Effect of Nitric Oxide on Acanthamoeba castellanii

Abstract

Purpose: Acanthamoeba keratitis is a well-known intractable corneal infectious disease. We investigated the anti-Acanthamoeba effect of exogenous nitric oxide (NO).

Methods: Acanthamoeba castellanii was axenically cultured and exposed to various concentrations of NO donors, such as sodium nitrite, sodium nitroprusside (SNP), and NO-releasing silica nanoparticles (coated in branched polyethylene imine, size:100 nm), for 1 to 7 days (sodium nitrite and SNP: 0, 0.1, 1, 10, 100, and 1000 μM; silica nanoparticles: 0, 6.25, 12.5, 25, 50, and 100 μg/mL). Human corneal epithelial cells (HCECs) were cultured and exposed to sodium nitrite, SNP (0, 0.1, 1, 10, 100, and 1000 μM), and silica nanoparticles for 1, 2, and 3 days.

Results: Sodium nitrite and SNP showed a dose-dependent inhibitory effect on A. castellanii viability. A more prominent inhibitory effect was observed with SNP (less than 10% of organisms survived at 7-day culture with 1000 μM) compared with sodium nitrite. However, more cytotoxicity on HCEC was observed with SNP. NO-releasing silica nanoparticles were successfully internalized into the amoebic cytoplasm and accumulated in large vacuoles. Although blank silica nanoparticles had no inhibitory effect on A. castellanii viability, NO-releasing silica nanoparticles showed a dose-dependent amoebicidal effect. Furthermore, no cystic transformation of A. castellanii was observed under a phase contrast microscope or transmission electron microscope after exogenous NO treatment.

Conclusions: Our results demonstrated the anti-Acanthamoeba effect of exogenous NO. This finding suggests that NO-releasing drug platforms, including nano-carriers, can be a promising therapeutic strategy for Acanthamoeba keratitis.

Acanthamoeba keratitis (AK) is among the most intractable of corneal infectious diseases leading to blindness.^1,2 Acanthamoeba is ubiquitously distributed in the environment, as evidenced by the presence of anti-Acanthamoeba antibodies in 85% to 100% of healthy populations.^3,4 However, human infection by Acanthamoeba is rare; the incidence of AK is estimated to be 0.15 to 1.4 per million in developed countries.^5–7 Contact lens wear and exposure to contaminated water significantly increase the risk; the incidence of AK has been reported as 0.33 to 1.0 per 10,000 hydrogel contact lens wearers per year.^7,8 Acanthamoeba castellanii is the most common species to cause human Acanthamoeba infection, accounting for 94.3% of AK.⁸ 

Acanthamoeba is a eukaryote, so most antibiotic agents targeting bacteria and fungi have limited treatment efficacy against AK.⁹ Acanthamoeba shares functional homology with mammalian cells, and the available drugs at amoebicidal concentrations can also induce unwanted side effects in host cells and tissues.⁹ Therefore, it is necessary to develop a therapeutic agent to kill Acanthamoeba with minimal toxicity to the host cells. 

The current most common therapeutic strategy against AK includes two types of topical biaguanides: chlorhexidine 0.02% and polyhexamethylene biguanide (PHMB) 0.02%.¹ These two topical agents induce structural and permeability changes, ionic leakage, and cytoplasmic disruption of Acanthamoeba.⁹ However, treatment with topical biguanide is lengthy (months to years), toxic to corneal cells, and the effect against the cystic form of Acanthamoeba is not fully established.^10,11 The vegetative Acanthamoeba trophozoite transforms into the dormant cyst form when exposed to harsh environments, and the cyst is resistant to most commercial chemical and physical agents.^2,12 The cyst is the major source of recurrent AK when the disease is improperly treated.^2,10 

Nitric oxide (NO) is a radical gas and a key player in a common antibacterial mechanism in mammalian hosts.¹³ As a small molecule gas, NO can diffuse freely across cellular membranes.¹⁴ When generated locally in micromolar concentrations, NO is a broad-spectrum bactericidal agent that can kill both gram positive and negative bacteria.^15–17 NO is synthesized by activated macrophages, and it is well known that activated macrophages play a critical role in resolving AK.^18,19 However, the role of exogenous supply of NO in Acanthamoeba treatment has not been fully elucidated. 

In the current study, we investigated the effect of NO on A. castellanii viability. Various concentrations of two commonly used NO donors, sodium nitrite and sodium nitroprusside (SNP), were added to planktonic cultures of A. castellanii, and cell viability was assessed. In addition, we developed NO-releasing silica (SiO₂) nanoparticles (BPEI-NO-SiNPs) and investigated the efficacy of BPEI-NO-SiNPs on cultured A. castellanii. 

Materials and Methods

Acanthamoeba Culture

A. castellanii (catalog number: ATCC-30010) was purchased from American Type Culture Collection (ATCC; Rockville, MD, USA) and grown axenically in T25 tissue culture flasks containing 5 mL ATCC medium 712 (basal medium containing proteose peptone 2% [w/v], yeast extract 0.1%, and 400 μM CaCl₂, 4 mM MgSO₄, 2.5 mM Na₂HPO₄, 2.5 mM KH₂PO₄, 0.1% Na citrate, and 50 μM Fe(NH₄)₂(SO₄)₂ as the final concentrations) and incubated at 25°C. When the cultures reached near peak density, 0.25 mL culture was transferred to a fresh tube containing 5 mL fresh medium. 

Human Corneal Epithelial Cell Culture

The human corneal epithelial cells (HCECs) (catalog number: PCS-700-010) were purchased from ATCC. Cells were resuspended in corneal epithelial cell basal medium supplemented with a growth kit supplied by ATCC. The cells were plated in 75-cm² tissue flasks and then were maintained at 37°C in a 5% CO₂ and 95% air humidified atmosphere. Culture medium was changed every 3 days, and the cells were passed using 0.05% Trypsin-EDTA (Gibco BRL, Grand Island, NY, USA), and cell with passage number ≤5 are used in this study. 

Viability Assay

Acanthamoeba viability assay was performed using AlamarBlue Cell Viability Reagent (catalog number; DAL1100; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's protocol. Briefly, A. castellanii was cultured at 0.1, 0.2, 0.5, and 1 × 10⁴ cells/well for 1, 3, 5, and 7 days, respectively. Sodium nitrite, SNP, and BPEI-coated NO-releasing silica nanoparticles (BPEI-NO-SiNPs) were used as the NO donors for the study. Cells were exposed to sodium nitrite, SNP, or various silica nanoparticles (SiNPs), which were added to the culture media in a dose-dependent manner (sodium nitrite and SNP: 0, 0.1, 1, 10, 100, and 1000 μM; SiNPs: 0, 6.25,12.5, 25, 50, and 100 μg/mL). After appropriate incubation, 100 μL AlamarBlue solution was added to each culture well, and absorbance was measured at 570 nm after a 12-hour incubation of A. castellanii with the reagent. The absorbance values were finally normalized to the wavelength values at 600 nm. 

To investigate the cysticidal effect of NO, Acanthamoeba cystic transformation was induced by culturing in Neff's encystment medium (NEM) (100 mM KCl, 0.4 mM CaCl₂, 8 mM MgSO₄, 1 mM NaHCO₃, and 20 mM Tris-HCl), and the pH was adjusted to 8.9 ± 0.2. The induced cysts were then transferred to axenical culture medium and exposed to NO donors (1 mM each of sodium nitrite and SNP and 100 μg/mL SiNPs) for 3, 5, 7, and 10 days. Reculturing in the axenical culture medium stimulated cyst–trophozoite transformation. Acanthamoeba viability was measured as previously described. 

HCEC viability assays were performed using cell counting kit (CCK-8) reagent (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) according to the manufacturer's protocol. Briefly, HCECs were cultured at 1 × 10⁴ cells/ well in a 96-well plate and incubated for 24 hours. Following the adherence of cells, cells were exposed to sodium nitrite or SNP, which was added to the culture media in a dose-dependent manner (sodium nitrite and SNP: 0, 0.1, 1, 10, 100, and 1000 μM) for 24, 48, and 72 hours. After the appropriate incubation, 10 μL CCK-8 solution was added to each cultured well, and the absorbance was measured at 450 nm after 2-hour incubation of HCECs with the reagent. 

NO Releasing SiO₂ Nanoparticle Synthesis

SiNPs (50, 100, 150 nm) were prepared using the Stöber synthesis method as previously described.²⁰ Size and distribution were analyzed by scanning electron microscopy (SEM) (SIGMA; Carl Zeiss, Oberkochen, Germany) using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The SiNPs were dispersed in the branched polyethyleneimine (BPEI) solution for 5 minutes using sonication and vortexed to coat their surface with BPEI. The BPEI-coated SiNPs (BPEI-SiNPs) were converted to NO-releasing silica nanoparticle (BPEI-NO-SiNPs) via high-pressure reaction maintained for 3 days while stirring in 10 atm of NO gas environment. The collected BPEI-NO-SiNPs were rapidly dried under vacuum and stored in a vacuum-packed bag at −20°C. NO release from BPEI-NO-SiNPs was measured in PBS (0.01 M) at pH 7.4 and 37°C using a chemiluminescence NO analyzer (Sievers NOA 280i; GE Analytical Instruments, Boulder, CO, USA). 

Electron Microscopy and Ultrastructural Analysis

The detailed procedure for transmission electron microscopic (TEM) observations was described previously.^21,22 A. castellanii was treated with three sizes of SiNPs, BPEI-SiNPs, or BPEI-NO-SiNPs for 24 hours and then fixed with 3.7% paraformaldehyde (Sigma-Aldrich Corp., St. Louis, MO, USA) and 2.5% glutaraldehyde (Sigma-Aldrich Corp.) in 0.1 M phosphate buffer (PB; pH 7.6) overnight. After washing with 0.1 M PB, A. castellanii was fixed with 1% osmium tetroxide (OsO₄) in the same buffer for 1 hour. The cells were then dehydrated with a series of graded ethanol (Merk, Kenilworth, NJ, USA). Next, the cells were embedded in Epon 812, and polymerization was performed at 60°C for 3 days. Ultrathin sections (60 to 70 nm) were obtained by ultramicrotome (Leica Ultracut UCT, Wetzlar, Germany), collected on grids (200 mesh), and examined by TEM (JEM-1010; JEOL, Tokyo, Japan) operating at 60 kV, and images were recorded by CCD camera (SC1000; Gatan, Pleasanton, CA, USA). Length on electron micrography was measured using GMS software (Gatan). 

Immunocytochemistry

A. castellanii was seeded at a density of 0.5 × 10⁴ cells/mL and grown on four-well Lab-Tek chamber slides (Nalgene Nunc, Penfield, NY, USA). Organisms were exposed to sodium nitrite (10 μM), SNP (10 μM), and SiNPs (50 or 100 μg/mL) for 1 and 3 days. Amoebic cells with no exposure to any NO-releasing agent were used as negative controls. A. castellanii was fixed with 3.7% paraformaldehyde for 10 minutes at room temperature, and permeabilization was carried out using 0.1% Triton X-100 for 5 minutes at room temperature. Following washing steps with Dulbecco's phosphate buffered saline (DPBS), cells were blocked using 1% bovine serum albumin in DPBS for 30 minutes at room temperature. The chamber slides were incubated overnight at 4°C with mouse monoclonal α-tubulin (1:2000; catalog number: T5168; Sigma-Aldrich). The chamber slides were then washed with DPBS and incubated with Alexa 488-conjugated donkey anti-mouse antibody (1:1000; catalog number R37114; Invitrogen, Carlsbad, CA, USA) for 2 hours at room temperature. Staining for F-actin was carried out using tetramethylrhodamine isothiocyanate (TRITC)-conjugated phalloidin (1 μg/mL; Sigma-Aldrich). Counterstaining of cell nuclei was carried out using 4′,6-diamidino-2′-phenylindole (DAPI; Roche, Basel, Switzerland; dilution 1:1000) for 10 minutes at room temperature. Slides were viewed using a confocal microscope (LSM 700; Carl Zeiss Co. Ltd., Oberkochen, Germany). 

Statistical Analysis

Data are presented as mean ± SE, and statistical significance was determined by 1-way ANOVA followed by the Dunnett's multiple comparison test. GraphPad Prism Ver. 5.01 (GraphPad Software, Inc., La Jolla, CA, USA) was used for analysis, and P < 0.05 was regarded as significant. 

Results

NO Inhibited A. castellanii Viability

Sodium nitrite and SNP successfully inhibited A. castellanii viability. However, the inhibitory effect of SNP was faster and more potent compared to that of sodium nitrite (Figs. 1A, 1B). The effect of sodium nitrite appeared after 3 days of exposure and was statistically significant only in the highest concentration (1000 μM). By contrast, SNP decreased A. castellanii viability after only 1 day of exposure, and the effect was significant at lower concentrations (1 to 10 μM). Ten micromolar sodium nitroprusside definitely and uniformly reduced the total organism number of A. castellanii following even 1 day of treatment. The inhibitory effect was dose dependent, and A. castellanii viability was consistently reduced, as demonstrated by the observation that almost total eradication of amoebic cells after 7 days of exposure to 1 mM of SNP. In addition, both morphologic changes and cellular density well correlated with viability assay (Fig. 1C). In general, F-actin staining identifies the cytoskeletons spread in cytoplasm, whereas α-tubulin staining represents the active motility of amoebic cells. No cystic form of A. castellanii was found when the cells were observed by phase contrast microscope (Fig. 2), immunocytochemistry, and TEM. 

Figure 1

 

(A) NaNO₂ and (B) SNP demonstrated dose-dependent inhibition of A. castellanii viability; the effect was more prominent with SNP. Longer incubation with higher concentrations of SNP inhibited A. castellanii viability to less than 10% compared with control. (C) Microscopic images taken at day 3 of exposure. Exposure to SNP (10 μM) significantly decreased amoebic cell numbers. However, no cystic form of A. castellanii was observed. (a–c) Phase contrast microscopic images. (d–f) Immunocytochemical staining images (blue: DAPI; red: F-actin; green: α-tubulin). Black scale bar denotes 100 μm. White scale bar denotes 20 μm. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2

 

Phase contrast microscopic images of cultured A. castellanii. (A) Negative control with no addition of NO. No cystic form was observed. (B) Positive control of amoebic cystic transformation 10 minutes after lactophenol cotton blue addition to culture media. Prominent cystic transformation was observed. Inserted image is the magnification of white dotted area. (C) Dramatic decrease in Acanthamoeba population was observed after 3-day exposure to 1 mM SNP. However, no cyst form was observed. (D) Number and morphology of Acanthamoeba was significantly deteriorated after 3-day exposure to 100 μg/mL BPEI-NO-SiNPs. However, no cyst form was observed. White scale bar denotes 100 μm.

Effect of Sodium Nitrite, SNP, and BPEI-SiNPs on HCEC Viability

Sodium nitrite and SNP showed markedly different cytotoxicity to HCECs (Fig. 3). Sodium nitrite showed no cytotoxic effect on HCEC up to 1000 μM exposure for 72 hours. In contrast, the dose-dependent cytotoxic effect of SNP on HCEC was evident. High dose (1 mM) of SNP reduced HCEC viability significantly up to 38% (24-hour exposure), 52% (48-hour exposure), and 69% (72-hour exposure), respectively. However, the cytotoxic effect of SNP on HCECs was not as severe as on A. castellanii. BPEI-coated silica nanoparticles showed significant cytotoxicity on HCEC only at higher concentrations (100 μg/mL). 

Figure 3

 

NaNO₂ and SNP demonstrated different effects of HCEC viability. (A) NaNO₂ showed no cytotoxic effect on HCECs. (B) SNP showed dose-dependent inhibition of HCEC viability. (C, D) Cytotoxic effect on HCEC was observed with BPEI-coated SiNPs only at higher concentration (100 μg/mL) whether it releases NO (BPEI-NO-SiNPs) or not (BPEI-SiNPs). *P < 0.05, ***P < 0.001.

SiNPs Did Not Affect A. castellanii Viability

Three sizes of SiNPs (50, 100, and 150 nm) were successfully internalized to the cytoplasmic vacuoles in A. castellanii (Fig. 4). No structural damage was observed on the cell membrane, nucleus, and mitochondria. SiNPs had no adverse effect on amoebic cell viability after up to 7 days of exposure to various concentrations (Fig. 5). A mild but significant increase in viability was observed inconsistently during the observation period, possibly due to the signal interfering effect of high concentrations of SiNPs.²² There was no morphologic alteration of F-actin and α-tubulin in A. castellanii after exposure to SiNPs (Fig. 5). 

Figure 4

 

TEM images after incubation with 50-, 100-, or 150-nm SiNPs. SiNPs were collected inside the large cytoplasmic vacuoles. Cell membranes were well maintained. Neither nuclear entry nor mitochondrial damage was observed. (E–H) Magnified images of rectangle area of A–D. Arrowheads indicate SiNPs and arrows indicate intact mitochondria. Black scale bar denotes 2 μm. White scale bar denotes 0.5 μm.

Figure 5

 

(A) SiNPs (50-, 100-, or 150-nm SiNPs) have no inhibitory effect on A. castellanii viability at up to 100 μg/mL concentration for 7 days of incubation. (B) Microscopic images taken at day 3 of exposure to SiNPs revealed normal F-actin and α-tubulin expression, as well as population numbers of amoebic cells (blue: DAPI; red: F-actin; green: α-tubulin). Scale bar denotes 20 μm. *P < 0.05.

BPEI-NO-SiNPs Inhibited A. castellanii Viability

NO was successfully released from BPEI-NO-SiNPs (Fig. 6). Significant release of NO was detected for more than 10 hours from BPEI-NO-SiNPs. NO released from BPEI-NO-SiNPs significantly reduced A. castellanii viability, whereas blank BPEI-SiNPs had no effect on A. castellanii viability (Fig. 7). This inhibitory effect of BPEI-NO-SiNPs was observed for days 1 to 7 after exposure and was dose dependent. The effect of exposure to 100 μg/mL BPEI-NO-SiNPs was most potent and resulted in more than a 60% decrease of A. castellanii viability even after 1 day of exposure. No cystic form of A. castellanii was found when the cells were observed by phase contrast microscope (Fig. 2), immunocytochemistry, and TEM. 

Figure 6

 

Morphology of silica nanoparticles taken by TEM (A–C) and SEM (D–F). (A, D) 100-nm SiNPs. (B, E) 100-nm BPEI-SiNPs. (C, F) 100-nm BPEI-NO-SiNPs. (G) NO release from BPEI-NO-SiNPs, measured by a chemiluminescence NO analyzer over time. [NO]m = maximum NO flux.

Figure 7

 

(A) BPEI-NO-SiNPs successfully demonstrated dose-dependent inhibition of A. castellanii viability for 7 days of incubation. (B) Microscopic images taken at day 3 of exposure to SiNPs revealed a significant decrease of amoebic cells (blue: DAPI; red: F-actin; green: α-tubulin). Scale bar denotes 20 μm. (C) TEM images revealed both BPEI-SiNPs and BPEI-NO-SiNPs were collected inside the large cytoplasmic vacuoles. (b, d) Magnified images of rectangle area of a and c. Black scale bar denotes 2 μm. White scale bar denotes 0.5 μm. *P < 0.05, **P < 0.01, ***P < 0.001.

NO Inhibited Trophozoite Transformation of A. castellanii Cysts: Indirect Marker of Cyst Viability

Sodium nitrite, SNP, and BPEI-NO-SiNPs successfully inhibited trophozoite transformation of Acanthamoeba cysts. The inhibitory effect was most prominent with SNP compared with BPEI-NO-SiNPs and sodium nitrite (Fig. 8) When considering that all live cysts eventually transform to trophozoites in a favorable environment (axenical culture media), the decreased number of transformed trophozoites and lowered viability may reflect the possible cysticidal effect of sodium nitrite, SNP, and BPEI-NO-SiNPs. 

Figure 8

 

NaNO₂, SNP, and BPEI-NO-SiNPs demonstrated a significant inhibitory effect on the cyst–trophozoite transformation of A. castellanii. The decreased numbers of trophozoites are represented by a viability assay (A) and phase contrast microscopic images (taken at day 10) (B). (A) SNP showed the most prominent inhibitory effect compared with NaNO₂ and BPEI-NO-SiNPs. ***P < 0.001.

Discussion

In this study, we investigated the potential anti-Acanthamoeba effect of NO on cultured A. castellanii, as demonstrated by the effects of two NO-releasing chemicals (sodium nitrite and SNP) and one NO-releasing nanoparticle (BPEI-NO-SiNPs). 

In the clinical setting, early diagnosis and effective treatment are two main requirements for the control of AK.⁸ However, it is widely accepted that early and accurate diagnosis of AK is quite challenging, and many cases of early AK are misdiagnosed as herpetic keratitis and prescribed with ineffective antiviral agent.⁸ Furthermore, there have been no commercialized therapeutic agents specifically targeting AK. Currently, the treatment of AK has been largely based on membrane-attacking chemicals that are highly positively charged, such as PHMB or chlorhexidine.¹ However, these chemicals are highly toxic to corneal cells and can sometimes aggravate corneal scar formation even after the elimination of the pathogen. Therefore, unfavorable final outcomes such as corneal opacity and thinning are common with significant deterioration of visual acuity.^5,8 

Recently, the desperate quest to development new therapeutic or preventative agents for AK has been attempted.⁹ First, controlling Acanthamoeba contamination of the contact lens can be an effective initial step to prevent AK. Recently, Lee et al. reported that adding autophagy inhibitors, such as 3-methyladenine or chloroquine, to commercial multi-purpose contact lens care solution could dramatically reduce Acanthamoeba contamination, which suggests the importance of autophagy in AK pathogenesis.²³ In addition, ergosterol, the abundant cell membrane sterol of Acanthamoeba, was suggested as a therapeutic target to kill the pathogen. It was reported that azoles, such as voriconazole, inhibited ergosterol synthesis of Acanthamoeba and demonstrated significant amoebicidal effect.²⁴ Other experimental agents, such as riboflavin, combined with ultraviolet A, oleic acid, α-bisabolol, lactoferrin, and siRNA technology, have also been actively investigated to develop promising amoebicidal agents.^25–28 

NO has been widely studied as a therapeutic agent against pathogenic bacteria and other organisms.^29,30 Sodium nitrite and SNP are two widely accepted NO donors commonly used in experimental settings. Our results demonstrated that SNP had more potent amoebicidal effect (>90% killing effect, 1000 μM, 3 days of exposure) compared to sodium nitrite (<50% killing effect, 1000 μM, 3 days of exposure). We postulate that the discrepancy between the effects of the two chemicals might originate from their different mechanisms of action to release NO. NO release from SNP is mainly dependent on whether the organism has a cyanide group or thiol group, which can transform glutathione free radical to S-nitrosoglutathione (GSNO).^31,32 In contrast, NO release from sodium nitrite mainly depends on the existence of the enzyme, nitrite reductase, inside cells. In addition, SNP was previously known to have both NO dependent and independent pro-apoptotic effect on human cells.³³ Therefore, the combination effect of both NO dependent and independent mechanisms may explain the stronger cytotoxic effect of SNP both on Acanthamoeba and HCEC observed in this study. 

Although NO has a significant amoebicidal effect, any potential toxicity of topical application of NO donors should be considered. It is noteworthy that no HCEC cytotoxicity was observed with sodium nitrite treatment while a significant HCEC cytotoxicity was found by SNP exposure. As demonstrated, HCEC viability was well maintained with sodium nitrite treatment up to 1000 μM. This finding is consistent with the previous report that low concentrations of sodium nitrite (up to 6.25 mM) had little cytotoxic effect on human gastric epithelial cells.³⁴ Our findings suggest that topical application of sodium nitrite is relatively safe method to booster the control AK with minimal damage to cornea although the potency is somewhat limited. The safety and epithelial healing effect of the topical application of 10 M of sodium nitrite solution was reported in mice.³⁵ The in vivo safety of different NO donors, such as S-nitrosoglutathione and S-nitroso-N-acetylcysteine, was also demonstrated up to 10 mM concentration in rabbit eyes.³⁶ In contrast, we observed that SNP exposure showed a dose dependent and significant cytotoxic effect on HCEC even with a relatively low concentration (1 μM). This finding is consistent with the previous report that SNP (≤500 μM) was toxic to mammalian cells.^33,37 However, it is interesting that there observed SNP induced cytotoxicity discrepancy between Acanthamoeba and HCECs. Exposure to 100 μM of SNP for 3 days induced only 30% of cytotoxicity on HCEC, whereas >90% of Acanthamoeba were killed under the same condition. Although some corneal toxicity is anticipated, our result suggests that sodium nitroprusside can also be used to treat AK but with a considerably narrow therapeutic window. 

To further confirm that NO was the main driver of the antiamoebic effect of sodium nitrite and SNP, we developed NO-releasing silica nanoparticles by coating 100-nm SiNPs with BPEI and adding NO by high-pressure reaction and investigated the pure NO effect. The resulting BPEI-NO-SiNPs showed favorable NO-releasing profiles lasting more than 10 hours. Although the safety and eventual biodegradation issues remain debatable, SiNPs are actively investigated for various applications, such as acting as a drug carriers or appearing in food, functional fabrics, and cosmetics.³⁸ In our previous studies, 50-, 100-, and 150-nm SiNPs had no significant cytotoxicity in cultured human corneal cells.^21,22 The in vivo safety of topical application of 100-nm SiNPs was also reported in rats.³⁹ Our results demonstrated BPEI cytotoxicity to HCEC at high concentration. BPEI was previously known to be toxic to mammalian cells including epithelial cells and fibroblasts.^40,41 In the current study, we demonstrated that SiNPs and BPEI-SiNPs themselves did not affect Acanthamoba viability. However, BPEI-NO-SiNPs induced a significant and dose-dependent toxicity to amoebic cells (about 60% amoebicidal effect with 100 μg/mL, 1 to 7 days of exposure). This suggests that NO is the key element of amoebicidal effect observed with the exposure to sodium nitrite and SNP. When considering the surface or structural modification of nanoparticle is one of the major advantages of nanoparticle-based drug delivery, future development of Acanthamoba targeting nanodelivery system with NO can be a promising field of investigation. 

NO can kill pathogens by various mechanisms of action, including DNA damage, bacterial protein alteration, strong oxidative stress, and critical enzyme alteration.^42–44 It has been previously reported that NO is an important amoebicidal factor. It is known that mice deficient in NO synthase (iNOS) are more vulnerable to amoebic liver abscess.⁴⁵ Another study reported that murine macrophages exposed to bacteria cocultured with Acanthamoeba polyphaga produced levels of NO 10 times higher than from exposure to bacteria alone.⁴⁶ As well known, the activation of neutrophils and production of NO is the key host response to eliminate hepatic infection by Entamoeba histolytica.⁴⁷ Amoebicidal effect of exogenous NO supply using S-nitrosoglutathione and S-nitroso-N-acetylcysteine was also reported.⁴⁸ However, the exact target or mechanism to explain how NO control amoebic infections is still unclear and needs future investigation. 

Cystic transformation of Acanthamoeba when the organism encounters a harsh circumstance is a characteristic biological feature.^11,12,49 Amoebic cysts are highly resistant to any type of antiamoebic treatments, which is the most significant drawback in treating patients with AK. Therefore, we investigated whether NO treatment encouraged cyst formation. By using phase contrast microscopic images, immunofluorescent staining, and TEM, we confirmed that A. castellanii cysts were not observed after treatment with SNP or BPEI-NO-SiNPs in our experiments. Furthermore, we demonstrated that the NO donors used in this study significantly inhibited the cyst–trophozoite transformation, which suggests a possible cysticidal effect. 

Our study has several limitations. First, the exact mechanism behind the effects of NO on the delicate biological system inside Acanthamoeba remains unknown. Second, definite cysticidal effects of NO still remain unclear. Although we showed no induction of cystic transformation as well as the inhibition of cyst–trophozoite transformation after NO exposure, it is noteworthy that Acanthamoeba cyst can remain in the dormant state for several months and eventually reactivate the disease. Although we used three different platforms of NO donors, it would have provided more robust evidence of an NO effect if an NO quenching agent had been reversed the amoebicidal effects of NO donors. Furthermore, the mechanism why the same concentrations of NO donors had different cytotoxic effect on HCEC and Acanthamoeba still remained unclear. Finally, the lack of in vivo verification how exogenous NO enhances the host immune cell defense against A. castellanii represents another limitation of our study. 

Conclusions

We verified that NO demonstrates dose-dependent inhibition of A. castellanii viability by using NO donor chemicals and NO-releasing silica nanoparticles. Considering the ongoing quest for effective antiamoebic agents, NO-releasing drug platforms may be a promising solution for AK. 

Acknowledgments

Supported in part by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (Grant HI-15C1653), and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2016R1D1A1B03931724). 

Disclosure: B. Yim, None; J.-H. Park, None; H. Jeong, None; J. Hong, None; M. Kim, None; M. Chang, None; R.S. Chuck, None; C.Y. Park, None