The Role of CCL Chemokines in Experimental Staphylococcus aureus Endophthalmitis

Abstract

Purpose: To test the hypothesis that (C-C motif) ligand 2 (CCL2) and CCL3 impact retinal function decline and inflammation during Staphylococcus aureus endophthalmitis.

Methods: Experimental endophthalmitis was initiated by intravitreal injection of 5000 colony-forming units of S. aureus into the eyes of C57BL/6J, CCL2^−/−, or CCL3^−/− mice. At 12 and 24 hours post-infection, retinal function, bacterial load, and myeloperoxidase levels were quantified.

Results: During S. aureus endophthalmitis, we observed a significant improvement in retinal function in CCL2^−/− mice relative to C57BL/6J mice at 12 hours but not at 24 hours. In CCL3^−/− mice, retinal function was significantly improved relative to C57BL/6J mice at 12 and 24 hours. The absence of CCL2 did not alter intraocular S. aureus intraocular concentrations. However, CCL3^−/− mice had significantly lower intraocular S. aureus at 12 hours but not at 24 hours. No difference in myeloperoxidase levels was observed between C57BL/6J and CCL2^−/− mice at 12 hours. CCL3^−/− mice had almost no myeloperoxidase at 12 hours. At 24 hours, increased myeloperoxidase was observed in CCL2^−/− and CCL3^−/− mice relative to C57BL/6J mice.

Conclusions: Although the absence of CCL2 resulted in improved retinal function retention at 12 hours, CCL3 deficiency resulted in improved retinal function at 12 and 24 hours. CCL3 deficiency, but not CCL2 deficiency, resulted in almost no inflammation at 12 hours. However, at 24 hours, the absence of CCL2 or CCL3 resulted in significantly increased inflammation. These results suggest that, although both CCL2 and CCL3 impact intraocular infection outcomes, CCL3 may have a more significant impact in S. aureus endophthalmitis.

Endophthalmitis is an intraocular infection resulting from invasion of the anterior and/or posterior segments of the eye by a pathogen. Sources of endophthalmitis can be categorized as endogenous, resulting from sepsis, or exogenous, resulting from iatrogenic, postoperative causes, or penetrating ocular trauma.^1–7 Of the exogenous etiologies, postoperative complications are the most common, occurring after intravitreal injection, cataract surgery, or other sources of eye penetration.^8,9 Diagnosis hinges on cultures or genetic identification of the pathogen from the vitreous or aqueous humor and may be supported by blood cultures if contracted from an endogenous source.³ Positive cultures are most often acquired from vitrectomy samples, but aqueous samples have lower yields.¹⁰ More recently, genetic identification of pathogens by techniques such as polymerase chain reaction (PCR) have elevated the standard of diagnosis.^11–13 Conventional treatment of endophthalmitis includes intravitreal injection of antibiotics (including amikacin, fourth-generation fluoroquinolones, ceftazidime, and/or vancomycin), anti-inflammatory drugs (such as corticosteroids), and/or vitrectomy in severe cases.^3,8 Visual outcomes in endophthalmitis cases may vary widely given the possible pathogenic etiologies, but cases often result in total vision loss if not appropriately treated.^10,14,15 Due to the deleterious, fulminant nature of this disease, endophthalmitis is a medical emergency that must be treated without delay to preserve vision. 

Endophthalmitis is associated with a variety of bacteria and fungi, with most cases being caused by Staphylococcus species.^16–19 Of the Staphylococcus species, S. aureus has been considered the most pathogenic.^14,20–23 Much of the severity of S. aureus endophthalmitis stems from combined virulence factor/toxin production and multidrug resistance (MDR) that are present in both community and clinical isolates.²⁴ Toxin production is notorious for its contributions to ocular inflammation and loss of retinal function in S. aureus endophthalmitis.²⁵ Pore-forming toxins and coordinated regulation of these toxins and other virulence factors have been described as facilitating S. aureus endophthalmitis.^9,26 Furthermore, MDR is an ever-increasing threat to treatment success. Resistance of S. aureus ocular isolates to aminoglycosides, b-lactams, cephalosporins, imipenem, and semisynthetic penicillins has been previously described, resulting in therapeutic failures.^{14,20–23,27} Intraocular infections, particularly due to S. aureus, are increasing in their difficulty to treat with conventional therapeutics, leading to poor visual outcomes. Therefore, it is paramount to investigate new possible treatment avenues by studying the host inflammatory immune responses and identifying potential targets. 

The upregulation of molecules important in innate immune signaling in response to bacterial intraocular infection has been studied in mouse and rat models of endophthalmitis and in human cases.^28–31 S. aureus interacts with host innate immune signaling pathways by activating pattern recognition receptors that regulate innate immune responses leading to inflammatory cell recruitment and influx into the eye.^2,28,31–34 Of these pathways, Toll-like receptor 2 (TLR2) is important in S. aureus recognition with TLR2 activation resulting in the upregulation of (C-C motif) ligand (CCL) chemokines, including CCL2, CCL3, monocyte chemoattractant protein 1 (MCP-1) and macrophage inflammatory protein-1 alpha (MIP-1a), as well as other inflammatory mediators.^32–34 Furthermore, downstream myeloid differentiation primary response 88 (MyD88) signaling was also shown to be important in the immune response to S. aureus intraocular infection.³⁵ S. aureus was also shown to activate the Nod-like receptor P3 (NLRP3) inflammasome complex during experimental S. aureus endophthalmitis.³⁶ 

Innate immune signaling interference has been shown to downregulate the expression of complement factors and important pathway genes, including tlr2, myD88, and nod2, and chemokines, including CCL2 and CCL3, and (C-X-C motif) ligand 1 (CXCL1), CXCL2, CXCL3, CXCL5, CXCL9, and CXCL10, in experimental murine Bacillus cereus intraocular infection.³⁷ The roles of CC chemokines CCL2 and CCL3 have been previously studied in this B. cereus model of endophthalmitis. CCL2 and CCL3 function in regulating chemotaxis, signaling, and inflammation via several pathways activated during the innate immune response to pathogens.^38–42 Infected eyes of CCL2- or CCL3-deficient mice showed reduced inflammation and greater retained retinal function and architecture, as compared to wild-type mouse eyes infected with B. cereus.⁴³ Overall, experimental B. cereus endophthalmitis in CCL2^−/− mice was observed to have a better outcome than this infection in CCL3^−/− mice.⁴³ Additionally, anti-CCL2 or anti-CCL3 antibodies administered with the fourth-generation fluoroquinolone gatifloxacin also resulted in significantly lower inflammation and greater retention of retinal function in eyes infected with B. cereus.⁴³ CCL2 and CCL3 are also expressed in eyes infected with S. aureus, but little is currently known about their contributions and the effects of their absence on retinal function, bacterial growth, and inflammation severity during this infection. 

Similar groups of chemokines have been identified in eyes with S. aureus and B. cereus endophthalmitis, and these chemokines influence inflammation in experimental B. cereus endophthalmitis.^2,43,44 We therefore tested the hypothesis that CCL2 and CCL3 might similarly influence inflammation and retinal function decline during S. aureus endophthalmitis. Here, the absence of CCL2 resulted in a delay in retinal function decline at an earlier time point post-infection with no change in inflammation, and increased inflammation at a later time point, despite no change in retinal function. In contrast, the absence of CCL3 resulted in improved retinal function and almost no inflammation at an earlier time point and improved retinal function despite increased inflammation at a later time point. These results point to CCL3 as a potential therapeutic target for S. aureus endophthalmitis. 

Materials and Methods

Mice

All procedures followed the recommendations of the Guide for the Care and Use of Laboratory Animals, the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and University of Oklahoma Health Sciences Center Institutional Animal Care and Use Committee policies. The following mouse strains were purchased from The Jackson Laboratory (Bar Harbor, ME, USA): C57BL/6J (cat. no. 000664), CCL2^−/− (C57BL/6.129S4-Ccl2^tm1Rol/J, cat. no. 004434), and CCL3^−/− (C57BL/6.129P2-Ccl3^tm1Unc/J, cat no. 002687). CCL2^−/− and CCL3^−/− mice were maintained on the C57BL/6J background. The mice were kept in biosafety level 2 microisolation conditions, following a 12-hour on/12-hour off light cycle, for a minimum of 2 weeks before intraocular infections. This period allowed for the equilibration of their microbiota and ensured physiological and nutritional stabilization. 

Murine S. aureus Endophthalmitis Model

S. aureus strain 8325-4 was cultured in brain heart infusion (BHI) medium at 37°C for 18 hours. The culture was then diluted in BHI to 10⁷ colony-forming units (CFU)/mL prior to intravitreal injection. Anesthetization of mice was accomplished with a cocktail of ketamine (ketamine hydrochloride, 85 mg/kg body weight; Covetrus, Portland, ME, USA) and xylazine (AnaSed, 14 mg/kg body weight; Akorn Pharmaceuticals, Decatur, IL, USA). Eyes were injected into the mid-vitreous with sterile, borosilicate glass micropipettes (Kimble Glass Company, Vineland, NJ, USA), which were beveled to a bore size of approximately 10 to 20 µm (BV-10 KT Brown Type micropipette beveller; Sutter Instrument Company, Novato, CA, USA). Eyes were visualized via stereomicroscope, and micropipettes were inserted just posterior to the superior limbus. The right eyes were infected with 5000 CFU in 0.5 µL of diluted culture mentioned above, and the left eyes served as contralateral, uninfected controls.⁴⁴ 

Scotopic Electroretinography

At specific time points following intravitreal injection, infected mice were dark-adapted for 6 hours prior to electroretinography (ERG). At either 12 or 24 hours post-infection, mice were anesthetized as described above. Prior to ERG, topical phenylephrine (10% phenylephrine hydrochloride; Paragon BioTeck, Portland, OR, USA) was administered for dilation, and topical anesthetic (0.5% proparacaine HCl; Alcon Laboratories, Fort Worth, TX, USA) was applied to both eyes. Gold-wire electrodes were placed in direct contact with the corneas of each eye, and reference and ground electrodes were attached to the head and tail, respectively. For the ERG procedure, five white-light flashes (1200 cd·s/m²) were administered consecutively at 60 seconds apart (each of 10-ms duration) to initiate a measurable retinal response. Following the light flash, scotopic A-wave and B-wave amplitudes were recorded for each eye (Espion E2; Diagnosys, Lowell, MA, USA). Percentages of retained retinal function in the infected eye (right eye) were calculated in comparison with the uninfected (left) eye controls as follows: 100 – {[1 – (experimental A-wave or B-wave amplitude/control A-wave or B-wave amplitude)] × 100} as previously described.^43–46 Values in the results represent the mean ± standard error of the mean (SEM) for at least seven eyes per group from at least two independent experiments. Immediately following the ERG procedure, mice were euthanized by CO₂ inhalation, and the eyes were harvested, placed in PBS (pH 7.4), and homogenized prior to quantify myeloperoxidase (MPO) and bacterial CFUs. 

Myeloperoxidase Assays

At 12 or 24 hours post-infection, estimates of inflammatory cell infiltration in infected eyes were assessed through measurements of MPO concentrations derived from whole-eye homogenates. Eyes were enucleated, placed into separates tubes consisting of 400 µL of sterile PBS (pH 7.4) supplemented with proteinase inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA) and 1.0-mm sterile glass beads (BioSpec Products, Bartlesville, OK, USA), and homogenized for 60 seconds at 5000 rpm in a Mini-Beadbeater (BioSpec Products). MPO concentrations were quantified by sandwich ELISA (Hycult Biotech, Plymouth Meeting, PA, USA), as previously described.^43,46,47 The negative controls used as a comparison were MPO levels derived from uninfected eye homogenates. Values in the results represent mean ± SEM for at least four eyes per group over at least two independent experiments. 

Bacterial Quantitation

At either 12 or 24 hours post-infection, enucleated eyes were homogenized as described above. Eye homogenates were serially diluted 10-fold before being plated onto BHI agar. After incubation overnight, the CFUs per eye were determined as previously described.⁴⁴ Values in the results represent mean ± SEM for at least six eyes per group from at least two independent experiments. 

Statistical Analyses

Data are represented as the arithmetic mean ± SEM of all samples in the corresponding experimental groups run in at least duplicate. Differences between comparative groups were interpreted as being statistically significant for P < 0.05. We compared experimental groups for ERG, MPO, and CFUs per eye using the Mann–Whitney U test. All statistical analyses performed for experiments described in this paper were executed using Prism 8.4.3 for Windows (GraphPad, Boston, MA, USA). 

Results

Absence of CCL2 on the Evolution of S. aureus Endophthalmitis

We previously showed that mice genetically deficient in CCL2 or treated with anti-CCL2 antibodies experienced reduced inflammation and greater retainment of retinal function as compared to wild-type C57BL/6J mice in an experimental model of B. cereus endophthalmitis.⁴³ These findings, coupled with observed similarities of chemokine importance in infection outcome in the B. cereus and S. aureus endophthalmitis models,^2,43,44 led us to investigate the effects that a genetic absence of CCL2 might have on retinal function decline, intraocular inflammation, and bacterial replication in experimental S. aureus endophthalmitis.^2,43,44 

At 12 hours post-infection, the mean ERG A-wave retention of infected eyes of CCL2^−/− mice was significantly higher than that of infected eyes of C57BL/6J mice (90.41% vs. 67.65%; P = 0.0131) (Fig. 1A). At the same time point, the mean B-wave retention was significantly higher in infected eyes of CCL2^−/− mice as compared to infected eyes of C57BL/6J mice (82.41% vs. 57.7%; P = 0.0102) (Fig. 1B). However, at 24 hours post-infection, no significant differences in mean A-wave (P = 0.7377) or mean B-wave (P = 0.9529) retention were found between infected eyes of CCL2^−/− and C57BL/6J mice (Figs. 1C, 1D). 

Figure 1.

 

Absence of CCL2 improved retinal function only at 12 hours after infection, but the absence of CCL3 improved retinal function at both 12 and 24 hours after infection with S. aureus. Right eyes of C57BL/6J and CCL2^−/− mice or CCL3^−/− mice were infected with 5000 CFU of S. aureus 8325-4. Retinal function was assessed by electroretinography at 12 hours post-infection (A, B, E, F) or 24 hours post-infection (C, D, G, H). Values represent means ± SEM of n ≥ 7 eyes per group in two independent experiments (P ≥ 0.05); ns, not significant.

Eyes were harvested to assess bacterial replication (CFUs) and to estimate the extent of intraocular inflammation (MPO). Intraocular replication was not affected by the absence of CCL2 in CCL2^−/− mice and was similar to that in eyes of C57BL/6J mice at 12 and 24 hours post-infection: 328,000 ± 126,913 versus 264,480 ± 82,913, respectively (P = 0.895) at 12 hours and 1,371,007 ± 881,844 versus 109,867 ± 66,753, respectively (P = 0.218) at 24 hours (Figs. 2A, 2B). MPO concentrations were not significantly different in eyes of CCL2^−/− mice compared with those of eyes of C57BL/6J mice at 12 hours post-infection (269.6 ng/eye vs. 313.5 ng/eye; P > 0.05) (Fig. 3). However, at 24 hours post-infection, MPO levels in eyes of CCL2^−/− mice were significantly higher than in eyes of C57BL/6J mice (1319 ng/eye vs. 461.3 ng/eye; P < 0.0472) (Fig. 3). Together, these results showed that the deficiency in CCL2 resulted in greater retained retinal function but no difference in inflammation at 12 hours post-infection, but increased inflammation and no difference in retained retinal function at 24 hours post-infection. The absence of CCL2 did not affect staphylococcal intraocular growth during the course of infection. 

Figure 2.

 

Absence of CCL2 did not alter bacterial growth at 12 or 24 hours post-infection, and absence of CCL3 decreased bacterial growth only at 12 hours post-infection. Right eyes of C57BL/6J and CCL2^−/− mice or CCL3^−/− mice were infected with 5000 CFU of S. aureus 8325-4. Eyes were harvested from mice at 12 hours post-infection (A, C) or 24 hours post-infection (B, D), followed by the determination of S. aureus CFU/eye values. Values represent means ± SEM of n ≥ 6 eyes per group in two independent experiments (P ≥ 0.05); ns, not significant.

Figure 3.

 

Absence of CCL2 increased inflammation at 24 hours post-infection but did not affect inflammation at 12 hours after infection. Absence of CCL3 decreased inflammation at 12 hours after infection but increased inflammation at 24 hours after infection with S. aureus. Right eyes of C57BL/6J and CCL2^−/− mice or CCL3^−/− mice were infected with 5000 CFU of S. aureus 8325-4. Inflammatory cell influx at 12 or 24 hours post-infection was measured as a function of MPO concentrations determined by sandwich ELISA of harvested and homogenized whole eyes. Values represent means ± SEM of n ≥ 4 eyes per group in two independent experiments (P ≥ 0.05); ns, not significant.

Absence of CCL3 on the Evolution of S. aureus Endophthalmitis

In addition to the important role played by CCL2 in a murine B. cereus endophthalmitis model, CCL3 has also been demonstrated to play a crucial role in intraocular inflammation and retinal function decline in this model.⁴³ We sought to determine the effect that a genetic absence of CCL3 would have during experimental S. aureus endophthalmitis. Like the results obtained in CCL2-deficient mice at 12 hours post-infection, the mean A-wave retention of eyes of CCL3^−/− mice was significantly higher than in eyes of C57BL/6J mice (87.17% vs. 67.65%; P = 0.0048) (Fig. 1E). The mean B-wave retention in eyes of CCL3^−/− mice was also significantly greater than in eyes of C57BL/6J mice (76.28% vs. 57.7%; P = 0.0159) (Fig. 1F). However, in contrast to our findings in CCL2-deficient mice at 24 hours post-infection, the mean A-wave retention in eyes of CCL3^−/− mice was significantly higher than in eyes of C57BL/6J mice (83% vs. 47.99%; P = 0.0078), and the mean B-wave retention of eyes of CCL3^−/− mice was similarly significantly greater than that of eyes of C57BL/6J mice (67.39% vs. 42.01%; P = 0.0482) (Figs. 1G, 1H). 

In contrast to our results in CCL2-deficient mice, bacterial growth was significantly lower at 12 hours post-infection in eyes of CCL3^−/− mice compared to the eyes of C57BL/6J mice (63,778 ± 17,647 CFU/eye vs. 264,480 ± 82,913 CFU/eye; P = 0.0008) (Fig. 2C). However, at 24 hours post-infection, the numbers of S. aureus in eyes of CCL3^−/− and C57BL/6J mice were similar (2,028,581 ± 879,614 vs 109,867 ± 66,753, respectively; P = 0.4617) (Fig. 2D). MPO concentrations were significantly lower in eyes of CCL3^−/− mice compared to those of eyes of C57BL/6J mice at 12 hours post-infection (1.007 ng/eye vs. 313.5 ng/eye; P < 0.0001) (Fig. 3). However, at 24 hours post-infection, MPO concentrations in eyes of CCL3^−/− mice were significantly greater than in eyes of C57BL/6J mice (1,751 ng/eye vs. 461.3 ng/eye; P < 0.0120) (Fig. 3). These results suggest that the absence of CCL3 resulted in a greater degree of retinal function retention at both early and later time points in S. aureus intraocular infection but impacted bacterial growth at only the early stages of infection. An absence in CCL3 also resulted in almost no inflammation at the early stages of infection but significantly greater inflammation during the later stages of infection relative to C57BL/6J eyes. 

Discussion

The emergence of MDR among ocular isolates of S. aureus emphasizes the need for treatment strategies that function synergistically with conventional antibiotic therapies. Resistance of ocular S. aureus isolates to antibiotics has been widely reported, often leading to treatment failure.^{14,20–23,27} Inquiry into the interactions of S. aureus with host innate immunity and exploring potential therapeutic avenues for those targets are necessary in developing more effective treatment options for this fulminant eye infection. 

Innate immune signaling is prominent in the early cellular response to S. aureus intraocular infection. Giese et al.²⁸ showed upregulation of CXCL1, IL-1β, and TNF-α at 24 hours post-infection in rats. More recent studies of inflammatory pathways in S. aureus endophthalmitis have corroborated the upregulation of IFN-γ, IL-1β, IL-6, IL-8, and TNF-α in the eyes of various animal species and in humans.²⁹ In mice, CCAAT/enhancer-binding protein beta (CEBPB), colony stimulating factor 1 (CSF1), CXCL2, insulin-like growth factor 1 (IGF-1), IL-1β, IL-6, JUN, nuclear factor kappa B subunit 2 (NF-κB2), protein tyrosine phosphatase non-receptor type 1 (PTPN1), secreted phosphoprotein-1 (SPP1), signal transducer and activator of transcription 1 (STAT1), and STAT3 were shown to be upregulated at earlier time points post-infection.³⁰ Upregulated pathways in response to S. aureus also include TLR2 and MyD88.^28,30,33,35 Additionally, S. aureus was shown to activate the NLRP3 inflammasome complex during experimental infection.³⁶ 

Interference with these pathways has shown promise as a therapeutic avenue. Pretreatment of mouse eyes with Pam3Cys, a TLR2 agonist, resulted in the downregulation of CXCL2, IL-6, and JUN; IL-10, Jak/Stat, and TLR signaling mediators; and tumor necrosis factor receptor 2 (TNFR2) signaling genes during S. aureus infection.³⁰ Pretreatment of mouse eyes with 2-deoxy-glucose, a glycolysis inhibitor, prior to infection with S. aureus reduced the expression of CXCL1, CXCL2, IL-6, and IL-1β and inhibited extracellular signal-regulated kinase 1/2 (ERK1/2) phosphorylation at 24 hours post-infection.³¹ Inhibition of ERK1/2 phosphorylation was postulated to be the mechanism for the observed suppression of an inflammatory response from innate immune cells.³¹ The TLR2 pathway and manipulation of its activity have also been shown to be important in pro-inflammatory gene expression, chemokine synthesis (including CCL2 and CCL3), and resulting inflammatory cell recruitment to sites of intraocular infection.^{30,32–34,47} 

In an experimental B. cereus murine model of endophthalmitis, the genetic absence of CXCL1 in CXCL1^−/− mice resulted in decreased inflammatory immune responses and increased preservation of retinal function.⁴⁸ Anti-CXCL1 antibody treatment showed similar improvement in inflammation and retinal function outcomes.⁴⁸ The absence of CXCL2 or CXCL10 in corresponding knockout mice also improved retinal function and reduced inflammation in the same B. cereus endophthalmitis model.⁴⁹ In experimental S. aureus endophthalmitis, CXCL1 was important only at early time points post-infection, resulting in decreased inflammation and improved retinal function.⁴⁴ However, in contrast with results in the B. cereus model, administration of anti-CXCL1 antibodies did not significantly change infection outcomes in the S. aureus model.⁴⁴ Also, in contrast with results in the B. cereus model, the absence of CXCL2 or CXCL10 had no impact on infection or inflammation at early or late time points in S. aureus endophthalmitis.⁴⁴ This suggests that blocking chemokines as a therapeutic may be effective for infections caused by some pathogens, but not others. 

The CCL chemokines are also important in experimental B. cereus endophthalmitis retinal function outcomes.⁴³ We previously showed that CCL2 and CCL3 play a role in the host immune response, retinal function decline, and inflammation in that model.⁴³ These chemokines are also upregulated in S. aureus infection.⁴⁷ However, little is known about the role of CCL2 or CCL3 in the innate immune response to S. aureus endophthalmitis. The current study explored CCL2 and CCL3 to delineate their importance in inflammation and to better understand them as potential therapeutic avenues in treating endophthalmitis due to S. aureus. 

CCL2 influences monocyte trafficking and basophil, dendritic cell, and memory T-cell recruitment to sites of infection or inflammation.^38–40 In our more recent studies, the absence of CCL2 limited inflammation and improved retinal function retention at early time points in experimental B. cereus endophthalmitis.⁴³ In the current study, retinal function showed improvement only at the earlier time point in the eyes of CCL2-deficient mice but resulted in no differences in growth of S. aureus during infection (Figs. 1A, 1B, 2A, 2B). Also, no differences in inflammation were seen at 12 hours post-infection in eyes of CCL2-deficient or C57BL/6J mice (Fig. 3), despite the improved retinal function. There was also significantly increased inflammation at 24 hours post-infection in CCL2^−/− mice (Fig. 3), despite no changes in retinal function compared to the eyes of wild-type mice (Figs. 1C, 1D). This was unexpected, given the intimate role of CCL2 in immune cell chemotaxis and its pro-inflammatory effects. However, the increased levels of inflammation seen at 24 hours post-infection in the eyes of CCL2^−/− mice relative to eyes of C57BL/6J mice did not correlate with a significant decline in retinal function relative to eyes of wild-type mice. Similarly, the improvements in retinal function retention observed at 12 hours post-infection in CCL2-deficient mice relative to C57BL/6J mice did not correlate with significant changes in inflammation levels. To measure inflammation in this study, we followed MPO levels as a marker for neutrophils in the early response to bacterial infection.⁵⁰ These findings suggest the possible presence of CCL2-mediated mechanisms influencing retinal function beyond that of facilitating inflammatory cell chemotaxis into the eye, corroborating other studies highlighting the myriad roles CCL2 holds beyond that of chemotaxis.⁴¹ Specifically, in addition to calcium ion influx and integrin expression-mediated chemotaxis, CCL2 is important in instigating cytokine expression and the respiratory burst.^51,52 Other studies have also expanded on the role of CCL2, implicating it in influencing secretion of effector molecules and leukocyte behavior and survival.⁴¹ Taken together, these findings suggest that CCL2 may influence the progression of inflammation and the effects in retinal function observed in this study. Due to its effects on retinal function outcomes only early on in infection, CCL2 seems to be less important in governing S. aureus infection outcomes compared to CCL3. Together, these results suggest a less important role for CCL2 in regulating retinal function and inflammation in the eye and implicates redundancies in chemokine function or other mechanisms for inflammatory cascade regulation in the eye. 

CCL3, like CCL2, is important in macrophage, natural killer cell, and T-cell/dendritic cell interactions.⁴⁰ CCL3 also facilitates chemotactic behavior in inflammatory cells and influences the recruitment and activation of neutrophils, an important immune cell in both B. cereus and in S. aureus endophthalmitis.³⁹ Importantly, the absence of CCL3 resulted in attenuated retinal function decline and intraocular inflammation without changes in bacterial growth in our model of B. cereus endophthalmitis.⁴³ In the current study, retinal function was improved in the eyes of CCL3^−/− mice relative to that in wild-type mice at both 12 and 24 hours post-infection, implicating its importance in retinal function decline both early and later in the infection process (Figs. 1E–1H). Interestingly, we observed almost no inflammation in eyes of CCL3^−/− mice at 12 hours post-infection but increased inflammation relative to that in wild-type mouse eyes at 24 hours (Fig. 3). These findings suggest the presence of redundancy in chemokine functional coverage or the possibility of mechanisms for governing retinal function decline and inflammation beyond that of merely immune cell chemotaxis. Previously, we observed a similar pattern of improved retinal function and decreased inflammation in CXCL1^−/− mice compared to C57BL/6J wild-type mice at 12 hours following intraocular S. aureus infection,⁴⁴ as we did with CCL3^−/− mice (Figs. 1E, 1F, 3). This similarity in effect at 12 hours post-infection could be due to neutrophil predominance at an early time point in infection and is further supported by the previously documented importance of CCL3 in neutrophil recruitment and activation.^39,53,54 We found improved retinal function despite the increased levels of inflammation seen at 24 hours, again raising the possibility of CCL3-mediated mechanisms for influencing retinal function beyond changes due to inflammatory cell influx (Figs. 1G, 1H, 3). The improved retinal function retention at both the early and late time points suggests that CCL3 plays a more important role than CCL2 in governing the innate immune response and the course and severity of S. aureus endophthalmitis. 

In the current study, we assessed the effects of CCL2 and CCL3 at early and late time points of infection on retinal function and inflammation. We previously examined the effects of an absence of CXCL1 up to 36 hours post-infection and the absences of CXCL2 and CXCL10 up to 24 hours post-infection and observed no differences.⁴⁴ Because of the lack of changes seen at the later time point, we did not extend that study further. We paralleled this experimental design in the current study, analyzing infection up to 24 hours to observe whether CCL2 or CCL3 exerted an influence on retinal function later in the infection timeline.⁴⁴ Further studies into the specific mechanism responsible for the increased MPO levels despite improved retinal function seen in this study would be beneficial, such as examining the degree of respiratory burst production by neutrophils in or near the retina. Overall, the findings regarding changes in retinal function support our hypothesis regarding the potential roles of CCL2 and CCL3 in S. aureus endophthalmitis. However, the results we observed on inflammation do not seem to support our initial hypothesis, implicating further complexity and suggesting potential areas of study into the specific roles of these chemokines at early and later time points in S. aureus endophthalmitis. 

Over the past several years, we have studied various chemokines and their roles in the innate immune response to Gram-positive pathogens in bacterial endophthalmitis—namely, B. cereus and S. aureus.^43,44,48 The CXCL chemokines CXCL1, CXCL2, and CXCL10 and the CCL chemokines CCL2 and CCL3 contributed to intraocular inflammation and retinal function decline in B. cereus endophthalmitis.^43,48 Due to the fulminant nature of B. cereus infection, time points were only gathered to 16 hours post infection in this experiment. In S. aureus endophthalmitis, the absence of CXCL1, CCL2, and CCL3 but not CXCL2 or CXCL10 resulted in improved retinal function retention at 12 hours post-infection (Figs. 1A, 1B, 1E, 1F).⁴⁴ Importantly, among the chemokines that we tested, only the absence of CCL3 improved retinal function retention at the later 24-hour time point. Taken together, these observations suggest an integral role for CCL3 in governing severity and outcomes during S. aureus endophthalmitis, although the mechanism of doing so may be more complex than regulation of immune cell entry into the eye. This further suggests that CCL3 might serve as a target for therapeutic intervention, given the effect of CCL3 disruption on outcomes at both earlier and later time points during infection. This work highlights the potential future utility of targeting chemokines as a potential adjunct therapy to antibiotics in assuaging the deleterious effects of these Gram-positive causes of bacterial endophthalmitis. 

Acknowledgments

The authors thank Roger Astley (Department of Ophthalmology, University of Oklahoma Health Sciences Center) and the Dean McGee Eye Institute Animal Facility for their invaluable technical assistance. 

Supported by a National Eye Institute P30-Center Core Grant for Vision Research (P30EY021725) awarded to the University of Oklahoma Health Sciences Center for development of the Live Animal Imaging and Functional Analysis Module and the Cellular Imaging and Morphometric Analysis Module; by grants from the National Institutes of Health (R21EY028066, R01EY032073, R21EY021802, and R01EY028810 to MCC); and by an unrestricted grant to the Dean A. McGee Eye Institute from Research to Prevent Blindness. 

This study was presented at the 2023 ARVO Annual Meeting, April 23–27, 2023, New Orleans LA. 

Disclosure: A.C. Parrott, None; P.S. Coburn, None; F.C. Miller, None; A.L. LaGrow, None; M.H. Mursalin, None; M.C. Callegan, None