The Glycoprotein 340’s Scavenger Receptor Cysteine-Rich Domain Promotes Adhesion of Staphylococcus aureus and Pseudomonas aeruginosa to Contact Lens Polymers

ABSTRACT

Contact lenses are biomaterials worn on the eye to correct refractive errors. Bacterial adhesion and colonization of these lenses results in adverse events, such as microbial keratitis. The adsorption of tear proteins to contact lens materials enhances bacterial adhesion. Glycoprotein 340 (Gp340), a tear component, is known to promote microbial colonization in the oral cavity; however, it has not been investigated in any contact lens-related adverse event. Therefore, this study examined the adsorption of Gp340 and its recombinantly expressed scavenger receptor cysteine-rich (iSRCR₁^Gp340) domain on two common contact lens materials, etafilcon A and lotrafilcon B, and the concomitant effects on the adherence of clinical isolates of microbial keratitis causative agents, Pseudomonas aeruginosa (PA6206; PA6294), and Staphylococcus aureus (SA38; USA300). Across all strains and materials, iSRCR₁^Gp340 enhanced adherence of bacteria in a dose-dependent manner. However, _iSRCR₁^Gp340 did not modulate the lysozyme’s or lactoferrin’s effects on bacterial adhesion to the contact lens. The Gp340 binding serine-rich surface protein (SraP) significantly enhanced the binding of USA300 to iSRCR₁^Gp340-coated lenses. In addition, iSRCR₁^Gp340-coated surfaces had significantly diminished biofilms with the SraP mutant (ΔSraP), and there was a further reduction in biofilms with the sortase A mutant (ΔSrtA), indicating the likely involvement of additional surface proteins. Finally, the binding affinities between iSRCR₁^Gp340 and SraP were determined using surface plasmon resonance (SPR), where the complete SraP binding region displayed nanomolar affinity, whereas its smaller fragments adhered with micromolar affinities. This study concludes that Gp340 and its SRCR domains play an important role in bacterial adhesion to the contact lens.

INTRODUCTION

Contact lenses are a medical device worn on the surface of the eye to correct refractive errors, improve cosmesis, and deliver ocular drugs. Despite its usefulness, contact lens wear is associated with adverse events, such as microbial keratitis, contact lens-related acute red eye, contact lens-related peripheral ulcers, and infiltrative keratitis (1–3). These events are clinically relevant as they can result in significant vision loss and in contact lens discomfort, and cause patients to drop out of contact lens wear (4). The adherence of bacteria to a worn contact lens is considered a primary factor in the development of these complications (1). During contact lens wear, proteins in the tear film deposit on the contact lenses and promote microbial adherence (1, 5). Tears contain more than a thousand proteins (6) that exert various functions, such as antimicrobial defense, lubrication, wound healing, and regulation of an inflammatory response (7). The adsorption of tear proteins and subsequent bacterial adhesion depend on several factors. Key factors among these are the effective (net) charge of the protein, modality or period of lens wear, properties of the lens materials (such as water content, surface charge, and hydrophobicity), and the surface characteristics of the adhering bacteria (1, 8–10).

Tear proteins, such as lysozyme and lactoferrin, promote adhesion of bacteria when they adsorb to contact lens material (5, 11). However, this enhanced bacterial adhesion does not potentiate the risk of microbial keratitis because these proteins have antibacterial functions (12–14). Among the many tear proteins, glycoprotein 340 (Gp340) is known to promote microbial infections (15–18). Glycoprotein 340 is a normal component of mucosal fluids, such as tears, saliva, and breast milk, and is expressed in mucosal epithelial tissues, such as the gastrointestinal tract, oral cavity, lung alveoli, and pancreas (16, 19). Glycoprotein 340, also known as salivary agglutinin and “deleted in malignant brain tumor 1” (DMBT1), is a 340 kDa innate immune protein that belongs to the scavenger receptor cysteine-rich (SRCR) superfamily (15, 16, 19). It contains multiple domains: SRCR, SRCR interspersed domain (SID), C1r/C1s, urchin embryonic growth factor and bone morphogenetic protein-1 (CUB), and zona pellucida (ZP) (16, 20, 21). Functionally, Gp340 can either be beneficial or harmful depending on its form or conformation (15). In the solution form or soluble conformation, Gp340 aggregates and inhibits pathogenic microbes, such as Streptococcus mutans, influenza A virus, and HIV-1 (15, 16). However, it promotes microbial attachments in dental caries and vaginal epithelial transcytosis of HIV-1 when it is surface-adsorbed (15–18).

On the ocular surface, Gp340 is a normal component of tears and the ocular surface tissues, lacrimal gland, cornea, and conjunctiva (22–24). Tear Gp340 has been shown to inhibit twitching motility of P. aeruginosa and promote corneal wound healing (25). While it was previously shown to bind to contact lens polymers (26), the impact of contact lens-bound Gp340 in adverse events, such as microbial keratitis, has yet to be elucidated.

Given that the surface adsorption of Gp340 can promote infections (17, 18), we hypothesized that the adsorption of Gp340 on a worn contact lens would promote bacterial adhesion and contribute to contact lens-related microbial keratitis. Therefore, this study aimed to examine the potential adsorption of Gp340 and its recombinantly-expressed first SRCR domain (iSRCR₁^Gp340) on two common commercially available contact lens polymers, etafilcon A and lotrafilcon B (Table 1), and the concomitant effect on the adhesion of the pathogenic bacteria, P. aeruginosa and S. aureus. In addition, this study explored the components on the microbial surface that aid in bacterial binding to the lens surface through Gp340. Understanding the mechanisms by which tear protein binds to contact lens materials and how it enhances microbial infections and/or other adverse events is crucial toward developing therapeutical interventions.

TABLE 1.

Characteristics of the lenses used in the in vitro studies

	Etafilcon A (Acuvue 2)	Lotrafilcon B (air optix)
FDA group	IV	V
diam (mm)	14.0	14.2
Base curve (mm)	8.3	8.6
Water content (%)	58	33
Oxygen transmissibility  (Dk/t at 35°C)	20	138
Surface treatment	None	25 nm plasma coating with high refractive index
Principal monomers	polyHEMA + MA	DMA + TRIS + siloxane

DMA, N,N-dimethylacrylamide; MA, methacrylic acid; polyHEMA, poly-2-hydroxyethyl methacrylate; TRIS, trimethylsiloxy silane.

In this study, we report that Gp340 and its recombinantly expressed scavenger receptor cysteine-rich (iSRCR₁^Gp340) domain adsorb onto etafilcon A and lotrafilcon B contact lens polymers. In addition, the causative agents of microbial keratitis, clinical isolates of Pseudomonas aeruginosa (PA6206; PA6294) and Staphylococcus aureus (SA8; USA300), displayed enhanced adherence in the presence of iSRCR₁^Gp340 in a dose-dependent manner. This study also determined that _iSRCR₁^Gp340 does not affect lysozyme- and lactoferrin-mediated bacterial adhesion to the contact lens. In addition, the interaction between S. aureus and iSRCR₁^Gp340 is predominantly mediated by the bacterial surface protein SraP. The SraP mutant (ΔSraP) and the sortase A mutant (ΔSrtA) displayed a significant reduction in biofilm formation. Finally, using surface plasmon resonance (SPR), we quantified the nanomolar affinity interaction between iSRCR₁^Gp340 and SraP. This study concludes that Gp340 and its SRCR domains play an important role in mediating bacterial adhesion to the contact lens.

RESULTS

Gp340 and iSRCR₁^Gp340 adsorb on etafilcon A and lotrafilcon B.

To determine the effect of contact lens-adsorbed Gp340 and iSRCR₁^Gp340 on bacterial adhesion, we first assessed whether Gp340 adsorbed on worn etafilcon A (more hydrophilic) and lotrafilcon B (more hydrophobic) lenses. We obtained 10 etafilcon A and 10 lotrafilcon B lenses from contact lens wearers, and, using a dot blot assay, confirmed the adsorption of Gp340 to both lens materials. This established that tear-secreted Gp340 adsorbs onto contact lens surfaces (Fig. 1A). Using a sandwich enzyme-linked immunosorbent assay (ELISA), we also determined the amount of tear Gp340 that adsorbs on the two lens polymers as 14.64 ng/lens and 22.49 ng/lens for etafilcon A and lotrafilcon B, respectively. Subsequently, we investigated whether the recombinantly expressed iSRCR₁^Gp340 would adsorb onto these lens polymers. Our results showed that iSRCR₁^Gp340 adsorbed to all of the lenses in a dose-dependent manner (Fig. 1B and C; Kruskal-Wallis H test, P = 0.017 for etafilcon A and P = 0.007 for lotrafilcon B). More importantly, at coating concentrations of 50, 500, or 5000 pg/μL, iSRCR₁^Gp340 adsorbed at 25%, 34%, or 22% higher, respectively, for lotrafilcon B compared to etafilcon A.

FIG 1.

Gp340 and iSRCR₁^Gp340 adsorbed on etafilcon A and lotrafilcon B lenses. Total tear protein was extracted from patient-worn etafilcon A and lotrafilcon B lenses and probed for Gp340 using a dot blot assay. Gp340 was detected on both polymers (A). Unworn etafilcon A and lotrafilcon B lenses were also coated in iSRCR₁^Gp340 solutions of varying concentrations (0 to 5000 pg/μL) and iSRCR₁^Gp340 binding to the lenses was observed with fluorescence microscopy (B to C). Scale bar for the images in B and C is 100 μm.

iSRCR₁^Gp340 promotes bacterial adhesion to the contact lens.

Glycoprotein 340 immobilized on the tooth surface and vaginal epithelial cells are known to promote infection (15–18). By coating etafilcon A and lotrafilcon B lenses in an iSRCR₁^Gp340 solution (50, 500, or 5000 pg/μL) or in phosphate-buffered saline (PBS; control) and subsequently incubating with clinical isolates of microbial keratitis, SA38, USA300, PA6206, and PA6294, we determined the effect of iSRCR₁^Gp340 on bacterial adhesion. Across all strains, lenses coated with 5000 pg/μL of iSRCR₁^Gp340 exhibited the largest amounts of adherent bacteria (Fig. 2; Kruskal-Wallis H test with pairwise comparisons, P < 0.05). At 500 pg/μL, USA300 exhibited higher adherence to both etafilcon A and lotrafilcon B, whereas PA6294 showed higher adherence to etafilcon A (Fig. 2; Kruskal-Wallis H test with pairwise comparisons, P < 0.05). At 50 pg/μL there were no differences in adhesion across all strains (P > 0.05). Finally, irrespective of the lens type, at every iSRCR₁^Gp340 coating concentration, the P. aeruginosa strains adhered in higher numbers compared to the S. aureus strains (Fig. 2).

FIG 2.

iSRCR₁^Gp340 enhances bacterial adhesion to etafilcon A and lotrafilcon B lenses. Lenses were coated in iSRCR₁^Gp340 solutions of varying concentrations (0 to 5000 pg/μL) for 12 h and subsequently incubated with bacterial suspension consisting of SA38 (A), USA300 (B), PA6206 (C), or PA6294 (D) for 24 h. The amount of total bacterial adhesion was determined using a live/dead bacterial assay kit. Significantly higher adhesion on iSRCR₁^Gp340-coated lenses (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). Error bars represent the standard error.

Contact lens-adsorbed iSRCR₁^Gp340 does not inhibit the proliferative ability of adherent bacteria.

In the case of the lens-adhered bacteria, only viable and culturable bacteria pose the threat of contact lens-related microbial keratitis. Through culturing and counting the colony forming units (CFU) of contact lens-adherent bacteria, we determined the proliferative capacity of iSRCR₁^Gp340-mediated adherent bacteria. Both etafilcon A and lotrafilcon B coated in 5000 pg/μL iSRCR₁^Gp340 displayed higher quantities of viable culturable bacteria (Fig. 3; Kruskal-Wallis H test with post hoc pairwise comparisons, P < 0.05). At 500 pg/μL coating, only SA38 showed a significant difference in viable, culturable bacteria with etafilcon A, whereas USA300, PA6206, and PA6294 did not. At 50 pg/μL there were no differences observed in the viable, culturable, adherent bacteria across all strains and materials (Fig. 3; Kruskal-Wallis H test with post hoc pairwise comparisons, P > 0.05).

FIG 3.

iSRCR₁^Gp340 does not affect the ability to culture bacteria that are adherent on iSRCR₁^Gp340-coated lenses. SA38 (A), USA300 (B), PA6206 (C), and PA6294 (D) were detached from the lens and plated on TSB agar plates, and the number of colonies were enumerated. Significantly higher bacterial colonies were observed on iSRCR₁^Gp340-coated lenses (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). Error bars represent the standard error.

_iSRCR₁^Gp340 does not modulate effects of lysozyme or lactoferrin on bacterial adhesion to contact lens.

The adsorption of lysozyme or lactoferrin on contact lens polymers enhances bacterial adhesion (1, 5, 11), and Gp340 interacts with these two abundant tear proteins (20, 27, 28). To determine whether iSRCR₁^Gp340 modulates the lysozyme- or lactoferrin-mediated bacterial adhesion to the contact lens, the amounts of bacterial adhesion on lenses coated in iSRCR₁^Gp340 (500 pg/μL), lysozyme (2 μg/μL), lactoferrin (1.8 μg/μL), lysozyme + iSRCR₁^Gp340, or lactoferrin + iSRCR₁^Gp340 were compared. Across all strains and lens types, lysozyme- or lactoferrin-coated lenses had significantly higher bacterial adhesion compared with the lenses coated in iSRCR₁^Gp340 (Fig. 4; one-way analysis of variance (ANOVA) with pairwise comparisons, P > 0.05). The addition of iSRCR₁^Gp340 to lysozyme or lactoferrin did not influence bacterial adhesion across all strains and lenses (Fig. 4; one-way ANOVA with pairwise comparisons, P > 0.05).

FIG 4.

iSRCR₁^Gp340 does not modulate lactoferrin- or lysozyme-mediated bacterial adhesion on etafilcon A and lotrafilcon B. Lenses were coated in solution containing iSRCR₁^Gp340 (500 pg/μL), lactoferrin (1.8 μg/μL), lactoferrin + iSRCR₁^Gp340 (1.8 μg/μL + 500 pg/μL), lysozyme (2 μg/μL) or lysozyme + iSRCR₁^Gp340 (2 μg/μL + 500 pg/μL) followed by incubation with the bacterial suspension consisting of SA38 (A), USA300 (B), PA6206 (C), or PA6294 (D) for 24 h. Significantly higher adhesion on lactoferrin- or lysozyme-coated lenses (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). Error bars represent the standard error.

SraP adhesin is involved in S. aureus adhesion to iSRCR₁^Gp340-coated lenses.

The S. aureus surface adhesin SraP/SasA adheres to Gp340 (29, 30). Hence, we tested for its role in promoting bacterial adherence to lenses that were coated with iSRCR₁^Gp340. Sortase A covalently anchors bacterial surface proteins to the peptidoglycan cell wall using the LPxTG motif in Gram-positive bacteria (31–34). The USA300 sortase A mutant, therefore, will not display LPxTG-containing surface proteins and, thus, aids the study of S. aureus adhesion to iSRCR₁^Gp340-coated lenses in the presence or absence of these proteins. Here, lenses coated in either iSRCR₁^Gp340 (500 pg/μL) or PBS (control) were tested for adherence with wild-type USA300 (USA300 WT), the SraP mutant (USA300ΔSraP), and the sortase A mutant (USA300ΔSrtA). Both USA300 WT and USA300ΔSraP adhered to iSRCR₁^Gp340-coated lenses in larger amounts compared to the control lenses (Fig. 5A and B; independent t test, P < 0.05), whereas the USA300ΔSrtA adhered at similar levels to PBS- and iSRCR₁^Gp340-coated lenses. Upon normalizing bacterial adherence of the mutants to the controls, the SraP mutant had significantly reduced binding compared to the wild-type (Fig. 5C and D), indicating that SraP mediates the interaction of S. aureus with iSRCR₁^Gp340-coated lenses. Finally, the lowest adherence to the lens surface was observed with the sortase A mutant (Fig. 5C and D), suggesting that additional surface adhesins are involved in S. aureus adhesion to iSRCR₁^Gp340-coated lenses.

FIG 5.

SraP plays a significant role in USA300 adhesion to iSRCR₁^Gp340-adsorbed etafilcon A and lotrafilcon B. Etafilcon A and lotrafilcon B lenses coated in a solution containing iSRCR₁^Gp340 (500 pg/μL) or PBS (control) were incubated with USA300 WT, USA300ΔSraP, or USA300ΔSrtA, and total bacterial adhesions were determined. Higher adhesion on iSRCR₁^Gp340-coated lenses compared to PBS-coated lenses for USA300 WT and USA300ΔSraP only (A to B). Significantly lower USA300ΔSraP and USA300ΔSrtA adhesion to iSRCR₁^Gp340-coated lens compared with that of USA300 WT (C to D; *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). Error bars represent the standard deviation.

SraP and other surface proteins on S. aureus USA300 mediate biofilm formation on an _iSRCR₁^Gp340-coated surface.

The results from the previous section showed that SraP facilitates the interaction between USA300 and _iSRCR₁^Gp340. To further confirm the role of SraP, we analyzed biofilms formed by USA300 WT, USA300ΔSraP, or USA300ΔSrtA in _iSRCR₁^Gp340-coated surfaces. Significant reductions in biofilms were observed with the SraP and sortase A mutants compared to the USA300 WT (Fig. 6; Kruskal-Wallis H test with pairwise comparisons, P < 0.05). However, between the two mutants, the sortase A mutant resulted in a higher reduction in biofilm, indicating that other surface adhesins of S. aureus bind to _iSRCR₁^Gp340 and agreeing with our results from the lens studies.

FIG 6.

SraP mediates USA300 biofilm formation on iSRCR₁^Gp340-coated solid surfaces. Ninety six-well microtiter plate coated with 50 μg/mL iSRCR₁^Gp340 was incubated for 24 h with cultures of USA300 WT, USA300ΔSraP, or USA300ΔSrtA, and the relative sizes of biofilm were measured on a plate reader. Significantly smaller biofilm sizes were observed with USA300ΔSraP and USA300ΔSrtA compared to USA300 WT (**, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). Error bars represent the standard deviation.

SraP binds to iSRCR₁^Gp340 with nanomolar affinity.

We analyzed the kinetics of the interaction of iSRCR₁^Gp340 with SraP using surface plasmon resonance (SPR) studies with the SraP constructs as analytes and iSRCR₁^Gp340 as the ligand. The full-length binding region (BR) of SraP that consists of L-lectin, β-grasp fold (β-GF), CDHL1 (cadherin-like), and CDHL2 domains (30) adhered with nanomolar affinity. The two truncated SraP fragments, namely L-lectin-β-GF-CDHL1^SraP and β-GF-CDHL1-CDHL2^SraP, also adhered with micromolar affinity, 2 orders of magnitude less than the full-length BR region of SraP.

SraP contributes to S. aureus adhesion to patient-worn contact lenses.

Having established that SraP promotes the adherence of S. aureus to iSRCR₁^Gp340-coated lenses, we determined the role of SraP in the adhesion of S. aureus to etafilcon A and lotrafilcon B lenses worn by patients. These lenses were incubated with USA300 WT, USA300ΔSraP, or USA300ΔSrtA, and the quantities of adherent bacteria were determined for all three strains. Significantly different amounts of adhesion were observed between the three USA300 strains across the two lens types (Fig. 7A and B; one-Way ANOVA, P < 0.05). The ΔSraP mutant did not show a statistically significant reduction in adhesion compared to the wild-type (etafilcon A, P = 0.09; lotrafilcon B, P = 0.34); however, the ΔSrtA mutant showed significant decreases in adherence to etafilcon A (P < 0.0001) and lotrafilcon B (P = 0.0001) lenses. In relative terms, USA300ΔSraP displayed 22% (etafilcon A) and 20% (lotrafilcon B) reductions in bacterial adherence, whereas USA300ΔSrtA showed 68% (etafilcon A) and 74% (lotrafilcon B) reductions in bacterial adherence (Fig. 7C and D).

FIG 7.

SraP and other CWA proteins contribute to adhesion of USA300 to contact lenses worn by patients. Patient-worn contact lenses were incubated with cultures containing USA300 WT, USA300ΔSraP, or USA300ΔSrtA. Adherent bacteria were detached and plated on TSB agar plates, and the number of colonies were enumerated (A to B). Relative to USA300 WT (C to D), the ΔSraP mutants were reduced by 21.86% (etafilcon A) and 20.21% (lotrafilcon B), and the ΔSrtP mutants were reduced by 67.63% (etafilcon A) and 74.05% (lotrafilcon B). These results indicated that other cell wall-associated surface proteins on USA300 contributed to the adherence. (ns, no significant difference; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). Error bars represent standard error.

DISCUSSION

Glycoprotein 340 is a normal tear component expressed by the lacrimal gland, cornea, and conjunctiva (19). It has been previously shown to adsorb on contact lenses (26). However, whether this adsorption contributes to contact lens-related adverse events, such as microbial adhesion or infection, remains to be determined. With the hypothesis that Gp340 would bind to the contact lens and enhance the adhesion of bacteria through microbial surface proteins, this study aimed to investigate the adsorption of Gp340 and its first SRCR domain (iSRCR₁^Gp340) on contact lens polymers, the possible impact of iSRCR₁^Gp340 on bacterial adhesion, and the microbial proteins that mediate this interaction.

The adsorption of Gp340 to patient-worn etafilcon A and lotrafilcon B lenses were investigated because any effect induced by iSRCR₁^Gp340 on bacterial adhesion would depend on the extent of Gp340 binding. Dot blot analysis of the total tear protein extracted from worn lenses confirmed that Gp340 binds to both lens polymers (Fig. 1), and that the quantity of bound Gp340 was 14.64 ng/lens for etafilcon A and was 22.49 ng/lens for lotrafilcon B. This finding laid the foundation for further assessments of iSRCR₁^Gp340 binding to lens material and the concomitant effects, if any, in contact lens-related adverse effects. Comparatively, iSRCR₁^Gp340 binding to lotrafilcon B was higher than that of etafilcon A. To explain this, we considered two major factors that are known to influence protein adsorption to biomaterials. These are the surface properties of the material and the effective charge of the protein that is determined by the isoelectric point (pI) and the pH of the medium containing the protein (35). The theoretical pI of iSRCR₁^Gp340 is 5.96 (ExPASy server [36]) and, thus, it would carry a net negative charge in PBS buffer whose pH is 7.4. Because etafilcon A and iSRCR₁^Gp340 both have an anionic surface, their repulsions could have led to lower binding, whereas the nonionic surface on lotrafilcon B results in higher deposition of iSRCR₁^Gp340 on its surface. These observations are consistent with a previous study by Zhao et al. (26), which reported a higher frequency of Gp340 deposition on lotrafilcon B.

Previously, investigating the influence of contact lens-adsorbed proteins on bacterial adhesion, Subbaraman et al. (5) reported higher adhesion for P. aeruginosa strains compared to S. aureus. In subsequent studies, Subbaraman et al. (11) further reported that the amount of P. aeruginosa bound to etafilcon A was higher (41 to 51 times) than that of S. aureus, whereas P. aeruginosa showed higher adhesion (26 to 85 times) to lotrafilcon B compared to that of S. aureus. Similarly, Borazjani et al. (37) reported higher adhesion (185 times) of P. aeruginosa to etafilcon A compared to that of S. aureus. In this study, the adherence of bacteria that was mediated by iSRCR₁^Gp340 displayed a similar trend, where P. aeruginosa strains adhered in higher quantities compared to that of S. aureus strains across all coating concentrations of iSRCR₁^Gp340 and lens polymers (Fig. 2).

Four type III secretion system (T3SS) effector toxins, namely ExoS, ExoT, ExoU, and ExoY, have been identified in P. aeruginosa, but all four are rarely present in any given strain (38–40). While three are conserved, most strains carry either the ExoS or ExoU gene. The invasive strain, PA 6294, has ExoS but not ExoU (ExoS⁺/ExoU-). The cytotoxic strain, PA 6206, on the other hand, has ExoU but lacks ExoS (ExoU⁺/ExoS-). Despite this difference, in the current study, the two strains had similar adherence to both lens polymers under all conditions, suggesting that their adhesion to Gp340-coated contact lenses is independent of the effector toxins and is consistent with a previous observation by Shen et al. (41).

When bacteria adhere to the lens surface, the estimated total bacterial counts would normally include both viable and dead cells. Because viable adherent bacteria are the ones that could propagate and potentially become infectious, and, among these, some may remain in a viable but nonculturable state (42), we also investigated the proliferating ability of the contact lens-bound bacteria mediated by iSRCR₁^Gp340. Our results showed that iSRCR₁^Gp340 did not impact the ability of the adherent bacteria to proliferate because the patterns of total adhesion and bacterial proliferation were similar (Fig. 3).

Investigations into the role of two other major tear proteins, lactoferrin and lysozyme, showed that each enhanced bacterial adhesion more than iSRCR₁^Gp340 (Fig. 4). This could partly be explained by the higher relative physiological concentrations of lactoferrin (1.8 μg/μL) and lysozyme (2 μg/μL) compared to iSRCR₁^Gp340 (500 pg/μL). At these higher concentrations, both lactoferrin and lysozyme would potentially adsorb on the lenses much more than iSRCR₁^Gp340, resulting in them displaying higher bacterial adhesion. While both lactoferrin and lysozyme have broad-spectrum antibacterial actions (14, 43) and can adversely affect the ability of any bound bacteria to cause infection, the effect of Gp340 during infection is dependent on its conformation. It is well known that Gp340 exists in two different conformations: soluble and immobilized (surface-associated/surface-bound) forms (15, 16, 19). In the soluble form, it inhibits cariogenesis, influenza A virus, and oral HIV-1 transmission (15). However, in the surface-bound conformation, it promotes cariogenesis and vaginal HIV transmission (15–18). Thus, the presence of contact lens surface bound Gp340/iSRCR₁^Gp340, even if limited, could promote bacterial adherence and/or viability to propagate and increase the risk of infection. Thus, despite iSRCR₁^Gp340 showing comparatively lower bacterial adhesion, the potential risk still exists for infection-enhancing effects.

Previously, Nichols and Green-Church (23) reported an increased level of tear Gp340 in subjects with contact lens-related dry eye in a study investigating the tear film proteome of contact lens wearers. With our studies conducted at physiological concentrations (500 pg/μL) in tears, iSRCR₁^Gp340 increased the adhesion of only two bacterial strains, PA6294 (etafilcon A) and USA300 (both polymers), but much higher adhesion was observed among all the strains at10-fold higher concentration (5000 pg/μL) (Fig. 2). It is here that we reason that deposition of Gp340 onto a contact lens surface will be cumulative and perhaps much larger than the concentration in tears, which, in turn, would increase bacterial adherence to a worn contact lens and elevate the risk of contact lens-related complications, such as microbial keratitis and peripheral ulcers. Further investigation into the potential effects of Gp340 during infections related to dry eye is needed because contact lens wear increases the risk of infection in dry eye disease (44).

As the third component of this study, the microbial component that mediates adherence of S. aureus to iSRCR₁^Gp340 was evaluated. In S. aureus, among the cell wall-anchored (CWA) proteins, SraP mediates the adherence to Gp340 (29), which is partly mediated by the glycosylation present on Gp340. The binding region (BR) of SraP at its N terminus consists of an L-lectin domain followed by one β-grasp fold (β-GF) and two tandem cadherin-like (CDHL) modules (29). The L-lectin domain binds one Ca²⁺ ion and coordinates the adhesion to host cells by recognizing the N-acetylneuraminic acid (Neu5Ac) of extracellular receptors (30). The CDHL domains also bind Ca²⁺ and, together with the β-GF domain, they project the L-lectin domain from the bacterial cell surface (30, 31). The CDHL domains are known to promote cell-to-cell accumulation and biofilm formation (30, 31).

The ΔSraP mutant displayed diminished binding to the contact lenses that were coated with iSRCR₁^Gp340, and the sortase A mutant (ΔSrtA) displayed even lower binding (Fig. 5), indicating that there are other surface components on S. aureus that are involved in the interaction with Gp340. This was further confirmed with patient-worn lenses, where there was a >20% reduction for the ΔSraP mutant and >65% reduction for ΔSrtA mutant (Fig. 7). Similar observations were made in our biofilm studies, where both the ΔSraP and ΔSrtA mutants reduced the biofilm on the iSRCR₁^Gp340-coated surface with higher reductions observed with ΔSrtA mutant (Fig. 6). Previous studies with S. aureus ISP479C showed that the SraP promoted biofilm formation (45), whereas a S. aureus NCTC 8325 SraP mutant did not show significantly decreased biofilm formation (30). Thus, the role of SraP in biofilm formation was adjudged to be strain-dependent (30). In addition to SraP, several other surface proteins, such as SasX, SdrC, FnBPA, FnBPB, and SasG, are known to promote biofilm formation (31), but their roles in Gp340-driven biofilm formation have yet to be determined. Taken together, the mutant adhesion and biofilm studies clearly established that SraP and other CWA proteins mediate interaction (adhesion and/or biofilm formation) of S. aureus on a Gp340-conditioned contact lenses.

Through SPR studies, the interaction between iSRCR₁^Gp340 and the SraP binding region was determined to be at a nanomolar affinity level (Table 2, Fig. S2). Studies with a combination of domains indicate that the constructs encompassing L-lectin-β-GF-CDHL1 and β-GF-CDHL1-CDHL2 display micromolar affinity (Table 2, Fig. S2), thus indicating cooperative elements that result in the nanomolar affinity are present within these domains. This study has now established the nanomolar interaction between the recombinant SraP BR domain and iSRCR₁^Gp340. Glycoprotein 340 is highly glycosylated (16, 20, 21, 24, 46, 47), and our biofilm and SPR binding studies confirmed that the L-lectin module would interact with sialylated/glycosylated moieties on Gp340 (30). In addition, the homophilic interactions of the CDHL domains in concert with L-lectin could further promote adhesion and biofilm formation.

TABLE 2.

SraP adherence to iSRCR₁^Gp340

Analyte	Ligand	ka (1/Ms)	kd (1/s)	KA (1/M)	KD (M)	χ2
L-lectin-β-GF-CDHL1-CDHL2SraP	iSRCR1Gp340	4.4 × 104	1.04 × 10−3	4.23 × 107	2.37 × 10−8	4.48
L-lectin-β-GF-CDHL1SraP	iSRCR1Gp340	1.24 × 103	2.34 × 10−3	5.31 × 105	1.88 × 10−6	3.28
β-GF-CDHL1-CDHL2SraP	iSRCR1Gp340	159	4.37 × 10−4	3.63 × 105	2.76 × 10−6	6.09

The results observed with SraP confirm previous studies (29) and has also now identified potential means to develop inhibitors to this interaction. For this to happen, the mechanistic aspects of this interaction are to be determined through site-directed mutagenesis studies to establish the binding-site/motif/region on SraP. While we have some idea of the interacting partners between Gp340 and S. aureus, little is known about the interacting partners with P. aeruginosa. Future studies could be aimed at the interaction between the minor pilins, FimU, PilV, PilW, PilX and PilE, in P. aeruginosa (48), which we suspect could be involved in binding to Gp340. Once these interactions are mapped at the amino acid level, the potential exists to selectively target pathogenic strains that are involved in Gp340-mediated contact lens-related microbial keratitis.

This study is the first to investigate the adsorption of Gp340 and its SRCR domain on a contact lens polymer and the potential effect on bacterial adhesion on the lens. This study has established that Gp340 and/or iSRCR₁^Gp340 bind to both etafilcon A and lotrafilcon B polymers and mediate the adhesion of various strains of S. aureus and P. aeruginosa to these lenses, that this interaction is mainly mediated by the surface protein SraP of S. aureus, which displays nanomolar affinity, and S. aureus potentially has other proteins on its surface that interact with Gp340.

MATERIALS AND METHODS

Contact lens polymers.

A hydrogel polymer, etafilcon A (Acuvue 2; Johnson & Johnson) and a silicone hydrogel polymer, lotrafilcon B (Air Optix; CIBA Vision) were investigated in this study. Table 1 summarizes the characteristics of the lenses used within in vitro studies.

Expression and purification of iSRCR₁^Gp340.

The recombinantly expressed scavenger receptor cysteine-rich (iSRCR₁^Gp340) domain of Gp340 was recombinantly expressed using the Drosophila S2 expression system as previously described by Purushotham and Deivanayagam (21). Briefly, a synthesized sequence-optimized DMBT1 gene was cloned in the recombinant expression vector, pMT/BiP/V5-HisA (Invitrogen, Inc.). The pMT/V5-HisA containing iSRCR₁^Gp340 was transfected stably into S2 cells (Invitrogen) using calcium phosphate transfection kit (Invitrogen, Inc.). These clones were expanded in selective growth media and copper sulfate (500 μM) and used to induce iSRCR₁^Gp340 expression. The growth medium was dialyzed to remove the copper ions, centrifuged, and the supernatant was filtered and loaded onto a HisTrap™ affinity column (GE Healthcare, Inc.) for purification using an AKTA purifier (GE Healthcare Inc.). Appropriate fractions containing iSRCR₁^Gp340 were pooled, dialyzed, and loaded onto a Mono Q column (GE Healthcare, Inc.) for further purification. Purified fractions identified from SDS-PAGE gels were pooled and used for the in vitro iSRCR₁^Gp340 adsorption and bacterial adhesion studies.

Cloning of S. aureus SraP constructs.

The SraP gene was custom-synthesized (GeneArt; Thermo Fisher Scientific), and subclones were developed from this gene. The three constructs, L-lectin-β-GF-CDHL1-CDHL2^SraP (245-751), L-lectin-β-GF-CDHL1^SraP (245-660), and β-GF-CDHL1-CDHL2^SraP (492-751) were designed with a C-terminal histidine tag utilizing the fast-digest restriction enzymes NcoI and XhoI (Thermo Fisher Scientific) for cloning into the pET23d vector (Novagen). The primers used for cloning are listed in Table S3. PCR amplification of the fragments was done using Phusion^TM DNA polymerase (Thermo Fisher Scientific) followed by digestion of both the PCR fragments and vector with appropriate enzymes. The products were ligated with T4 DNA ligase (New England BioLabs), transformed into E. coli DH5 cells and grown on LB-agar plates supplemented with ampicillin (50 μg/mL). Single colonies were grown in fresh 5 mL LB cultures, and the plasmids were harvested using the Mini-Prep kit (Zymo Research). DNA sequencing was carried out at the University of Alabama Birmingham Heflin Center, which confirmed the presence of the appropriate inserts in the pET23d vector. After confirmation, these plasmids were transformed into E. coli BL21(DE3) cells for protein expression.

Expression and purification of the S. aureus SraP binding region.

E. coli BL21(DE3) cells harboring the plasmids for each SraP construct were inoculated into a 20 mL starter Terrific Broth (TB) culture overnight at 37°C. The next morning, these cultures were transferred into shaker flasks containing 1 L of TB, and the cells were grown to an optical density at 600 nm (OD₆₀₀) of 1.0, induced with 1 mM IPTG for 5 h at 30°C, supplemented with additional antibiotics, and grown overnight at 18°C. These cells were harvested by centrifugation at 5000 × g for 20 min using a Beckman Avanti JL-25 centrifuge, and the cell pellets were resuspended in nickel affinity column binding buffer (50 mM Tris, pH 8.0, 500 mM sodium chloride) augmented with a cOmplete EDTA-free protease inhibitor (Sigma-Aldrich). E. coli cells were ruptured by sonication outfitted with a temperature Fisensor (Fisherbrand Sonicator) for a total of 5 min while maintaining a maximum temperature of 10°C. The lysed cells were then centrifuged at 35,000 RPM for 1 h using a Ti70 rotor, and the supernatant was collected and filtered through a 0.22 μm filter before being loaded onto a 20 mL HisPrep Nickel Column (GE Healthcare, Inc.). Using a first step gradient of 50 mM imidazole, the nonspecifically bound proteins were gently removed from the column and the bound protein was eluted with a 50 to 300 mM imidazole gradient. The purity of the eluted samples was again visualized using an SDS-PAGE gel, and appropriate fractions were pooled and dialyzed overnight into the Mono Q binding buffer (50 mM Tris, pH 8.0, 50 mM sodium chloride, and 1 mM EDTA). The dialyzed sample was filtered and loaded onto a Mono Q column (GE Healthcare, Inc.), and the protein was eluted with a 0 to 400 mM NaCl gradient. The purest single-banded fractions as identified by SDS-PAGE gels (Fig. S1) were then pooled and concentrated under 55 lb/in² nitrogen gas using an Amicon stirring concentrator. The protein concentration was measured using a modified method described elsewhere (49).

Surface plasmon resonance studies.

To determine the affinity coefficients between SraP and iSRCR₁^Gp340 surface plasmon resonance studies were performed. Serving as the ligand, iSRCR₁^Gp340 was immobilized on a CM5 chip (GE Healthcare, Inc.) using ethanolamine chemistry. Analytes L-lectin-β-GF-CDHL1-CDHL2^SraP, L-lectin-β-GF-CDHL1^SraP, and β-GF-CDHL1-CDHL2^SraP were injected over SRCR₁ at various concentrations that were optimized for each analyte (serial dilutions within 0.250 to 32 μM), and dissociations were measured for 600 s following injections. The running buffer used for all analytes was 10 mM HEPES pH 8.0, 150 mM NaCl, 1 mM CaCl₂. Each experiment was carried out in triplicate with a BIAcore2000 instrument. The sensorgrams were fitted using the BIAevaluation software (Biacore AB), where both the residuals and χ² values were refined to convergence and the results are presented in Table 2 and Fig. S2.

Determination of iSRCR₁^Gp340 adsorption on the contact lens.

(i) Ex vivo study. This portion of the study was approved by the Institutional Review Board of the University of Alabama at Birmingham (UAB) and followed the tenets of the Declaration of Helsinki (50). Ten etafilcon A and ten lotrafilcon B lenses worn for one month and were obtained from ten healthy, non-dry eye contact lens wearers (five subjects per lens type). The total protein adsorbed into each lens was extracted using trifluoroacetic acid-acetonitrile buffer as described earlier (51, 52). These extracted samples were then pooled and concentrated. The presence of Gp340 in the extracted protein was then detected by a dot blot assay after blotting equal amounts of total protein on polyvinylidene fluoride (PVDF) membrane and probing with monoclonal primary antibody (Thermo Fisher Scientific) and Alexa Fluor 488-conjugated IgG secondary antibody (Thermo Fisher Scientific). Unworn etafilcon A and lotrafilcon B lenses were included as a quality control.

The amount of Gp340 in tears that adsorbs on the lens polymers was quantitated using a sandwich ELISA kit for Gp340 (MyBioSource) and following manufacturer’s recommendation. Briefly, the extracted samples and Gp340 standards (10 to 0 ng/mL) were added to the wells of microtiter plates (precoated with anti-Gp340 antibody; MyBioSource) and incubated for 1 h at 37°C. After washing, anti-Gp340 antibody was added to the wells followed by the addition of a secondary detection antibody, all of which occurred at 37°C. Thereafter, substrate solution was added and incubated for 20 min after which a stop solution was added. The absorbances were then read at 450 nm and adjusted for background noise. Standard curves were generated with the four-parameter logistic model in the Gen 5™ analysis software (BioTek, version 3.08) and used to interpolate the concentrations of Gp340 in the samples.

(ii) In vitro study. To determine the adsorption of iSRCR₁^Gp340 on contact lens materials, unworn etafilcon A and lotrafilcon B lenses were placed in triplicate for 12 h in 1.2 mL of histidine-tagged iSRCR₁^Gp340 at the following concentrations: 5000, 500, and 50 pg/μL at 37°C. Lenses (n = 3 lenses per polymer) were also placed in 1.2 mL of protein-free phosphate-buffered saline (PBS) to serve as control. Because Gp340 binding is calcium dependent (20), 1.5 mM Ca²⁺ was added to each solution. Subsequently, each lens was washed, blocked, and probed with a primary 6× His Tag antibody (Thermo Fisher Scientific), followed by an Alexa Fluor 488-conjugated IgG (Thermo Fisher Scientific). Four different areas of the lens surface were randomly visualized and imaged using fluorescence microscopy. Prior to imaging, the entire lens surface was scanned to confirm that there was uniform binding of iSRCR₁^Gp340. Any lens that had tears or abrasions was excluded from the study. The observed fluorescence was considered a measure of iSRCR₁^Gp340-adsorption and was quantified using ImageJ densitometry analysis (53). The experiments were repeated two more times. Kruskal-Wallis H test with post hoc pairwise comparisons was used to analyze the difference in the amount of bound iSRCR₁^Gp340 between the different coating iSRCR₁^Gp340 concentrations for both polymers. The ratio of adsorbed iSRCR₁^Gp340 between etafilcon A and lotrafilcon B was used to compare the amounts of iSRCR₁^Gp340 binding between the two lens types at each coating concentration.

Role of iSRCR₁^Gp340 on bacterial adhesion.

(i) Bacterial culture. Two Gram-positive, pathogenic bacterial strains, S. aureus 38 (SA38) and methicillin-resistant S. aureus (MRSA) USA300, and two Gram-negative strains, P. aeruginosa 6206 (PA6206) and P. aeruginosa 6294 (PA6294), were investigated in this study. They were selected because they are etiological agents for infections on the surface of the eye. Specifically, SA38, PA6206 (cytotoxic strain), and PA6294 (invasive strain) are clinical isolates from human microbial keratitis (5, 54) and were a gift from Mark Willcox (University of New South Wales, Sydney, Australia). USA300 is one of the most prevalent community-associated MRSA strains in the United States and a leading candidate in health care associated keratitis (55–57). The method previously reported by Subbaraman et al. (5, 11) was adopted to prepare bacterial cultures from stocks stored at −80°C in 30% glycerol. Briefly, bacteria were grown overnight in Tryptic soy broth (TSB; Oxoid) at 37°C for 18 h. Bacterial cells were harvested by centrifugation (Eppendorf 5810) for 10 min (3000 rpm at 18°C) and resuspended in PBS, where the optical density of the bacterial suspension was adjusted to OD₆₆₀ = 0.3.

(ii) Coating of contact lens and bacterial adhesion assay. First, both etafilcon A and lotrafilcon B lenses were coated in triplicate in 1.2 mL of iSRCR₁^Gp340 at three different concentrations: 5000, 500, and 50 pg/μL for 12 h at 37°C (n = 3 lenses per coating concentration for each polymer). Lenses that were coated in protein-free PBS served as a negative control. The 500 pg/μL concentration was selected because it falls within the physiological range of tear Gp340 concentration, which was determined through preliminary studies to be 620 pg/μL. Afterwards, the lens was rinsed with PBS to remove any unbound iSRCR₁^Gp340. Subsequently, the protocol previously employed by Subbaraman et al. (5) was adapted to determine the effect of iSRCR₁^Gp340 on bacterial adhesion. Briefly, each lens was placed in 1.2 mL of bacterial suspension at 37°C for 24 h. Following this, the lens was rinsed three times in PBS for 30 s to remove loosely adhered bacteria. Thereafter, each lens was resuspended in 1.5 mL of PBS containing 200 μl of 0.25% trypsin-EDTA and vortexed for 1 min with a magnetic stir bar to detach the adhered bacteria. This detachment strategy homogenizes the lens and removes more than 99.9% of adherent bacteria (5). The total bacterial adhesion on each lens was measured using the bacterial viability assay kit (Abcam) and following manufacturer’s protocol. The assay utilizes two highly specific fluorescent reagents, each staining either dead bacteria or both live and dead (total) bacteria. The experiments were performed three times for each bacterial strain. The amount of total bacterial adhesion at the different coating concentrations of iSRCR₁^Gp340 were compared using Kruskal-Wallis H test with a P value < 0.05 denoting statistical significance. Post hoc pairwise comparisons were done using the Bonferroni correction.

Role of adsorbed iSRCR₁^Gp340 on the proliferating potential of contact lens-adhered bacteria.

Briefly, 100 μL of detached bacterial suspension was diluted 1000-fold in the neutralizing broth (Difco Laboratories) and then serially diluted 1:10, 1:100, and 1:1000. Fifty microliters of these serially diluted samples were plated in triplicate on nutrient agar and incubated at 37°C for 16 h. The number of CFU on each plate was counted, and, accounting for the dilutions, the total CFU per lens was estimated. The amounts of culturable, viable adherent bacteria at the different coating iSRCR₁^Gp340 concentrations were compared using Kruskal-Wallis H test with pairwise comparisons and the Bonferroni adjustment.

Modulation of the effect of lactoferrin and lysozyme on bacterial adhesion in the presence of iSRCR₁^Gp340.

(i) Coating of contact lens and bacterial adhesion assay. Based on previous studies and the established individual tear protein concentrations (58, 59), the following coating solutions were prepared and used for the study: (i) iSRCR₁^Gp340 (500 pg/μL), (ii) lysozyme (2 μg/μL), (iii) lactoferrin (1.8 μg/μL), (iv) lysozyme + iSRCR₁^Gp340, and (v) lactoferrin + iSRCR₁^Gp340. Etafilcon A and lotrafilcon B lenses were placed in triplicate 1.2 mL of each solution for 12 h at 37°C. Each lens was then rinsed three times with PBS to remove any unbound protein. Subsequently, each lens was placed in 1.2 mL of bacterial suspension (OD₆₆₀ = 0.3) and incubated at 37°C for 24 h. Each lens was rinsed and bound bacteria were detached as described in the previous section. The total amount of adherent bacteria was measured using the total/dead bacterial assay. The experiments were performed three times with each strain. A Mann-Whitney U test was used to compare the total bacterial adhesions between the iSRCR₁^Gp340 solution and each of the other coating solutions.

Role of SraP on bacterial adhesion to contact lens.

Three different genotypes of S. aureus USA300, including the wild-type (USA300 WT), SraP mutant (USA300ΔSraP), and sortase A mutant (USA300ΔSrtA) were acquired from the Center for Staphylococcal Research, University of Nebraska Medical Center, and used in this study. The mutant strains were created by transposon mutagenesis. Briefly, etafilcon A and lotrafilcon B lenses (n = 3 lenses per polymer type) were placed in 1.2 mL of _iSRCR₁^Gp340 solution (500 pg/μL) or PBS (control) as described in the previous sections. Cultures of USA300 WT, USA300ΔSraP, and USA300ΔSrtA were also prepared as described above. Each lens was incubated in 1.2 mL of bacterial suspension (OD₆₆₀ = 0.3) for 24 h at 37°C and subsequently rinsed. Adherent bacteria were detached as described in the previous sections. The amount of total adherent bacteria was measured using the total/dead bacterial assay. These experiments were repeated two more times for each polymer for all three USA300 strains. An independent t test was used to compare the total number of bacterial adhesions from the iSRCR₁^Gp340-coated lenses and control lenses for each USA300 genotype. To determine the effect of SraP on USA300 adhesion to iSRCR₁^Gp340-coated lenses, the adhesion of the bacteria to iSRCR₁^Gp340-coated lenses was first normalized to the control lens by dividing the iSRCR₁^Gp340-coated lens fluorescence by the mean control lens fluorescence. Subsequently, a one-way ANOVA with post hoc pairwise comparison was used to determine the differences in adhesion between the USA300 WT, USA300ΔSraP, and USA300ΔSrtA strains.

The impact of _iSRCR₁^Gp340/SraP interaction on S. aureus biofilm formation.

A biofilm assay was performed using the methods previously described with minor modifications (60). Briefly, frozen stocks of the S. aureus strains were inoculated in TSB and grown overnight at 37°C with 5% CO₂. Overnight cultures were then diluted into 5 mL fresh TSB medium (1:100) and grown at 37°C with 5% CO₂ until the cultures reached an OD₄₇₀ = 0.6. These growth phase cultures were further diluted 1:200 into fresh TSB containing 1% glucose, from which 200 μL was aliquoted into 96-well microtiter plates coated with 50 μg/mL _iSRCR₁^Gp340. Bacterial cells were grown for 24 h at 37°C with 5% CO₂ under static conditions. Wells with only TSB were included as controls. After the 24 h incubation period, bacterial growth in each well was measured at OD₄₇₀ and the cultures were gently removed. Nonadherent bacteria were removed by washing each well three times with 200 μL of PBS. Thereafter, the biofilms were fixed by heating the plate at 65°C for 1 h then stained with 150 μL of 0.1% (wt/vol) crystal violet for 5 min. Excess crystal violet stain was discarded, and the wells were washed to remove any residual dye then air-dried for 30 min at room temperature. Following this, 150 μL of 30% (vol/vol) glacial acetic acid was added to each well and incubated for 1 h under constant shaking. The resulting biofilm was analyzed by measuring absorbance at 595 nm. Eight replicates of the assay for each of the USA300 strains were studied, and the experiments were performed three times for each strain. The absorbance values were averaged to obtain the relative amount of biofilm formation. The absorbance values from negative control wells were subtracted from the positive wells, and the resultant absorbance represented the amount of the biofilm. Potential differences in bacterial growth rate were accounted for by normalizing OD₅₉₅ values to OD₄₇₀ values. The differences in the amount of biofilm formation between the three USA300 strains were determined by analyzing OD_595/470 values using a one-way ANOVA with post hoc pairwise comparisons.

The role of SraP in S. aureus adhesion to contact lenses won on the eye.

This component of the study was approved by the UAB Institutional Review Board and followed the tenets of the Declaration of Helsinki (50). Briefly 4 etafilcon A and 4 lotrafilcon B lenses, each worn for 28 d, were obtained from healthy, habitual contact lens wearers. After rinsing the lenses in PBS, each lens was cut into four equal parts and incubated with USA300 WT, USA300ΔSraP, or USA300ΔSrtA in TSB medium (OD₆₀₀ = 0.1) for 12 h at 37°C as described in an earlier section. One of the cut lenses was also incubated in bacteria-free TSB medium to serve as a quality control. Afterwards, the lens was rinsed in PBS to remove unbound bacteria. To determine the amount of bacterial adhesion on the lens, the adherent bacteria were detached and serially diluted (1:100, 1:1000, or 1:10000) with 50 to 100 μL plated on TSB agar plates, and the plates were incubated at 37°C for 14 h. The CFU per plate was enumerated and a one-way ANOVA with post hoc pairwise comparisons was used to determine the difference in CFUs between the three USA300 strains. A P value of  < 0.05 denoted statistical significance.