al. reported that silver NPs with small sizes (10 nm and 20 nm) were more effective than 100 nm particles with regard to biofilm biomass and biofilm bacterial cell viability reduction. However, no morphological change in biofilm structure or cellular morphology was observed [18]. Radzig and colleagues [87] reported that when 8.3 nm silver NPs were administered to Escherichia coli biofilms at an inhibitory concentration (for planktonic bacteria), the NPs failed to have an effect; however, when administered at a high concentration (100–150 μg/ml), the silver NPs killed most E. coli biofilm bacteria. Separately, Jena and colleagues [30] found that chitosan-coated silver NPs (sizes from 55 nm to 275 nm) did not disrupt 24-h-old P. aeruginosa and Staphylococcus aureus biofilms, but they did inhibit P. aeruginosa biofilm formation by 65%. A similar inhibitory effect on P. aeruginosa biofilm was also reported by Palanisamy et al. [31] using chemically synthesized silver NPs (20 nm–30 nm).

Several researchers have attempted combinatorial therapeutic strategies with both antibiotics and silver NPs. Interestingly, in one study, the antibiotic aztreonam administered alone did not eradicate biofilms and even stimulated biofilm growth, up to 250% of untreated biomass, after treatment was stopped [18]. Cationic antimicrobial peptides (CAMPs) are very promising in the treatment of multidrug-resistant P. aeruginosa lung infections experienced by CF patients. Nevertheless, there is an urgent need of inhalable formulations able to deliver the intact CAMP in conductive airways and to shield its interactions with airway mucus/bacterial biofilm, thus enhancing CAMP/bacteria interactions. Along these lines, the aim of this work was the design and development of nano-embedded microparticles (NEMs) for sustained delivery of CAMPs in the lung [23]. To this purpose, NPs made of poly(lactide-co-glycolide) (PLGA) containing a model CAMP, colistin (Col), were produced by emulsion/solvent diffusion technique. Engineering NPs with chitosan (CS) and poly(vinyl alcohol) (PVA) allowed to modulate surface properties and, in so doing, to improve NP transport through artificial CF mucus. In order to achieve a long-term stable dosage form useful for NP inhalation, NPs were spray-dried in different carriers (lactose or mannitol), thus producing NEM. The most promising NEM formulations were selected on the basis of bulk and flow properties, distribution of NPs in the carrier, and aerosolization performance upon delivery through a breath-actuated dry powder inhaler. Of note, selected Col-loaded NEMs were found to kill P. aeruginosa biofilm and to display a prolonged efficacy in biofilm eradication compared to the free Col. This effect was likely ascribable to the ability of NPs to penetrate into bacterial biofilm, as demonstrated by confocal analysis, and to sustain Col release inside it. Taken altogether, our results indicate that adequate engineering of PLGA NPs represents an enticing technological approach to harness novel antimicrobials for P. aeruginosa lung infection, such as CAMPs, especially in CF [23]. However, the combination of small silver NPs and aztreonam demonstrated rather promising results. The highest biomass reduction (~98%) and thickness reduction (~50%) as well as bacterial cell wall damage was observed after exposure to a combination of 10 nm silver NPs and aztreonam [18]. Furthermore, significantly higher reactive oxygen species (ROS) production was observed when silver NPs were combined with an antibiotic in comparison to the sole implementation of either agent alone [90].

Although the mechanism behind silver NPs’ antimicrobial ability is still unclear, it is generally believed that damage to the bacterial cells in the presence of silver NPs, as with other metal NPs, to a high degree, is induced by ROS such as hydroxyl radicals, superoxide anions, and hydrogen peroxide, which may cause damage to protein and DNA in bacteria [91]. Silver ions released from the silver NPs tend to react with thiol groups in protein and deactivate key respiratory enzymes to result in toxic ROS levels that damage and kill the cells [92]. It is also reported that the ROS generated on the surface of silver NPs can interact with cell membrane lipids, leading to the disruption of cell membrane functions [91]. Silver NPs are also known to be capable of binding and penetrating the bacterial cell wall, thus increasing cell wall permeability and cell death [93], [94]. Thus, the combination of silver NPs and conventional antibiotics appears promising for biofilm treatment.

3.2 Gold nanoparticles

Unlike silver, gold is not known to have intrinsic antimicrobial properties. However, the properties of nanoscale gold allow for robust particle functionalization, and researchers have investigated the possibility of using gold NPs in biofilm treatment. Like silver, gold NPs are most commonly synthesized by the reduction of gold salts via several well-established methods [43], [44], [76].

Using an antimicrobial plant extract from Plumbago zeylanica, silver NPs (~63 nm), gold NPs (aggregated into gold nanotriangles), and silver-gold bimetallic NPs (~93 nm) were synthesized from the bioreduction of silver nitrate (AgNO₃) and chloroauric acid (HAuCl₄) [45]. Gold NPs showed 97% S. aureus biofilm inhibition and disrupted 24-h-old S. aureus biofilms up to 95% and Acinetobacter baumannii biofilms up to 40%. Silver and bimetallic NPs outperformed gold NPs in all strains tested in terms of inhibition and disruption [45]. The anti-biofilm efficacy of gold-silver bimetallic NPs synthesized from γ-proteobacterium Shewanella oneidensis MR-1 was investigated by Ramasamy et al. [46]. E. coli, P. aeruginosa, Enterococcus faecalis, and S. aureus cultures were incubated with gold-silver bimetallic NPs (~209 nm) or gold NPs (~57 nm) for 24 h and further examined by crystal violet to determine biofilm inhibition. Gold-silver bimetallic NPs inhibited all strains effectively at 250 μM. E. coli biofilm formation was completely inhibited at particle concentrations as low as 10 μM [46].

The photosensitizer methylene blue (MB) was attached onto the surface of gold NPs (~26 nm after conjugation) to treat 24-h-old Candida albicans biofilms [39]. Photosensitizers can be activated under a light source and release highly ROS to damage bacteria. The conjugation was achieved through electrostatic interactions between negatively charged gold NPs and positively charged MB. After being treated with a 660 nm diode laser at 38.2 J/cm² for 40 s every 6 h during a 24-h treatment, C. albicans biofilm growth was inhibited up to 63.2% by 20 μg/ml MB alone and 82.2% by MB-conjugated gold NPs containing an equivalent amount of MB, based on crystal violet staining results. The 2,3-Bis (2-methoxy-4-nitro-5-sulfophenyl)-5-([phenylamino] carbonyl)-2H-tetrazolium hydroxide (XTT) biofilm reduction assay revealed that 81.9% and 95.4% of the biofilm bacteria were dead under the same treatment conditions as above. Moreover, the structure of the MB-conjugated gold NP-treated biofilms was highly distorted compared to the untreated control biofilms when examined by SEM. Additionally, it was also found that MB increased bacterial cell membrane permeability [39]. Under the appropriate activating conditions, photosensitizers like MB can generate highly toxic ROS that cause bacteria cell death [95].

The use of gold NPs improved the anti-biofilm efficacy of the photosensitizer MB, and it is of continuing interest to investigate whether gold NPs can elicit similar results with photosensitizers such as rose bengal (RB) and erythrosine (ER). Interestingly, the chemically synthesized gold NPs failed to inhibit or disrupt A. baumannii and E. coli biofilms. Also, the chemically synthesized silver NPs and bimetallic NPs had poor performance in both biofilm inhibition and disruption [45]. The authors concluded that the anti-biofilm efficacy of gold NPs reported was due to the antimicrobial plant extract synthesis. While results with gold NPs are promising, further research into using gold NPs alone or as part of a combinatorial therapy is warranted.

3.3 Iron nanoparticles

Magnetic NPs have been studied extensively for use in the biomedical field due to their multifunctionality. They have the potential to facilitate magnetic resonance imaging (MRI), targeted drug delivery, and hyperthermia-based therapies [96]. In addition, magnetic NPs are considered biocompatible and are inexpensive.

Magnetic NPs can be synthesized through either co-precipitation or thermal decomposition. In the co-precipitation method, a base is added into an aqueous mixture of Fe³⁺ and Fe²⁺ salts at room temperature or elevated temperature in an oxygen-free environment, and a black magnetite precipitate is formed [97]. Co-precipitation is convenient to carry out and provides a considerable amount of magnetic NPs in a stable colloidal state [98]. However, synthesizing NPs with a uniform size remains a challenge using the co-precipitation method because the size distribution can be affected by a variety of parameters including temperature, pH, salts used, and concentrations [99].

Superparamagnetic iron oxide NPs (SPIONs) were used by Durmus et al. and Taylor et al. to treat methicillin-resistant S. aureus (MRSA) [43], [44]. The SPIONs were coated with dimercaptosuccinic acid (DMSA) before being chelated with Fe³⁺ from FeCl₃ or Zn²⁺ from ZnCl₂. DMSA is a chelator that is able to improve the solubility of SPION and conjugate with metal ions [100]. Twenty-four-hour-old S. aureus biofilms were treated with iron- or zinc-conjugated SPION or DMSA-coated SPION for 24 h in vitro. The OD as well as crystal violet staining revealed that zinc-conjugated SPION inhibited the biofilm growth effectively (~20% compared to control) at 0.01 mg/ml, the lowest concentration tested; all SPION inhibited biofilm growth at 1 mg/ml. On the contrary, the application of ZnCl₂ alone actually stimulated biofilm growth. Combined with the result that SPIONs, as imaged by TEM, were found inside 48-h-old biofilm bacteria cells after 2 h of treatment, the authors concluded that SPION delivered the zinc and iron ions into biofilm bacteria cells where chemical balance was interrupted and eventually killed the biofilm bacteria [43].

Fructose was used along with the metal-conjugated SPION mentioned above by Durmus et al. to treat 24-h-old S. aureus biofilms for 24 h [44]. The anti-biofilm efficacy of SPION conjugated with or without metal ions was enhanced significantly when fructose was present. The use of fructose increased the amount of SPION deposited in the biofilms, leading to more ROS generation. Unconjugated SPION and 1 mM or 4 mM iron-conjugated SPION achieved a killing rate of up to 99.99999% when 1 mM fructose was present. In fact, all SPIONs, conjugated or not, performed better than did vancomycin. This result was confirmed with fluorescence microscopy images of cells with live/dead staining. Moreover, biomass was reduced by almost 50% after treatment with iron-conjugated SPION and fructose, similar to the reduction caused by vancomycin. Even 0.1 mg/ml SPION achieved a similar biofilm eradication effect as 1 mg/ml SPION alone when used with 100 mM fructose, indicating that the use of fructose can strengthen the anti-biofilm efficacy of SPION. In turn, this reduces the potential for systemic toxicity of SPION because a lower dose of SPION is required for a therapeutic effect. Promising results were also observed on Gram-negative biofilms such as P. aeruginosa and E. coli, suggesting a broad-spectrum anti-biofilm efficacy [44]. Salunke et al. [45] also reported that metal NPs reduced by fructose containing plant extract outperformed the chemically synthesized NPs.

Magnetic NPs, when directed using an external magnetic field, are capable of penetrating the full depth of biofilms, accumulating at the bottom, and killing bacterial cells deep inside biofilms. SPIONs were decorated with carboxyethylsilanetriol (CES) by Subbiahdoss et al. [40] to treat 24-h-old gentamicin-resistant Staphylococcus epidermidis biofilms. The graft of CES did not change the size of SPIONs (~13 nm) but lowered the zeta potential from +43 mV to −15 mV. With a magnet placed under the culture well where biofilms were growing, CES-SPIONs could be concentrated at the center of the wells. The cross-sections of live/dead stained biofilms further demonstrated that bacterial cells at the bottom of the biofilms accounted for most of the dead bacteria. The use of gentamicin did not change the anti-biofilm efficacy of SPIONs. Thirty-eight percent of mature biofilm bacteria cells were killed when treated with magnetically targeted CES-SPIONs, while only 9% were found dead when the magnetic field was not applied [40]. As there was no significant difference between the anti-biofilm efficacy of unmodified SPIONs and CES-SPIONs, it is possible that the ROS formation, caused by the extremely small size of NPs, attributes more to the anti-biofilm efficacy compared to the electrostatic interaction between SPIONs and biofilm bacterial cell walls [40].

Magnetic NP induced hyperthermia also facilitates anti-biofilm treatments. Up to 99.9% reduction was observed when planktonic S. aureus was heated in tryptic soy broth (TSB) for 20 min at 60°C [101]. When magnetic hyperthermia was implemented, a lower temperature (43°C) was needed to cause the same 99.9% reduction on S. aureus biofilms [14]. As for P. aeruginosa biofilms, when magnetic NP-mediated hyperthermia created a temperature rise of 5°C or more, the dispersion of preformed biofilms was induced. The hyperthermia did not show any toxic effects on the bacteria themselves but enhanced gentamicin treatments [102]. This conclusion was not unexpected, as planktonic bacteria are far more susceptible to conventional antibiotics than biofilm bacteria; therefore, the hyperthermia rendered the dispersed bacteria more susceptible to antibiotic treatment.

With their use as targeted delivery agents, magnetically targeted particles, or particles for inducing hyperthermia, iron oxide particles hold great promise. However, further investigation of the effect of size and chemical composition, as well as other formulation characteristics, is necessary. For example, in the case of Fe₂O₃ iron oxide NPs, 2 nm particles did not inhibit biofilm growth but actually induced its formation. It was hypothesized by Borcherding et al. that these very small Fe₂O₃ NPs may have been an iron source for biofilm bacteria growth in this study. In addition, Fe₂O₃ NPs were found to inhibit antimicrobial peptide function and deactivate ciprofloxacin [103], [104]. Additional studies into the effect of iron oxide NPs are necessary to understand their usefulness in biofilm treatment.

3.4 Copper nanoparticles

Copper NPs exhibit characteristics favorable to their utilization as therapeutic agents with antimicrobial/antifungal properties and have a lower cost compared to other metals such as silver and gold [105]. Copper NPs have found applications as coating agents on biomedical devices to inhibit the growth of pathogens [106], [107], [108]. Copper NPs can be used as industrial fungicides, in water purification, and as antifouling agents [107], [109]. A variety of research has been performed on the effect of copper NP use on the formation of biofilms based on their cell counts and ability to be cultured [110]. Given their ability to inhibit the formation of biofilms and limit the growth of pathogens, copper NPs demonstrate potential for their use in hospital and other healthcare facilities to prevent spreading of pathogens on equipment and nosocomial infections in patients.

A variety of methods for the synthesis of copper NPs exist in literature including the use of polyol synthesis, thermal plasma, inert-gas condensation, one-pot synthesis, and the solvothermal method [30], [32], [89], [91], [92]. Nonetheless, chemical reduction frequently appears as a prominent method for the synthesis of copper NPs. Generally, chemical reduction is a simple and economical method to produce metal NPs given the use of a reducing agent to reduce metal salts and a capping agent to curtail NP synthesis at a particular size. Due to copper’s sensitivity to air, copper NPs often form a surface oxide layer during synthesis [109], but some methods exist that utilize an environment void of oxygen in order to inhibit the formation of this layer. As with the other metallic NPs, the plethora of synthesis methods available provides flexibility in the ability of a facility to produce copper NPs and may allow for scaled-up manufacturing of copper NPs [105].

Many of the aforementioned studies utilized NPs in various experimental settings to assess the practical application of copper NPs as inhibitors against biofilm formation. The proliferation of pathogenic specimens quantifies the change in bacterial counts on agar plates after treatment with copper NPs [111]. This experiment involved the extraction of biofilm bacteria from the slurry of the tooth crown from a healthy, young male. The slurry was then suspended in phosphate-buffered saline (PBS) and dosed with 40 nm cupric oxide NPs at concentrations of 10, 50, or 100 μg/ml for 16 h at 37°C. These cultures were subsequently spread on agar plates, incubated for 3–5 days, and analyzed using crystal violet under light microscopy. The images showed that EPS biomass was inhibited by almost 61% when bacteria were treated with 50 μg/ml cupric NPs [111].

Shankar and Rhim [112] devised an environmentally friendly method to produce copper NPs through chemical reduction. They investigated the effect of copper salts and reducing agents on the size and shape of copper NPs and subsequently determined the effect of copper NPs on foodborne pathogens including the Gram-positive bacterium Listeria monocytogenes and the Gram-negative bacterium E. coli. Using two types of reducing agents, sodium hydroxide (NaOH) and ascorbic acid (AA), and three types of copper salts, they compared the effects of particles prepared with each reducing agent against the two types of bacteria through the use of an agar well diffusion method where zones of inhibition were measured. L. monocytogenes (10.5 mm–12.5 mm) were found to exhibit a greater susceptibility toward copper NPs compared to E. coli (11.7 mm–18.5 mm) due to a greater zone of growth inhibition. Additionally, NPs prepared using sodium hydroxide induced a more potent effect against both L. monocytogenes and E. coli than those synthesized using ascorbic acid [112].

LewisOscar et al. [107] assessed how antibiotic resistance of P. aeruginosa increases in more mature specimens and described how copper NPs help prevent the formation of biofilms at more nascent stages. The methodology employed utilized two control strains of P. aeruginosa as well as 17 other clinical isolates grown in Luria-Bertani (LB) agar at 37°C for 24 h. This study reported a maximum biofilm inhibition of 94% and exemplified the use of copper NPs as a cheaper alternative to silver NPs to impede the formation of biofilms [107]. Based on the current results, copper NPs appear to be best suited for biofilm prevention rather than biofilm eradication.

3.5 Liposomes

Biocompatible amphiphilic phospholipids form spherical vesicles readily, which are composed of one or more lipid bilayers. Liposomes can be categorized into multilamellar and unilamellar classes depending on the number of lipid bilayers, and they are capable of encapsulating hydrophilic or hydrophobic compounds [113]. High encapsulation efficiency synthesis methods of liposomal NPs include the Bangham, detergent-depletion, ether/ethanol injection, reverse phase evaporation, and emulsion methods [114]. The bilayer membrane structure grants liposomes exceptional permeability, which facilitates the cellular uptake of liposomes. Hydrophilic antimicrobial agents can be protected from deactivating factors in vivo by encapsulation into the large aqueous core of the liposomal NP with high drug loading.

The efficacy of the liposomal formulation of the antibiotic amikacin (Arikace^®), a nebulized solution for CF treatment, was tested in vivo on P. aeruginosa biofilms by Meers et al. Uninfected rats were subject to inhalation of the same amount of free tobramycin and liposomal amikacin (~300 nm) for 60 or 80 min [115]. About 65% of the amikacin remained in liposomal form after deposition in the lung at an amount of 782 μg compared to 645 μg of free tobramycin, which is likely due to the fast clearance of the free aminoglycoside throughout the inhalation process. The free tobramycin only lasted fewer than 2 h until it was mostly removed from the lungs, while the liposomal amikacin exhibited a sustained drug release for up to 175 h after a burst release due to the transformation of liposomal amikacin into free form during inhalation. Nevertheless, the sustained release of amikacin lasted for only 48 h when incubated with diluted P. aeruginosa sputum from a patient with CF, which led to the speculation that the lytic components of biofilms may trigger the release of amikacin. Almost all amikacin was released over 24 h when liposomal amikacin was incubated with Di-rhamnolipid (RHL) and mono-rhamnolipid, indicating amikacin is more readily released at biofilm-infected areas. Rats infected with P. aeruginosa for 4 days were treated with equal doses of free amikacin and liposomal amikacin three times per week for 2 weeks. No obvious change of colony-forming units (CFUs) was seen in the free amikacin-treated group, while liposomal amikacin reduced CFUs by 40%. Over 1300 μg of amikacin was detected 3 days after treatment, indicating a sustained release of amikacin [115].

As the biofilms in CF patients are surrounded by thick mucus, it is crucial to deliver a sufficient amount of antimicrobial agents through the mucus layer to the biofilm-infected sites, which is a size-dependent process. In a mixture with 1 μm polystyrene particles, 62% of fluorescent liposomes penetrated a layer of sputum in a period of 24 h as well as 60-h-old pure P. aeruginosa biofilms. In contrast, only 9% of the 1 μm polystyrene particles were recovered at the other end of the sputum layer, yet even a few 200 μm polystyrene particles were found to penetrate the sputum [115].

This size-dependent penetration was also mentioned in a separate study. Cationic, smaller sized (128 nm–141 nm), unilamellar particles penetrated both P. aeruginosa and S. aureus biofilms at a much higher rate with deeper penetration than larger, multilamellar particles [34]. The smaller, blank, cationic liposomes inhibited the growth of P. aeruginosa biofilms by 75% based on relative viability after 24-h exposure, and S. aureus biofilms were inhibited by 43%. It is hypothesized that the unilamellar cationic liposome particles may disrupt the electrostatic equilibrium of the colonies within the biofilm, allowing the anti-biofilm effects to be more potent [34].

Neutral and anionic liposomal tobramycin were prepared by dissolving a mixture of cholesterol and 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) and a mixture of 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) and 1,2-dipalmitoyl-sn-glycero-3-phosphoglycerol sodium salt (DPPG) in chloroform, respectively. However, neither liposomal tobramycin performed better than free tobramycin in this study [116].

Liposomal antibiotics have been studied as an alternative to traditional free antibiotics that only deliver low concentrations at the site of biofilm infections and are cleared quickly in the lungs. However, the clinical application of liposomal formulations is limited by a liposome’s weak physiochemical properties. Meer et al. reported that 35% of the amikacin did not arrive at the lungs in liposomal form in vivo. The side effects caused by the leakage of encapsulated antibiotics from liposomes are potentially unfavorable. Despite the promising results reported by Meer et al. there is still doubt as to whether liposomal antibiotics can penetrate the harsh conditions of the biofilm biomass in infected patients.

3.6 Chitosan nanoparticles

Chitosan is a natural polymer deacetylated from chitin. Its biocompatible, biodegradable, and inherent broad-spectrum antimicrobial properties make it suitable as a carrier for biofilm treatment [117]. Chitosan NPs have been effectively used against 24-h-old P. aeruginosa biofilms of six different strains derived from clinical cases [118]. Chitosan NPs carry a positive surface charge, while bacterial cell walls and biofilm EPS surfaces are negatively charged. Chitosan NPs are thus expected to have a higher affinity to infected areas than negatively charged NPs. One common method used to synthesize chitosan NPs is the ionic gelation method. Amine groups present in chitosan acetic acid solution react with polyanions like sodium tripolyphosphate (TTP) readily, and chitosan NPs are then formed through cross-linking [119], [120]. Preparing chitosan NPs for biofilms at neutral pH was proposed by Chávez de Paz et al. [28]. The anti-biofilm effect was examined through treatment of 24-h-old S. mutans biofilms with low molecular weight chitosan NPs. More than 95% of the biofilm bacterial cells were damaged after treatment, although no significant biofilm structure change was observed [28].

The photosensitizer ER was encapsulated into chitosan NPs to facilitate photodynamic therapy by Chen and coworkers [27]. ER-loaded chitosan NPs were used to treat S. mutans, P. aeruginosa, and C. albicans biofilms in vitro. Gram-positive S. mutans biofilms were eradicated completely based on viable cell count after incubation with ER-loaded chitosan NPs for 12 h at room temperature in the absence of light followed by 50 J cm⁻²dose. Under the same conditions, NP-treated Gram-negative P. aeruginosa biofilm cells were reduced only by around 28%, indicating a less effective treatment [27].

Another photosensitizer, RB, was conjugated onto chitosan NPs by carbodiimide chemistry to treat 24-h-old E. faecalis biofilms [42]. RB-loaded chitosan NPs increased the cell membrane permeability followed by cell content leakage after 15 min of incubation in a dark environment, which can be attributed to chitosan’s antimicrobial properties. Encapsulated RB showed a significantly higher cell uptake compared to free RB. It is worth noting that encapsulated RB eliminated biofilm completely when photodynamic therapy was divided. Two dosages of 20 J/cm² were more effective than one 60 J/cm² dosage. Although the time interval between the two dosages was not mentioned in this paper, it is likely that molecular oxygen was replenished during this time, enhancing the production of toxic highly reactive species around and inside the biofilm. Chitosan NPs also help slow down the release rate of singlet oxygen, elongating the bacterial killing effect [42].

In another study, linoleic acid modified chitosan NPs were used to immobilize the enzyme β-N-acetyl-glucosaminidase (DspB), which has been shown to be capable of inhibiting and detaching biofilms by degrading poly-β(1,6)-N-acetyl-glucosamine (PNAG), a major polysaccharide found in S. epidermidis, S. aureus, and Actinobacillus actinomycetemcomitans EPS [29]. At 37°C, free enzyme lost its activity completely after incubation for 7 h, yet encapsulated enzyme still possessed 33.4% activity. Immobilized enzyme was more stable upon temperature variation, which may result from stabilization of encapsulated enzyme into the chitosan NPs. Chitosan NPs showed a slightly anti-biofilm effect in biofilm treatment, and biofilm treated with immobilized DspB had the lowest absorbance reading at 590 nm to measure the amount of crystal violet after 1-h treatment. Longer treatment may result in biofilm eradication because immobilized enzyme remained active for longer than 7 h [29].

The biocompatibility, degradability, and antimicrobial nature of chitosan make it a strong candidate for biofilm antibiotic delivery. However, the hydrophobicity of chitosan prompts aggregation under biological condition and limits its use. Aggregation of chitosan NPs was observed after an hour in cell medium [121]. The addition of poly(ethylene glycol) (PEG) to the surface of chitosan NPs has been shown to be an effective way to overcome this drawback [122], which will likely have implications for the future of this drug delivery technology.

3.7 Poly(lactic-co-glycolic acid) nanoparticles

Poly(lactic-co-glycolic acid) (PLGA) is a biocompatible and degradable copolymer. The ester linkages in PLGA degrade through hydrolysis when water is present, and the polymer degradation products lactic acid and glycolic acid are natural byproducts of several metabolic pathways. Several drug delivery devices made of PLGA are FDA approved. A list of FDA-approved PLGA-based particles on the market is given elsewhere [123].

The use of PLGA NPs as a colloidal drug delivery system has been explored extensively. Numerous studies have shown that PLGA NPs possess the ability to deliver therapeutic agents in a controlled and sustained manner through polymer degradation and diffusion of therapeutic agents encapsulated while the degradation rate of PLGA NPs can be manipulated by changing the ratio of poly lactic acid (PLA) and poly glycolic acid (PGA), which controls the material’s hydrophobicity and crystallinity. Higher PGA ratios lead to higher crystallinity, while more PLA increases hydrophobicity [124]. Unmodified PLGA NPs, capped with COOH, possess a negative zeta potential because of the disassociation of the hydrogen of the carboxyl group in an aqueous environment. In the case of OH-capped PLGA NPs, the deprotonation of the hydroxyl group only happens under very basic conditions, which is not considered relevant for most biological applications. Generally, the size of PLGA NPs ranges from 100 nm to 400 nm. Longer sonication time during fabrication process can lead to a smaller size [125].

Ciprofloxacin (CPX) was encapsulated into PLGA NPs coated with poly(L-lysine) (PL) using an oil-in-water nano-precipitation method by Baelo et al. [35]. The PLGA core was functionalized with deoxyribonuclease I (DNase I) to facilitate the eradication of biofilm by degrading the extracellular DNA in EPS that is a key component in early stages of biofilm formation and abundant in EPS. Burst release and a slower sustained release after that were observed in an in vitro release study. PLGA-PL-CPX-DNase I, PLGA-CPX, PLGA-PL-CPX NPs, and free CPX were compared in terms of antimicrobial ability, prevention of biofilm formation, and eradication of preformed biofilm. The growth rate of biofilm was inhibited by PLGA-PL-CPX-DNase I NPs up to 80% at the minimum CPX concentration of 0.0078 μg/ml. The biomass of 2-day-old biofilm reduced by 95% after treatment with PLGA-PL-CPX-DNase I NPs. These results indicate that active DNase I on the surface of PLGA NPs helps increase the mobility of the particles and facilitate diffusion into the EPS [35].

In another study, gentamicin was loaded into PLGA NPs by Abdelghany et al. using a modified water-in-oil-in-water double emulsion to treat P. aeruginosa biofilms [37]. By adjusting the pH of the outer aqueous phase, drug-loading increased from 6.1 μg/ml to 22.4 μg/ml in comparison with a similar work done by Concepcion in 2006. Increased pH made gentamicin less hydrophilic, leading to higher entrapment in the hydrophobic PLGA core. After a burst release in the first 24 h, sustained release of gentamicin from PLGA NPs lasted for another 15 days at 37°C incubation. The existing biofilm was placed in the external chamber in a dialysis cell, and gentamicin-loaded PLGA NPs and free gentamicin were added into dialysis tubing, respectively. Half of the media was replaced with fresh media every hour to mimic the typical clearance rate of the lung. After 36 h of incubation, the CFU in the NP-treated biofilm was significantly lower than the free gentamicin-treated biofilm, suggesting that a sustained release of gentamicin from PLGA NPs is more effective in biofilm treatment than a single dosage of gentamicin. Gentamicin-loaded PLGA NPs and free gentamicin were delivered intraperitoneally to P. aeruginosa-infected mice. Peritoneal lavage was done in infected mice at certain time points, and CFUs were measured in each aspiration. After 48 h, a similar reduction was shown in CFU from both NP-encapsulated gentamicin and free gentamicin-treated mice. After 96 h, NP-encapsulated gentamicin managed to keep CFU levels low in the peritoneal lavage while the mice treated with free gentamicin returned to the same level as the positive control treated with saline buffer. The result of the in vivo test corresponds to the in vitro biofilm treatment results, indicating that these PLGA NPs improved the effectiveness of gentamicin by maintaining a working concentration of gentamicin at biofilm-infected areas [37].

Ciprofloxacin and magnetic NPs (CMX) were encapsulated into PLGA NPs by Xin and co-workers [36]. An external low-energy oscillating magnetic field was used to induce hypothermic conditions, which led to the disruption of the PLGA core. An external magnetic field was turned on and off alternatively for 6 h every other hour during the drug release experiment. The results showed that the CPX release rate surged when a magnetic field was on and plateaued when the magnetic field was off, suggesting a controlled release. CPX-CMX-loaded PLGA NPs released CPX for 14 days after being triggered by the magnetic field, and the total release amount was 95%. Mature biofilm was treated with CPX-CMX-loaded PLGA NPs, blank PLGA NPs, and free CPX to test the in vitro functionality of released CPX. XTT assay was conducted to measure the metabolic activity of biofilms after a 24-h treatment. The reduction of biological activity caused by triggered CPX-CMX-loaded NPs was 33.3%, slightly lower than free CPX due to the incomplete release of CPX [36].

Cationic antimicrobial peptides (CAMP) area promising treatment for P. aeruginosa lung infections. CAMP can cause disruption of Gram-negative cell walls by the interaction with anionic lipopolysaccharide on those cell walls. However, its cationic nature causes CAMP to react with anionic proteins and other polysaccharides of the lung extracellular environment, leading to poor bioavailability. Colistin, a model cationic antimicrobial peptide, was encapsulated into PLGA NPs coated with chitosan or PVA to achieve desired properties to overcome the mucus barrier and penetrate biofilms [23]. Colistin-loaded PLGA NPs were then embedded into water-soluble inert lactose microparticles using a co-spraying drying process to ensure enough NPs can be deposited onto lung-infected areas without being exhaled because of their small size. Both the size and zeta potential of CS-PLGA and PVA-PLGA NPs stayed consistent after the lactose coating was dissolved in water, suggesting that the lactose carrier did not significantly affect the properties of PLGA NPs. PVA-PLGA NPs showed a much slower burst release in the first 6 h compared to CS-PLGA NPs, yet both NPs released peptide for up to 15 days. Similar to other antimicrobial agents, free colistin had a robust anti-biofilm performance in the first 24 h, reducing 90% of biomass. Nevertheless, for high concentrations used in the biofilm treatment, biomass recovered completely after 72 h, with recovery being quicker at a lower concentration. Although not as strong as free colistin at the beginning, the anti-biofilm activity of PVA-PLGA particles and CS-PLGA particles was maintained after 72 h, and confocal microscopy imaging of rhodamine-encapsulated particles confirmed that both of them penetrate into biofilm. Interestingly, CS-PLGA NPs released twice as much colistin as PVA-PLGA NPs in the first 24 h in the release study. However, in biofilm treatment, CS-PLGA particles reduced approximately the same amount of biomass as PVA-PLGA particles after 24 h. The release kinetics of PLGA NPs may be changed after embedding into lactose [23].

Lipid-PLGA NPs were synthesized by Cheow and colleagues to encapsulate levofloxacin [38]. A lipid coat over the PLGA core was expected to help PLGA NPs penetrate sputum and target biofilm. Phosphatidylcholine (PC) was incorporated into the aqueous phase to generate the lipid coating. Levofloxacin-loaded lipid-PLGA NPs had an encapsulation efficiency of 21%. In the in vitro test, Lipid-PLGA NPs showed higher efficacy in eradicating P. aeruginosa biofilms compared to PLGA NPs. The author hypothesized that the incorporation of lipid may facilitate the diffusion of levofloxacin into the biofilm matrix. In another study, rhoamnolipid, a biosurfactant produced by P. aeruginosa, has shown the potential to trigger the release of low solubility-low permeability antibiotic such as Levofloxacin and calcein from lipid-PLGA NPs by disrupting the lipid coating. However, neither an in vivo nor in vitro biofilm susceptibility test was conducted in the study [81].

Instead of antibiotics, Iannitelli et al. [126] encapsulated carvacrol (Car.) into PLGA NPs using an oil-in-water single emulsion method. This was done to see its capability as an adjunctive therapy to promote the eradication of biofilm when an antibiotic was co-delivered. Car. is present in essential oil and able to inhibit the growth of several bacterial strains. Due to the drug’s hydrophobic nature, PLGA NPs were used to improve its bioavailability. Sixty percent of Car. was released after 3 h in an in vitro drug release study, and 95% was released in the first 24 h. Based on a significant reduction of elasticity and mechanical stability from rheological testing, Iannitelli and colleagues suggested the architecture of 2-day-old biofilm was altered, which made it more susceptible to antimicrobial agents [126].

PLGA NPs provide a robust drug delivery system. Most PLGA-encapsulated antibiotics show a biphasic sustained release profile in vitro. With the decoration of PEG on the surface, the biocompatibility of PLGA NPs can further be improved and the circulation time elongated, if the application necessitates this. The low encapsulation efficiency, especially of hydrophilic agents, is the main disadvantage of PLGA NPs.

3.8 Other polymeric particles

In addition to PLGA NPs, many investigators have had success in using other polymeric particles. Amoxicillin (AMX) was loaded into lipid polymer NPs containing pectin sulfate (PECS) by Cai et al. [47] to treat Helicobacter pylori biofilms and prevent biofilm adhesion. PECS was dissolved in water and then mixed with CaCl₂ to form electrolyte complexes before it was emulsified with a mixture of rhamnolipid and phospholipids (phosphatidylcholine, PC) in methylene dichloride. The particles were then freeze dried and soaked in an AMX aqueous solution to load AMX into NPs. The potential of pectin to inhibit the adhesion of H. pylori onto host cells such as gastric epithelial AGS cells and red blood cells was studied by Song and Li [127]. Rhamnolipid is a bacterial surfactant produced by P. aeruginosa and has been found to be involved in several stages of biofilm development. It is hypothesized that rhamnolipid may have the ability to induce biofilm EPS breakdown as well as fuse with the EPS of biofilms so that the antibiotic can be released inside the EPS. Cheow and Hadinoto [81] incorporated rhamnolipid into PLGA NPs as mentioned before. In this research, AMX-loaded RHL-PC lipid polymer NPs had a mean size of 181 nm and a negative zeta potential. Due to the solubility of AMX in water, the majority of the AMX was released in the first 2 h and the rest was released at a relatively stable rate for another 22 h. Live/dead staining was used to visualize biofilm bacteria by confocal scanning laser microscopy and quantify viability by fluorescence intensity. A percentage of 82.4 of biofilm bacteria was not viable after treatment with AMX-loaded RHL-PC lipid polymer NPs. The minimum inhibitory concentration of AMX to mature H. pylori biofilms was reduced by nearly 7-fold after treatment with RHL-PC lipid polymer NPs for 24 h at 37°C. Fluorescein isothiocyanate was used to tag polysaccharides in EPS as an indicator of the change of the amount of EPS after mature biofilms were treated with RHL-PC lipid polymer NPs for 24 h. The fluorescence intensity of the treated group accounted for 15% of the control group, suggesting a significant EPS loss, which coincides with the previous two results that more susceptibility was observed after treatment [127].

In addition, broad spectrum antimicrobial activity against intraoral biofilm of dental resin composite containing 1% w/w quaternary ammonium polyethylenimine (QPEI) NPs in vivo was observed by Beyth et al. [128]. A specimen of resin composite containing QPEI NPs and resin composite control were recovered from 10 volunteers after letting biofilm buildup on the composite for 4 h. Confocal laser scanning microscopy revealed that the viability of biofilm bacteria decreased significantly compared to the control group, while the thickness of biofilms on QPEI incorporated resin composite increased. This contradicts established knowledge that an anti-biofilm effect of bactericidal-immobilized material is only observed when biofilm bacteria come into contact with that material directly. The author thus hypothesized that the stress caused by QPEI NPs in the resin composite interacting with biofilm bacteria cell wall may electrostatically trigger a programmed cell death of other biofilm bacteria that were not in contact with the resin composite [128].

Another polymeric NP system consisting of silver NPs attached to poly(L-lactic acid) (PLLA) particles was created by Taheri et al. [41] using an emulsion method with sodium dodecyl sulfate (SDS) and PVA as surfactants. Although SDS stabilized silver-loaded PLLA NPs exhibited a smaller size (~90 nm) compared to those stabilized with PVA (~225 nm), more silver NPs (~17 nm after the synthesis process) were found attached to the PVA group under both SEM and TEM. About 0.6% of silver was found in the hybrid NPs using XPS. Thus, only the PVA stabilized NPs were further immobilized onto a thin layer of allylamine plasma polymer (AApp) through electrostatic interaction between COO- groups on the surface of the particles and NH3⁺ groups on the AApp film to test their capability of inhibiting S. epidermidis biofilm formation, and Safranin-O staining showed that up to 98% of biofilm development was inhibited compared to control treated with blank PLLA NPs. The author attributed this result to the combination of PLLA and silver NPs. Lactic acid, which degrades from PLLA, increases the permeability of the bacterial cell wall, allowing more silver ions into the bacterial cells. Additionally, the degradation resulted in local pH drop that accelerated oxidation and dissolution of silver NPs [41].

A study by Kho et al. [129] focused on encapsulation of LEX into poly(caprolactone) (PCL) NPs at an encapsulation efficiency of 5–10%. LEX-loaded PCL NPs were then spray-dried with lactose and leucine to be delivered by dry-powder inhalers in lung infection treatments. The inclusion of lactose and leucine prevented the melting of PCL during the spray-drying process, increased re-dispersability of PCL NPs, and may facilitate the deposition of PCL NPs onto lung tissues. In vitro E. coli biofilm treatment results showed that the encapsulation of LEX into PCL NPs maintained the anti-biofilm efficacy of free LEX. A total of 99.9% biofilm bacterial cells were killed after 24 h of incubation, which was determined using the drop plate method [129]. The drop plate method enumerates viable bacterial cells but has not been widely standardized [130], [131], [132]. Biofilms need to be removed from attached surfaces and disaggregated before using this method to measure viability. More details about utilizing this method are described elsewhere by Zelver et al. [133].

3.9 Silica nanoparticles

Biodegradable materials are expected to provide a way to deliver antimicrobial agents to a biofilm in a controlled manner. However, as revealed by in vitro experiments, organic biodegradable NPs tend to start releasing the active agent (attached or encapsulated) molecules once in contact with water. In some cases, burst release is caused by antimicrobial agents attached onto the NP surface like PLGA NPs. Yet in other cases, like liposomes, relatively weak mechanical properties are responsible for premature leakage of encapsulated molecules. Thus, stronger and biocompatible but not biodegradable silica NPs become an option. In terms of anti-biofilm uses, silica NPs are commonly paired up with nitric oxide (NO) as a carrier to deliver NO donor molecules. NO is an endogenous free radical and is involved in several biological processes [134]. In terms of the antimicrobial effects, researchers expect NO and its byproducts to eradicate the biofilm without giving rise to resistance like antibiotics. Another advantage of NO is its broad-spectrum antimicrobial capability and low toxicity at working concentration [135].

NO releasing inorganic-organic hybrid silica NPs were synthesized in a novel one-pot method reported by Shin et al. [136]. Generally, silica NPs were prepared through cocondensation of tetramethoxysilane (TMOS) or tetraethoxysilane (TEOS), and various aminoalkoxysilanes such as N-(6-aminohexyl)aminopropyltrimethoxysilane (AHAP3), N-(2-aminoethyl)-3-aminopropyltrimethoxysilane (AEAP3), and (aminoethylamino-methyl)phenethyltrimethoxysilane (AEMP3) were used in this study with ethanol, water, and ammonia. Aminoalkoxysilanes provided functional amine groups that were further converted to N-diazeniumdiolate NO donors under high pressures of NO and TMOS or TEOS as the initial silicon alkoxide backbone. Particle sizes tended to pseudolinearly decrease with increasing AEAP3 concentration, and this characteristic made silica NP size tunable. When NP size was smaller, the NO release rate was faster possibly because water in smaller NPs requires a decreased distance to diffuse to NO donors. The highest total NO amount released from the AHAP3/TMOS combination was almost three times that of surface grafted control silica NPs. The longest half-life of NO release was up to 12 h, achieved by AEAP3/TEOS NPs. Higher NO amounts and longer release times were attributed to the one-pot method used to prepare silica NPs, as NO donors precursors were distributed evenly throughout silica NPs. On the other hand, the amount of NO donor precursors in surface grafted NPs was limited by surface area and exposure to water [136].

The method mentioned above was used to prepare silica NPs using methylaminopropyltrimethoxysilane (MAP3) or AHAP3 as a NO donor precursor. MAP3 silica NPs released much more NO than AHAP3 NPs in a shorter period. Both NPs showed more than 99% killing of P. aeruginosa biofilm bacterial cells. Yet, MAP3 NPs were 1000-fold more effective than AHAP3 NPs, indicating that a higher concentration of NO released rapidly at the biofilm infection site may be more effective than a lower concentration of NO released continuously due to NO diffusion into biofilms, which are facilitated by high concentrations. MAP3 NPs were then used to treat against biofilms of P. aeruginosa, E. coli, S. aureus, S. epidermidis, and C. albicans, and 99.999% reduction in Gram-negative biofilm, 99.9% in fungal biofilm, and 99% in Gram-positive biofilms were observed, demonstrating a broad-spectrum anti-biofilm capability. To be noted, more susceptible Gram-positive biofilms usually appeared more resistant to NO releasing NPs in this study. The authors hypothesized that electrostatic interaction between silica NPs and biofilms may play a role in these experiments [33].

Similar results were reported by Slomberg et al. [137] where Gram-negative biofilms were more susceptible to NO released from silica NPs than Gram-positive ones. Surface-grafting silica NPs with different sizes and shapes represented by the aspect ratio were used to treat P. aeruginosa and S. aureus biofilms in this study. The aspect ratio of the particles was tuned by varying reaction temperature and ammonia concentration. Furthermore, more effective biofilm bacterial killing was observed from smaller (14 nm) and more rod-like (AR4) NPs. Although higher fluorescent intensity from rhodamine B isothiocyanate labeled silica NPs was observed inside 150 nm NP-treated biofilm under confocal microscopy after 60 min, 14 nm NPs dispersed biofilm in 14 min and led to bacterial cell death. Fourteen-nanometer NPs also diffused into biofilm faster than larger 150 nm particles. As a result, this NP strategy showed a rapid release of NO in high concentration from silica NPs under 20 nm was promising in Gram-negative biofilm eradication. Regarding shape, a large portion of dead bacterial cells were found after 15 min of treatment with rod-like AR4 and AR8 NPs, and the biofilm was dispersed after 60 min. Spherical NPs were confined in a limited area in the biofilm, and the death of cells was not observed until 60 min later. As this group of NPs has a much bigger size, it is more likely that AR4 and AR8 released sufficient NO and dispersed the biofilm [137].

In addition to NO donors, chlorhexidine was encapsulated into mesoporous silica NPs by submerging silica NPs in chlorhexidine ethanolic solution [138]. Chlorhexidine released for 80 h and was effective against mono-species biofilms including S. mutans, Streptococcus sobrinus, Fusobacterium nucleatum, A. actinomycetemcomitans, and E. faecalis. However, chlorhexidine released from silica NPs was unable to eradicate 72-h-old multispecies biofilm [138]. Paired with NO, silica NPs are expected to remove the biofilm without giving rise to biofilm bacterial antibiotic resistance. Similar to conventional antibiotics, a sufficient amount of NO needs to be delivered to biofilms in order to eradicate them. The size, shape, and synthesis method all affect the anti-biofilm efficacy of silica NPs. Thus, it is paramount to find the optimum combination of these parameters in order to achieve biofilm annihilation.

3.10 Importance of size and surface charge

Regardless of material composition, NP size affects the uptake of NPs in biofilms and is therefore a critical parameter in the design of an effective treatment. Forier et al. studied the penetration cut-off size for polymer NPs on Burkholderia multivorans and P. aeruginosa biofilms using PEG-coated polystyrene NPs [115]. The size of fluorescent polystyrene NPs ranged from 40 nm to 555 nm, and the zeta potential ranged from −6 mV to −9 mV, similar to poly(vinyl acetate)-coated polymeric NPs. This PEG coating was meant to minimize the interaction between hydrophobic NPs and components of biofilm so that NP size was the only variable in this research. The NP concentration inside the P. aeruginosa biofilm decreased gradually as particle size became larger than 100 nm. Sixty-one percent of the 40 nm NPs were detected, suggesting the penetration rate of NPs is size dependent, and a smaller size (<~100 nm) is favorable in P. aeruginosa biofilm treatment. In another study conducted by Meer et al. 1 μm fluorescent polystyrene beads were unable to pass through P. aeruginosa biofilm, revealing that there is an effective size for NP penetration into P. aeruginosa biofilms [115]. A similar result was also found in B. multivorans biofilm except that a sharp decrease was observed when NPs were larger than 100 nm. Forier et al. [139] hypothesized that the size limit of channels inside B. multivorans biofilm may be more uniform than P. aeruginosa biofilm.

Polystyrene NPs with different surface charges but similar sizes (~ 200 nm) were used by Messiaen et al. [116] to study the mobility of these NPs inside Burkholderia cepacia complex (Bcc) biofilms. Positively charged NPs (~30 mV) were modified from fluorescent carboxylate-modified polystyrene FluoSpheres (~−48 mV) with N, N-dimethylethylenediamine (DMEDA). Single particle analysis revealed the sluggish movement of negatively charged NPs inside the Bcc biofilms. On the other hand, positively charged NPs were hindered by the negatively charged fibrous structures within the biofilm (hypothesized to be eDNA) [116].