Phospholipase A₂ in Rabbit Tears: A Host Defense against Staphylococcus aureus

Abstract

purpose. This study analyzed rabbit tears for anti-staphylococcal activity, the role of phospholipase A₂ (PLA2) in this reaction, and the ability of enzyme inhibitors to promote bacterial survival.

methods. Contact lenses with Staphylococcus aureus were applied to scarified rabbit eyes. The colony-forming units (CFU) per cornea or lens were determined and pathology was scored by slit-lamp examination (SLE). The bactericidal activity was measured by incubating bacteria with rabbit tears or PLA2 at 33° or 37°C. Radiolabeled S. aureus was incubated with PLA2 or tears to quantify the release of a membrane component that was identified by thin-layer chromatography. Inhibitors of these reactions were also analyzed.

results. Application of Staphylococcus, on contact lenses, to rabbit corneas resulted in bacterial killing and limited inflammation. Incubation of tears and bacteria (1:1; v/v) in tryptic soy broth at 33°C decreased CFU approximately 4 logs. Tears (≥30 μl) or PLA2 (≥30 U) incubated with bacteria in phosphate-buffered saline were bactericidal. PLA2 (≥0.2 U) or tears (≥2 μl) cleaved bacterial membranes, liberating arachidonic acid. Spermidine or tetracaine inhibited cleavage of bacterial membranes by tears or PLA2 and spermidine promoted bacterial survival and growth in tears. Tears (60μ l) killed >99% of the bacterial inoculum, whereas bacteria incubated in tears plus spermidine approximately doubled in number.

conclusions. PLA2 in rabbit tears kills Staphylococcus by hydrolyzing bacterial membranes to release arachidonic acid. Spermidine and tetracaine inhibited PLA2 activity and spermidine protected Staphylococcus from PLA2 in rabbit tears.

Human skin includes in its normal flora abundant numbers of Staphylococcus. Despite the fact that the tear film has many nutrients that could foster bacterial growth, the corneal surface, unlike the skin, contains relatively few staphylococci. The tear film thus possesses a potent defense system that limits the replication and even the survival of these bacteria. Tears have long been recognized to contain nonspecific immune mechanisms active against bacteria, including complement, lysozyme, defensins, and lactoferrin. ^{1 2 3 4 5 6 7} These molecules coupled with specific antibody, especially IgA, are generally recognized as key defense components of tears. 

Weinrauch et al. ⁸ and Dominiecki and Weiss ⁹ have reported that phospholipase A₂ (PLA2) from leukocytes can rapidly kill Staphylococcus aureus. This enzyme cleaves arachidonic acid from the bacterial membrane ^{10 11} and, when such cleavage is extensive, membrane damage activates autolytic bacterial enzymes that destroy the bacterial cell wall. ¹² Such bactericidal activity of phospholipase has been reported to be active in human tears and has been suggested to be a part of the host defense system. ¹³  

S. aureus, the leading cause of keratitis in many human populations ^{14 15 16 17} can cause severe keratitis in rabbits after intrastromal injection of log phase bacteria into the cornea. ¹⁸ Unlike Pseudomonas, which can topically infect and replicate extensively in a scratched rabbit cornea, ¹⁹ S. aureus fails to replicate in a scarified rabbit cornea after topical inoculation. ²⁰ The fate of the inoculated bacteria has not been analyzed, but the likelihood is that the host defense system either directly kills these bacteria or removes the organisms from the corneal surface. The defensive mechanism active in the rabbit tear film has not been analyzed. 

Inhibition of the activity of host defense molecules such as phospholipase A₂ could alter the ability of tears to kill bacteria, and in doing so help elucidate the mechanisms that are important to host defense. Inhibitors of phospholipase A₂ include spermidine, a polyamine, and topical anesthetics such as tetracaine. ^{21 22} Spermidine indirectly inhibits the activity of phospholipase A₂ by sterically hindering the interaction with its substrate. ²¹ Two hypotheses have been proposed concerning the inhibition of phospholipase A₂ by tetracaine. ²² One hypothesis is the possibility that tetracaine interacts with the substrate of phospholipase A₂. The second hypothesis involves the binding of the anesthetic to phospholipase A₂ possibly displacing Ca²⁺ from the enzyme. 

The present study was undertaken to determine the mechanisms that prevent bacterial replication after topical Staphylococcus inoculation of the rabbit cornea. The findings emphasize the importance of phospholipase A₂ as a defensive bactericidal molecule of the rabbit tear film and illustrate the ability of a phospholipase A₂ inhibitor to compromise this potent ocular defense. 

Materials and Methods

Bacteria

S. aureus strain 8325–4 was used in this study and has been previously used in the rabbit intrastromal injection model of keratitis. ^{18 23 24} Bacteria were grown overnight in tryptic soy broth (TSB; Difco Laboratories, Detroit, MI) or minimal medium (M9) with amino acids ^{25 26} at 37°C and then subcultured to log phase under the same conditions. 

Contact Lenses

Soft contact lenses (Surevue, Type IV: 42% etafilcon A, 58% water; Vistakon, Johnson & Johnson, Jacksonville, FL) were placed into 1.5 ml of log phase bacteria (8.0 log colony-forming units [CFU]/ml) in TSB. After 1 hour at room temperature, the lenses were rinsed in phosphate-buffered saline (PBS; 0.1 M phosphate, 0.85% NaCl, pH 7.0) and three lenses, selected at random, were assayed to determine the number of bacteria bound. Lenses were assayed by homogenizing each in sterile PBS (3.0 ml), diluting the homogenate in PBS and inoculating the homogenate and dilutions onto tryptic soy agar plates (TSA; Difco Laboratories). The plates were incubated at 37°C for 24 hours, and the colony count was used to calculate the log number of bacteria per lens. Approximately 7.0 log CFU were bound to each lens. Lenses to be used as controls free of bacteria were placed into the sterile PBS and then aseptically rinsed with PBS. 

Rabbits

New Zealand White rabbits (2.0–3.0 kg) used in these studies were maintained in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Inoculation of Rabbit Eyes

Rabbits were anesthetized by SC injection of a 1:5 mixture of 100 mg/ml xylazine (Rompun; Miles Laboratories, Shawnee, KS) and 100 mg/ml ketamine hydrochloride (Ketaset; Bristol Laboratories, Syracuse, NY). One drop of proparacaine hydrochloride (0.5% Alcaine; Alcon Laboratories, Fort Worth, TX) was instilled into each eye just before starting the inoculation procedure. Corneas were scarified by making three horizontal scratches (∼8 mm long) with a 22-gauge needle. Contact lenses were then placed onto the eyes under the nictitating membrane and a lateral tarsorrhaphy was performed, as described previously. ¹⁹  

Quantification of S. aureus per Contact Lens or Cornea

Rabbits were killed by intravenous injection of sodium pentobarbital solution (100 mg/ml; The Butler Co., Columbus, OH) at either 24 or 48 hours after inoculation (n = 4 corneas per time point). Contact lenses were aseptically removed and cultured as described above. Corneas were aseptically removed, dissected, and homogenized as previously described. ^{18 23 24} Corneal homogenates were serially diluted in sterile PBS, aliquots (0.1 ml) were inoculated onto TSA, and the plates were incubated at 37°C for 24 hours. Colonies were counted, and CFUs per cornea were expressed as log values. 

Slit-Lamp Examination

Ocular pathology was graded by slit-lamp examination (SLE) with a Topcon S1–5D slit-lamp biomicroscope (Koaku Kikai K.K., Tokyo, Japan) using a scoring system that has been previously described. ²⁷ Seven parameters were scored on a scale of 0 (absent) to 4 (severe), including conjunctival injection, conjunctival chemosis, corneal infiltrate, corneal edema, fibrin in the anterior chamber, hypopyon, and iritis. Two masked observers performed all SLE scoring. Rabbit eyes were examined at 8, 24, 48, and 72 hours after inoculation. 

Collection of Rabbit Tears

Capillary tubes were placed into the cul-de-sac of normal rabbits and allowed to collect 5 to 10 μl of tears. Tears (1000 μl) from two collections per day from 8 to 10 rabbits were pooled and frozen in aliquots at −70°C until assayed. 

Bactericidal Assay of Rabbit Tears

In initial assays, tears (25–90 μl) were mixed in a 1:1 (v/v) ratio with log phase bacteria in TSB (∼7.0 log CFU, initial concentration) and incubated at either 33° or 37°C for up to 24 hours. Optimal bacterial growth is seen at 37°C, whereas the normal temperature of the rabbit corneal surface is 33°C. ²⁸ In subsequent assays, the bacteria were grown in M9 medium, centrifuged to a pellet, and then resuspended in PBS. Bacteria (100–200 CFU in 15 μl) were mixed with rabbit tears (1–60 μl), and the reactions were incubated at 33°C for 4 hours. Bacteria in PBS were also incubated with pure porcine pancreatic PLA2 (2–36 U; Sigma Chemical Co., St. Louis, MO) in Tris-HCl buffer (pH 7.5) and calcium chloride (CaCl₂; 2 mM) for 4 hours at 33°C. Aliquots were inoculated onto TSA plates or diluted in PBS and then inoculated onto TSA plates. The log number of CFUs was determined from the colony count after 24 to 48 hours of incubation. 

Radioactive Labeling of S. aureus

S. aureus was radioactively labeled based on a modification of the procedure described by Dominiecki and Weiss. ⁹ S. aureus strain 8325-4 was grown overnight in M9 medium with amino acids. An aliquot (200 μl) of these bacteria was subcultured in M9 (20 ml) with 0.5 μCi/ml of ¹⁴C-labeled oleic acid (100 μl; specific activity of 55 mCi/mM; ICN Biomedicals, Irvine, CA) and incubated at 37°C for 2.5 hours. After incubation, the bacteria were centrifuged to a pellet. The supernatant was removed, and the bacterial pellet was resuspended in TSB and then incubated for 30 minutes at 37°C. The labeled bacteria were washed in 1% bovine serum albumin (BSA; Sigma) and then resuspended in PBS. Bacteria were stored at −70°C until needed for use. 

PLA2 Radioactivity Assay

A modification of the procedure of Wright et al. ²⁹ was used to assay PLA2 activity. Radiolabeled bacteria (∼1 × 10⁶ CFU with 5700 cpm in 20 μl) were incubated with PLA2 in Tris-HCl buffer (pH 7.5) or rabbit tears, both with 1 mM CaCl₂, in a total volume of 250 μl at 33°C for 30 minutes. Reactions were terminated by adding 250 μl of ice-cold 0.5% BSA. An aliquot (50 μl) of stationary phase nonradiolabeled S. aureus culture was added and the samples were centrifuged in a microfuge to pellet the bacteria. Nonradioactive S. aureus were added to increase the size of the subsequent bacterial pellet. Reactions with pure PLA2 were centrifuged at 6000 rpm for 2 minutes and reactions with rabbit tears were centrifuged at 6000 rpm for 6 minutes. An aliquot of supernatant (200 μl) was used to quantify, by liquid scintillation counting, the products of hydrolysis. When inhibitors were tested, they were added to the reaction mixtures just before incubation at 33°C. 

Thin-Layer Chromatography of Radioactivity Assay Samples

The supernatants of reactions of PLA2 or tears with ¹⁴C-oleic acid labeled Staphylococcus were pooled. The pooled samples (2.75 ml) were filtered through a 0.22-μm syringe filter (Fisher Scientific, Pittsburgh, PA) to remove any remaining bacteria. Aliquots of the filtrate (200 μl) were spotted in 20-μl volumes onto thin-layer chromatography (TLC) plates (PE Sil G, Whatman; Fisher). Pure arachidonic acid (30 mM, 50 μl; Sigma) was spotted onto the plate as a chromatography marker. The TLC plates were chromatographed using a solvent containing hexane:diethyl ether:glacial acetic acid (80:20:1.8; v/v/v). ³⁰ The lane containing pure arachidonic acid was placed into a chamber containing iodine vapors and allowed to develop for 5 to 10 minutes to determine the R_f of pure arachidonic acid. ³¹ The other lanes on the plate (20 × 3.5 cm) were separated, and each was cut into 10 equal sections (2 cm) for scintillation counting. 

Statistical Analysis

The SEM of Staphylococcus per cornea or contact lens as well as the SEM for CFU or counts per minute (cpm) per assay reaction were determined using a Statistical Analysis Systems program. ³² Statistical analysis was performed using a one-way nested analysis of variance on each group. Protected t-tests were then determined between least square means derived from each variance analysis on each group. P ≤ 0.05 was considered significant. 

Results

In Vivo Bacterial Killing

The fate of Staphylococcus adhering to contact lenses after inoculation onto scarified rabbit corneas was analyzed with time (Fig. 1) . There was, within the first few hours, a marked decline in the number of bacterial CFUs per contact lens (from 7.0 to 3.0 log CFU). The loss of bacteria from the contact lenses was essentially complete within 24 hours. The number of bacteria associated with the cornea was low at 24 hours (2.6 ± 0.92 log CFU) and even lower at 48 hours (1.03 ± 0.38 log CFU). 

Sterile contact lenses induced trace inflammation at 8 hours (SLE = 3.6 ± 0.39); however, this inflammation dissipated by 24 hours (SLE = 0). Contact lenses with adherent Staphylococcus caused greater inflammation than sterile lenses at both 8 hours (SLE = 5.25 ± 0.52) and 24 hours (SLE = 9.50 ± 0.71). SLE scores for rabbits with contact lenses containing bacteria steadily decreased until there were only mild corneal changes in inoculated eyes by 48 hours (SLE = 6.41 ± 0.62). This decline in SLE scores continued, and inoculated eyes appeared normal by 72 hours. 

In Vitro Killing of Bacteria by Rabbit Tears

To determine whether a host defense against Staphylococcus was associated with the tear film, the survival of Staphylococcus in rabbit tears was analyzed. Bacteria in TSB (7.0 log CFU, initial concentration) were mixed (1:1; v/v) with tears or saline and incubated at 33° or 37°C for 4 hours. Samples of bacteria in tears incubated at 33°C demonstrated a 4 log reduction in CFUs compared with bacteria incubated in saline at 33°C (3.99 ± 0.27 vs. 8.34 ± 0.20 log CFU, respectively; P ≤ 0.001). The bacteria incubated with saline at 37°C for 4 hours contained a similar quantity of bacteria as those incubated with tears at 37°C (7.82 ± 0.12 vs. 6.84 ± 0.32 log CFU, respectively; P = 0.0994). 

To determine the rate of bacterial killing by tears at 33°C, a mixture of tears and bacterial culture, as well as a mixture of saline and bacterial culture, was incubated at 33°C and assayed periodically (Fig. 2) . The results show that the bacterial killing was minimal at 0.5 hours, but was clearly evident by 1 hour and became extensive by 4 hours compared with the saline control at each time point. 

Bacterial Killing by PLA2 or by Rabbit Tears

To further explore the bactericidal action of rabbit tears and to determine the role of PLA2 in this activity, the survival of Staphylococcus was analyzed quantitatively by mixing 100 to 200 CFU of log phase bacteria in PBS with increasing quantities of PLA2 for 4 hours at 33°C. Bacteria incubated with a small quantity of PLA2 (6 U) showed an increase in CFU relative to the control incubated in PBS (Fig. 3A ). However, PLA2 quantities of 18 to 24 U caused a concentration-dependent decline in the number of surviving bacteria. At PLA2 quantities of 30 U or higher essentially all bacteria were killed. 

The killing of bacteria by rabbit tears was quantitatively analogous to that of PLA2. Incubation of bacteria for 4 hours at 33°C in tear volumes of 1 to 20 μl caused the number of bacteria to increase above the control bacteria incubated in PBS (Fig. 3B) . However, for bacteria incubated in 25 μl of tears, there was a marked decline in the number of surviving bacteria. Bacteria incubated in tear volumes of 30 μl or higher were almost all killed. 

Release of Radioactive Arachidonic Acid from S. aureus by PLA2 or Tears

The activity of pure PLA2 or rabbit tears in cleaving arachidonic acid from bacterial membranes was tested using ¹⁴C-labeled S. aureus. Bacteria incubated without PLA2 or rabbit tears did not show significant bacterial membrane cleavage as measured by cpm released from the bacteria. However, S. aureus incubated with pure PLA2 (≥0.2 U) or with rabbit tears (≥2 μl) for 30 minutes at 33°C demonstrated significant cleavage of bacterial membranes (Fig. 4) . 

To determine whether membrane cleavage by PLA2 or rabbit tears released arachidonic acid, TLC was used to separate the reaction components. Reactions containing PLA2 (0.8 U) or rabbit tears (10 μl) both demonstrated significant amounts of radioactivity with an R_f identical with arachidonic acid (Fig. 5) . Bacteria incubated without PLA2 or tears contained low quantities of radioactive products and no product with the R_f of arachidonic acid. 

Effect of Inhibitors of PLA2 Activity on Release of Arachidonic Acid

Cleavage of bacterial membranes by PLA2 has been reported to be inhibited by spermidine, a polyamine, ²¹ and tetracaine, a topical anesthetic. ²² These two inhibitors were tested using radiolabeled S. aureus to determine their ability to inhibit the cleavage of ¹⁴C-labeled bacterial membranes by PLA2 or rabbit tears. Spermidine (50 mM) or tetracaine (50 mM) was added to the bacterial reaction with PLA2 or tears before incubation. The addition of spermidine or tetracaine markedly reduced the release of ¹⁴C-arachidonic acid in the reactions mediated by either PLA2 or rabbit tears (Fig. 6) . Neither spermidine nor tetracaine incubated with bacteria (without enzyme or tears) caused significant bacterial membrane cleavage as shown by the low cpm released in these reactions. 

Inhibition of Bacterial Killing by PLA2 or Rabbit Tears

The ability of tetracaine or spermidine to protect bacteria from PLA2-mediated killing in rabbit tears was investigated. Tetracaine, a potent inhibitor of PLA2 activity, was found to be toxic to S. aureus at those concentrations inhibitory for PLA2 activity (≥2 mM). However, spermidine, an inhibitor of PLA2 activity with less potency than tetracaine, at a concentration of 1 mM was minimally toxic for S. aureus, and partial inhibition of the PLA2-mediated release of arachidonic acid was noted. S. aureus incubated with a nonlethal quantity of tears (10 μl) and with spermidine (1 mM) were found to grow to approximately twice the CFU numbers obtained in tears without spermidine (Table 1) . Bacteria incubated with a lethal quantity of tears (60 μl) were killed unless spermidine (4 mM) was included. In the presence of spermidine (4 mM) the bacteria in tears grew to quantities exceeding that of the controls free of tears. 

Discussion

The findings of this study are the first to indicate that PLA2 is an important host defense molecule of the rabbit tear film and to demonstrate that this enzyme in rabbit tears kills S. aureus. The action of PLA2 in the tear film explains the observed death of S. aureus after bacterial inoculation topically onto the scarified rabbit cornea. Although the antibacterial action of tears has been traditionally associated with lysozyme, complement, lactoferrin, and IgA antibody, ^{1 2 4 5 6} the data presented herein and that of another study ¹³ demonstrate that PLA2 is a very potent bactericidal factor of the tear film. Although the present study demonstrates the susceptibility of S. aureus to PLA2, other ocular pathogens are reported to be susceptible to digestion by PLA2 from leukocytes. ³³ Qu and Lehrer ¹³ have shown that PLA2 in human tears can kill a variety of Gram-positive pathogens. Thus, PLA2 activity in rabbit tears probably protects against multiple types of bacteria pathogens that contact the tear film. 

The defensive role of PLA2 in rabbit tears was overcome by the addition of inhibitors of the PLA2 reaction. These inhibitors significantly reduced the release of arachidonic acid from bacteria incubated in tears and promoted the survival and growth of bacteria in potentially lethal quantities of tears (60 μl). The data do not exclude the possibility that, in addition to PLA2 digestion of bacterial membranes, other components of tears contribute to bacterial killing. However, the data show that PLA2 must be active for the bactericidal reaction to occur. 

The results of this study help explain the difference in bacterial survival and growth after inoculation by intrastromal injection as opposed to topical inoculation. Injection of relatively small numbers of S. aureus (100 CFU) into the rabbit corneal stroma (remote from the tear film) results in keratitis characterized by rapid increases in the number of CFU per cornea and by a steady increase in inflammation and tissue damage. ¹⁸ However, Matoba et al. ²⁰ showed that topical inoculation of S. aureus onto scarified rabbit corneas failed to result in keratitis. The present study as well as previous studies from this laboratory ²⁶ and from Rhem et al. ³⁴ demonstrates that the application of S. aureus onto the rabbit cornea can lead to inflammation. Inflammation noted by this laboratory included chemosis, injection, and accumulation of pus on the corneal surface. Such inflammation could be stimulated at least in part by the release of arachidonic acid during PLA2 digestion of the bacterial inoculum. The release of arachidonic acid could also stimulate inflammation during other ocular infections associated with Staphylococcus such as conjunctivitis or blepharitis. Although inflammation was apparent in previous studies, ^{20 26 34} an increase in bacterial CFUs in the cornea after topical inoculation has not been previously described. In fact, numerous methods to mechanically compromise the cornea before topical inoculation have failed to yield a topical model of Staphylococcus keratitis with extensive bacterial replication. ²⁶ The methods tested include enhanced scarification, vertical incisions of various depths into the stroma using a diamond knife, and removal of the epithelium. Pretreatment of the cornea with hydrolytic enzymes and the testing of numerous bacterial strains and growth conditions have also failed to yield an infection in which extensive and rapid bacterial replication followed topical inoculation. The successful infection of the rabbit cornea after inoculation by intrastromal injection and the repeated failures of topical Staphylococcus inoculations each appear appropriate now that the bactericidal potency of PLA2 in the rabbit tear film has been determined. 

The killing of Staphylococcus in rabbit tears appeared greater at 33°C than at 37°C. This observation, however, represents the sum of two reactions: the killing of bacteria by PLA2 in tears and the growth of bacteria in a rich medium (TSB) at 37°C. Release of radioactive arachidonic acid from bacterial membranes was as extensive at 37°C as at 33°C, indicating that PLA2 is fully active at 37°C (unpublished finding). Most experiments in the present study were conducted at 33°C because the lower temperature (33°C vs. 37°C) is essentially that of the normal rabbit corneal surface. ²⁸  

The effect of tears on Staphylococcus survival was clearly concentration dependent. Low volumes of tears mixed with bacteria in buffer resulted in bacterial growth, indicating that tears contain a relatively high concentration of nutrients suitable for bacterial replication. However, at higher tear concentrations the bactericidal properties of tears offset the nutritive effects, and the bacteria were killed. These results are significant because they suggest that even a partial loss of PLA2 activity in the tear film could leave the corneal surface covered with a nutritive layer instead of a bactericidal layer. A loss in the bactericidal activity of PLA2 in tears could be an unrecognized hallmark of some diseases that predispose patients to bacterial infection (e.g., Sjögren’s syndrome). If so, the topical application of exogenous PLA2 could represent a prophylactic therapy for such patient populations. Saari et al. ³⁵ have suggested that PLA2 of human tears declines with age. This decline of PLA2 could increase the possibility of ocular infection among the elderly. 

Two findings of potential significance were that tetracaine inhibits the PLA2 reaction responsible for the killing of bacteria in tears and that tetracaine is itself toxic to S. aureus. Tetracaine is only one topical anesthetic commonly used in ophthalmology that has reported inhibitory activity for PLA2. ²² We have found that proparacaine is another topical ocular medication that is inhibitory for PLA2 digestion of bacterial membranes (unpublished finding). These findings imply that the use of an ocular anesthetic inhibitory for PLA2 activity on bacteria could compromise this major host defense of the tear film and precondition the eye to infection with bacteria resistant to the anesthetic. The bactericidal activity of these anesthetics could be an under-recognized yet important factor in avoiding nosocomial ocular infections. Studies are needed to determine the ability of various drugs to inhibit the protective PLA2 reaction, the susceptibility of various ocular bacterial pathogens to the PLA2 activity, and the susceptibility of these ocular pathogens to ocular anesthetics. 

This study has demonstrated that bacterial killing in the tear film of rabbits is mediated by PLA2 and that this defense mechanism is apparently of sufficient potency to protect the scarified cornea against topical inoculation of Staphylococcus. Although the data presently available do not exclude the involvement of other host components in the bacterial killing reactions, inhibition of PLA2 activity by spermidine was sufficient to protect bacteria from the lethal effects of tears. Impairment of PLA2 activity by use of drugs inhibitory to this protective enzyme could be a predisposing factor for ocular infections, especially those after surgery or other invasive techniques. Further studies of this protective reaction as a broad-based ocular host defense system are needed to understand the interaction of the tear film with invading bacteria. 

²  Present affiliation: Cooperative Research Centre for Eye Research and Technology, University of New South Wales, Sydney, New South Wales, Australia.

 Supported by National Eye Institute Grant RO1-EY10974.

 Submitted for publication March 19, 2001; revised June 1, 2001; accepted June 7, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked“ advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Colony-forming units (CFU) per contact lens in rabbit eyes after application of contact lenses containing Staphylococcus aureus. Contact lenses with adherent Staphylococcus were applied to rabbit eyes and a tarsorrhaphy was performed. At various times, lenses were removed, homogenized, and cultured to determine the log number of bacterial CFU per lens. Error bars, SEM.

Figure 2.

 

Effects of Staphylococcus (in TSB) incubation in saline or rabbit tears at 33°C. Staphylococcus grown to log phase in TSB was mixed with an equal volume of tears or sterile saline and incubated at 33°C. Aliquots were removed periodically and cultured to determine the log number of CFU surviving the incubation. Error bars, SEM. ▪, log CFU of bacteria incubated with rabbit tears;□ , log CFU of bacteria incubated with saline.

Figure 3.

 

Effects of Staphylococcus incubation with phospholipase A₂ (PLA2) or rabbit tears at 33°C. Staphylococcus grown to log phase in M9 medium was centrifuged to a pellet, resuspended in PBS, mixed with rabbit tears or PLA2, and incubated at 33°C for 4 hours. An aliquot of each was removed and cultured to determine the log number of CFU surviving incubation. Error bars, SEM. (A) The CFU present after incubation with various quantities of PLA2; (B) the CFU present after incubation with various volumes of rabbit tears.

Figure 4.

 

Phospholipase A₂ (PLA2) or rabbit tear cleavage of ¹⁴C-labeled Staphylococcus aureus. Staphylococcus labeled with ¹⁴C-oleic acid was incubated with PLA2 or rabbit tears at 33°C for 30 minutes. Reactions were terminated with the addition of BSA and then centrifuged to pellet bacteria. An aliquot of each supernatant was removed and quantified by liquid scintillation counting. Error bars, SEM. (A) The cpm released when ¹⁴C-labeled Staphylococcus was incubated with various quantities of PLA2; (B) the cpm released when ¹⁴C-labeled Staphylococcus was incubated with various volumes of rabbit tears.

Figure 5.

 

TLC of radioactivity assay reactions. Supernatants of reactions of PLA2 or tears with ¹⁴C-labeled Staphylococcus were spotted onto TLC plates. Arachidonic acid was spotted on the plate as a chromatography marker. The TLC plates were chromatographed using a hexane:diethyl ether:glacial acetic acid (80:20:1.8; v/v/v) solvent. The lane containing arachidonic acid was developed in iodine vapors to determine the R_f of pure arachidonic acid. The lanes containing supernatant reactions were cut into 10 equal sections for scintillation counting. Bacteria incubated with buffer (no tears or PLA2) showed only low quantities of radioactivity and none with the R_f of arachidonic acid when chromatographed (not shown). Error bars, SEM. (A) The cpm released when supernatant from bacteria incubated with PLA2 (0.8 U) or rabbit tears (10 μl) was chromatographed; (B) the R_f of pure arachidonic acid.

Figure 6.

 

Inhibitors of phospholipase A₂ (PLA2) prevent cleavage of ¹⁴C-labeled Staphylococcus aureus by phospholipase A₂ or rabbit tears. An inhibitor of PLA2 activity, spermidine or tetracaine, was incubated with Staphylococcus labeled with ¹⁴C-oleic acid and PLA2 or rabbit tears at 33°C for 30 minutes. Reactions were terminated with the addition of BSA and then centrifuged to pellet bacteria. An aliquot of each supernatant was removed and quantified by liquid scintillation counting. Error bars, SEM. (A) The cpm released when ¹⁴C-labeled Staphylococcus was incubated with PLA2 (0.8 U) and inhibitors (50 mM, each); (B) the cpm released when ¹⁴C-labeled Staphylococcus was incubated with rabbit tears (10 μl) and inhibitors (50 mM, each).

Table 1.

 

Protective Effects of Spermidine for Staphylococcus in Rabbit Tears

Group	Tears	Spermidine	CFU*
1	10 μl (nonlethal)	—	760.0 ± 44.06, †
2	10 μl (nonlethal)	1 mM	1,426.0 ± 126.97, †
3	60 μl (lethal)	—	0.71 ± 0.71, ‡
4	60 μl (lethal)	4 mM	206.5 ± 60.29, ‡
5	—	—	127.5 ± 23.78, ‡

*  Log phase bacteria (100–200 CFU per assay) in M9 medium were added to each sample. Samples (≥3 per group) were incubated at 33°C for 4 hours, and aliquots were cultured on TSA agar plates to determine the number of CFU.

^†  The number of bacteria in group 1 was significantly lower than the number of bacteria in group 2 (P = 0.0026).

^‡  The number of bacteria in group 3 is significantly lower than the number of bacteria in groups 4 and 5 (P ≤ 0.0085).

The authors thank Hinh Keith Nyugen, Kiana Nelson, and Quentin Booker for their technical assistance with this manuscript.