Disarming Pore-Forming Toxins with Biomimetic Nanosponges in Intraocular Infections

Abstract

Intraocular infections are prevalent after traumatic injuries or after common ocular surgeries. Infections cause inflammation that can damage the retina and architecture of the eye, often resulting in poor visual outcomes. Severe cases may result in blindness or require enucleation of the eye. Treatments for intraocular infections include intravitreal antibiotics and corticosteroids or surgical vitrectomy in serious cases. The increase in multidrug-resistant infections calls for novel treatment options. In the present study, a biomimetic erythrocyte-derived nanosponge was tested for the ability to neutralize pore-forming toxins from the most frequent Gram-positive bacterial causes of intraocular infections (Staphylococcus aureus, Enterococcus faecalis, Streptococcus pneumoniae, and Bacillus cereus). Nanosponge pretreatment of supernatants reduced hemolytic activity in vitro. In a murine sterile endophthalmitis model, nanosponge pretreatment of injected supernatants resulted in greater retinal function and less ocular pathology compared to that in eyes injected with untreated supernatants from all pathogens except methicillin-resistant S. aureus In a murine bacterial endophthalmitis model, treatment with gatifloxacin and gatifloxacin-nanosponges reduced intraocular bacterial burdens, except in the case of methicillin-sensitive S. aureus For all pathogens, eyes in both treatment groups showed decreased ocular pathology and inflammation. Overall, reductions in retinal function loss afforded by gatifloxacin-nanosponge treatment were significant for E. faecalis, S. pneumoniae, and methicillin-resistant S. aureus but not for B. cereus and methicillin-sensitive S. aureus These results suggest that clinical improvements in intraocular infections following nanosponge treatment were dependent on the complexity and types of toxins produced. Nanosponges might serve as an adjunctive therapy for the treatment of ocular infections.IMPORTANCE Endophthalmitis is a blinding consequence of bacterial invasion of the interior of the eye. Because of increases in the numbers of ocular surgeries and intraocular injections, the incidence of endophthalmitis is steadily increasing. Staphylococcus aureus, Enterococcus faecalis, Streptococcus pneumoniae, and Bacillus cereus are leading causes of infection following ocular procedures and trauma and are increasingly more difficult to treat due to multidrug resistance. Each of these pathogens produces pore-forming toxins that contribute to the pathogenesis of endophthalmitis. Treatment of these infections with antibiotics alone is insufficient to prevent damage to the retina and vision loss. Therefore, novel therapeutics are needed that include agents that neutralize bacterial pore-forming toxins. Here, we demonstrate that biomimetic nanosponges neutralize pore-forming toxins from these ocular pathogens and aid in preserving retinal function. Nanosponges may represent a new form of adjunct antitoxin therapy for serious potentially blinding intraocular infections.

Keywords: antibiotic; endophthalmitis; eye; infection; nanoparticle.

FIG 1

Rabbit nanosponges reduced hemolytic activity in vitro. Undiluted filter-sterilized supernatants were obtained from 18 h cultures of B. cereus strain ATCC 14579 (BC), methicillin-resistant S. aureus ocular isolate 180 (MRSA), methicillin-sensitive S. aureus strain 8325-4 (MSSA), and S. pneumoniae strain TIGR4 (SP), mixed 1:1 with 8 mg/ml rabbit nanosponges (NS) or PBS only, and allowed to incubate at 37°C for 30 min. Rabbit NS were removed by centrifugation and hemolytic activity was assessed. Values represent the means ± the standard deviations from three independent experiments. *, P = 0.0006; **, P = 0.005; ***, P < 0.0001 versus untreated controls.

FIG 2

Rabbit nanosponge neutralization of PFTs improved retinal function retention in a sterile endophthalmitis model. Right eyes of C57BL/6J mice were injected with either 0.5 μl of rabbit nanosponge (NS)-treated bacterial supernatant or untreated bacterial supernatant from either B. cereus (BC), methicillin-resistant S. aureus (MRSA), methicillin-sensitive S. aureus (MSSA), or S. pneumoniae (SP). Retinal function was assessed by electroretinography 24 h postinjection. Eyes injected with either rabbit NS-treated BC, MSSA, or SP supernatants had significantly higher A-wave (*, P = 0.0274; ***, P = 0.0041; ****, P = 0.0052) (A) and B-wave (*, P < 0.0001; ***, P = 0.001; ****, P = 0.0052) (B) retention versus that in untreated controls. Eyes injected with rabbit NS-treated MRSA supernatant did not exhibit a significantly increased A-wave (**, P = 0.1213) (A) or B-wave (**, P = 0.1765) (B) retention relative to that in untreated controls. Values represent the means ± the standard deviations from at least 7. Two independent experiments were performed.

FIG 3

Rabbit nanosponges neutralized PFTs and protected the retina from PFT-mediated damage in a sterile endophthalmitis model. Right eyes were injected with either 0.5 μl of rabbit nanosponge (NS)-treated B. cereus (BC), methicillin-resistant S. aureus (MRSA), methicillin-sensitive S. aureus (MSSA), S. pneumoniae (SP) supernatant, or untreated supernatant. Eyes were then harvested 24 h later and processed for hematoxylin and eosin staining. Images are representative of at least 3 eyes from 2 independent experiments. Control uninjected mouse eyes did not exhibit inflammatory infiltrate in either the anterior or posterior chambers, and the corneas and retinal layers appeared normal. Eyes injected with BC, MRSA, MSSA, and SP typically showed retinal and corneal edema and cellular infiltrate and fibrin deposition in the anterior and posterior segments. In eyes injected with rabbit NS-treated BC, MRSA, MSSA, and SP supernatant showed less anterior and posterior chamber infiltrate, less retinal edema, intact retinal layers, and normal corneas.

FIG 4

Human nanosponges and human nanosponges plus gatifloxacin increased retinal function retention and protected retinal architecture following E. faecalis infection in a murine model of endophthalmitis. Right eyes of mice were infected with 100 CFU of E. faecalis. At 6 h postinfection, E. faecalis-infected right eyes were intravitreally injected with 0.5 μl PBS only (untreated control), 0.5 μl PBS containing 1.25 μg gatifloxacin (Gat), 0.5 μl PBS containing 1.25 μg gatifloxacin and 2 μg of human nanosponges (Gat+NS), or 0.5 μl PBS containing 2 μg of human nanosponges (NS). (A) Retinal function was assessed by electroretinography 24 h postinfection. Values represent means ± standard deviations (SDs) from at least 6 eyes per group in two independent experiments (A-wave: *, P < 0.0001; **, P = 0.0014; B-wave: *, P < 0.0001; **, P = 0.0021 versus untreated controls). (B) Eyes were harvested from mice and E. faecalis CFU/eye was determined. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (*, P = 0.0007; **, P = 0.0013; ***, P = 0.5728 versus untreated controls). (C) Histological analysis of eyes infected with E. faecalis revealed retinal and corneal edema and cellular infiltration and fibrin deposition in both the anterior and posterior chambers; partial disruption of retinal layers was also apparent. In eyes treated with Gat and Gat+NS, retinal dissolution and edema, cellular infiltration, and fibrin deposition were reduced relative to that in untreated eyes. Eyes treated with NS only showed intact retinal layers, but cellular infiltrates and fibrin were not reduced in the anterior and posterior segments relative to that in untreated eyes.

FIG 5

Human nanosponges increased retinal function retention and preserved retinal architecture but did not influence growth following S. pneumoniae infection in a murine model of endophthalmitis. Right eyes of mice were infected with 100 CFU of S. pneumoniae. At 6 h postinfection, S. pneumoniae-infected right eyes were intravitreally injected with 0.5 μl PBS only (untreated control), 0.5 μl PBS containing 1.25 μg gatifloxacin (Gat), 0.5 μl PBS containing 1.25 μg gatifloxacin and 2 μg of human nanosponges (Gat+NS), or 0.5 μl PBS containing 2 μg of human nanosponges (NS). (A) Retinal function was assessed by electroretinography 24 h postinfection. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (A-wave: *, P = 0.0047; **, P = 0.0156; ***, P = 0.0148; B-wave: *, P = 0.0070; **, P = 0.0047; ***, P = 0.0379 versus untreated controls). (B) Eyes were harvested from mice and S. pneumoniae CFU/eye was determined. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (*, P = 0.0022; **, P = 0.5887 versus untreated controls). (C) Histopathological analysis of untreated and treated eyes revealed an edematous retina and cornea and cellular infiltration and fibrinous deposition in both anterior and posterior segments in untreated eyes. Eyes treated with either Gat or NS showed less retinal edema, cellular infiltration, and fibrin deposition in both chambers compared to that in untreated eyes. Eyes treated with both Gat and NS appeared essentially normal.

FIG 6

Human nanosponges did not influence retinal function retention, retinal architecture, and growth following MSSA infection in a murine model of endophthalmitis. Right eyes of mice were infected with 5,000 CFU of MSSA. At 6 h postinfection, MSSA-infected right eyes were intravitreally injected with 0.5 μl PBS only (untreated control), 0.5 μl PBS containing 1.25 μg gatifloxacin (Gat), 0.5 μl PBS containing 1.25 μg gatifloxacin and 2 μg of human nanosponges (Gat+NS), or 0.5 μl PBS containing 2 μg of human nanosponges (NS). (A) Retinal function was assessed by electroretinography 24 h postinfection. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (A-wave: *, P = 0.7959; **, P = 0.4813; ***, P = 0.9682; B-wave: *, P = 0.0831; **, P = 0.3154; ***, P = 0.1333 versus untreated controls). (B) Eyes were harvested from mice and MSSA CFU/eye was determined. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (*, P = 0.1949; **, P = 0.0585; ***, P = 0.9591 versus untreated controls). (C) Histopathology revealed severe inflammation in untreated eyes, with cellular infiltration and fibrinous deposition in both chambers and severe retinal and corneal edema. The Gat- and Gat+NS-treated groups showed a decrease in inflammation in terms of decreased infiltrate and fibrin in both chambers compared to that in untreated eyes. The NS-treated group appeared similar to the untreated group. In all groups, the layers of the retina remained relatively intact.

FIG 7

Human nanosponges augmented gatifloxacin in increasing retinal function retention and preserving retinal architecture but did not influence growth following MRSA infection in a murine model of endophthalmitis. Right eyes of mice were infected with 5,000 CFU of MRSA. At 6 h postinfection, MRSA-infected right eyes were intravitreally injected with 0.5 μl PBS only (untreated control), 0.5 μl PBS containing 1.25 μg gatifloxacin (Gat), 0.5 μl PBS containing 1.25 μg gatifloxacin and 2 μg of human nanosponges (Gat+NS), or 0.5 μl PBS containing 2 μg of human nanosponges (NS). (A) Retinal function was assessed by electroretinography 24 h postinfection. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (A-wave: *, P = 0.0721; **, P = 0.0097; ***, P = 0.7577; B-wave: *, P = 0.0541; **, P = 0.0136; ***, P = 0.6806 versus untreated controls). (B) Eyes were harvested from mice and MRSA CFU/eye was determined. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (*, P = 0.0002; **, P = 0.0207 versus untreated controls). (C) Histopathological analysis showed severe retinal and corneal swelling, alterations in the architecture of the retina, and cellular infiltration of the posterior segment in untreated eyes. Gat-treated eyes showed reduced retinal and corneal edema and decreased cellular infiltration, and retinal layers appeared normal compared to those in untreated eyes. Eyes treated with Gat+NS were similar to uninfected normal controls. NS-treated eyes appeared to have reduced inflammation relative to untreated eyes; however, cellular infiltrate and fibrinous deposition was still observed in both the anterior and posterior segments.

FIG 8

Human nanosponges did not influence retinal function retention and growth but preserved retinal architecture following B. cereus infection in a murine model of endophthalmitis. Right eyes of mice were infected with 100 CFU of B. cereus. At 6 h postinfection, B. cereus-infected right eyes were intravitreally injected with 0.5 μl PBS only (untreated control), 0.5 μl PBS containing 1.25 μg gatifloxacin (Gat), 0.5 μl PBS containing 1.25 μg gatifloxacin and 2 μg of human nanosponges (Gat+NS), or 0.5 μl PBS containing 2 μg of human nanosponges (NS). (A) Retinal function was assessed by electroretinography 12 h postinfection. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (A-wave: *, P = 0.0813; **, P = 0.3450; ***, P = 0.2949; B-wave: *, P = 0.0813; **, P = 0.1812; ***, P = 0.2343 versus untreated controls). (B) Eyes were harvested from mice and B. cereus CFU/eye was determined. Values represent means ± SDs from at least 6 eyes per group in two independent experiments (*, P = 0.0407; **, P = 0.0236; ***, P = 0.1435 versus untreated control). (C) Histological examination of untreated B. cereus-infected eyes showed retinal and corneal edema, cellular infiltrates and fibrinous exudates in the anterior and posterior segments, and retinal layer disruption. In mouse eyes treated with Gat, Gat+NS, and NS alone, less anterior and posterior segment cellular infiltration and fibrin deposition were observed, retinal and corneal edema was reduced, and retinal layers were intact compared to untreated eyes.