Bacterial Coaggregation and Cohesion Among Isolates From Contact Lens Cases

Abstract

Purpose: The aim of this study was to examine cohesion, coaggregation, and coculture between bacteria commonly isolated from contact lens cases.

Methods: Staphylococcus epidermidis, Staphylococcus haemolyticus, Micrococcus luteus, and Acinetobacter radioresistens (two strains each) isolated from contact lens cases of two asymptomatic wearers were used in this study. In the cohesion assay, bacteria were grown, washed, and examined by incubating lens cases with two different types of bacteria sequentially and assessing the number of adhered cells of each isolate. The ability of isolates to interfere with the growth of other isolates was tested by growing strains in cocultures for 24 hours and determining the numbers of cells of individual strains. For coaggregation, equal proportions of two bacterial suspensions were mixed and allowed to coaggregate for 24 hours. Inhibition of coaggregation was tested by the addition of lactose (0.06 M) or sucrose (0.06 M) or pronase.

Results: The initial adhesion of M. luteus or A. radioresistens significantly (P < 0.05) enhanced the subsequent adhesion of the staphylococci. The addition of A. radioresistens in liquid media significantly (P < 0.05) enhanced the growth of staphylococci. S. epidermidis or S. haemolyticus coaggregated with M. luteus or A. radioresistens. The degree of coaggregation varied between 30% and 54%. The highest coaggregation (54% ± 5%) was seen between A. radioresistens 22-1 and S. epidermidis 22-1, isolated from the same lens case. Only lactose or sucrose treatment of staphylococci could partly inhibit coaggregation of some pairs.

Conclusions: Coaggregation, cohesion, and growth promotion may facilitate the process of bacterial colonization of contact lens cases.

Approximately 140 million people worldwide wear contact lenses to correct refractive error, as a therapeutic device or as a cosmetic device.¹ In a daily wear modality, contact lenses need to be soaked in lens storage cases in solution for disinfection and cleaning when not being worn. However, often contact lens storage cases are contaminated² despite the use of disinfectants.^3–5 Lens case contamination occurs in up to 85% of compliant asymptomatic lens wearers.^5–10 Contact lens case contamination has been identified as a major risk factor associated with contact lens–induced ocular inflammation and infection.^10,11 The use of contaminated lens cases also has been associated with ocular itchiness, foreign body sensation, and redness.¹² Poor lens case hygiene has been associated with contact lens–induced microbial keratitis.^13,14 In lens wearers with microbial keratitis, the microorganisms recovered from lens cases were identical or similar to the causative organisms.^15,16 

While many types of microorganisms have been isolated from contact lens cases, the major contaminants are Gram-positive bacteria^6,7,9 such as coagulase-negative staphylococci^5,6,17 which are often cultured from the ocular surface of healthy individuals^18,19 and from the ocular surface during keratitis.^20,21 Other species such as Micrococcus spp. have also been isolated from lens cases of asymptomatic wearers^6,22 and are often found on skin, in soil, water, on fomites, in dust and air.^23,24 Contact lens cases can also be contaminated by Gram-negative bacteria, though the rate of isolation of these is rare in asymptomatic wearers.^7,8,25,26 Pseudomonas aeruginosa, Serratia marcescens, and Acinetobacter spp. have been cultured from contact lens cases of people with microbial keratitis.^{3,5,8,27–31} 

Microorganisms become more resistant to disinfectants upon attachment and biofilm formation on surfaces.^32,33 Biofilm formation may facilitate eventual transmission to contact lenses and eventually to ocular surface.³⁴ Bacterial biofilm formation is complex and involves adhesion, microcolony formation, maturation, and eventual dispersal of microbial cells. During this process, coaggregation and cohesion may occur and contribute to the formation and maturation of biofilms.^35–37 Coaggregation and cohesion may be processes that result in the production of multispecies biofilms.^38–41 

Ocular isolates of P. aeruginosa and S. aureus can coaggregate, but our previous study showed that coaggregation was not associated with cohesion between the strains.⁴² Further work is required to understand the bacterial interactions that promote the build-up of multiple species biofilms within lens cases. This study aimed to understand the role of cohesion, coaggregation and growth in the buildup of biofilm formation of Staphylococcus spp. and other commonly isolated case contaminants. 

Methods

Microbial Strains

Strains of Staphylococcus epidermidis, Staphylococcus haemolyticus, Micrococcus luteus, and Acinetobacter radioresistens (two strains each) isolated from contact lens cases of two asymptomatic wearers, enrolled in a previous clinical trial⁴³ were used in this study (Table 1). Actinomyces naeslundii ATCC 12104 and Streptococcus sanguinis CR2B, isolated from dental plaque were used as positive controls for coaggregation.⁴² 

Table 1

 

Microorganisms Used in This Study

 

Cohesion Assay

Bacterial cells were grown in brain heart infusion (BHI; Becton Dickinson, Macquarie Park, Australia) for 18 to 24 hours at 37°C followed by washing and adjusting the optical density (OD) in PBS (pH 7.4 NaCl 8 g 1⁻¹, KCl 0.2 g 1⁻¹, Na₂HPO₄ 1.15 g 1⁻¹, KH₂PO₄ 0.2 g 1⁻¹ pH 7.2) to an OD of 0.1 at 660 nm (1 × 10⁸ CFU/mL) using a spectrophotometer (FLUOstar Omega; BMG Labtech, Ortenberg, Germany). Contact lens cases (ReNu Multiplus; Bausch & Lomb, Rochester, NY, USA) were incubated at 37°C sequentially with two different types of bacteria, each for 24 hours followed by washing the lens cases once with PBS to remove the loosely attached cells. Subsequently, 2 mL of PBS was added to each well of the lens case along with a sterile magnetic stirring bar and the case was vortexed for 1 minute to dislodge the adherent bacterial cells. The recovery of bacterial species was assessed using a selective bacterial growth medium, as described previously.⁴² Controls used a single bacterial type incubated for 24 hours followed by the addition of sterile PBS and further incubation of 24 hours. 

The total number of adherent bacteria was measured after growth at 37°C for 24 hours on nutrient agar (NA) plates (Thermo Fisher Scientific, Thebarton, Australia). The growth of Staphylococcus and Micrococcus spp. was estimated by growth for 24 hours at 37°C on the selective medium mannitol salt agar (MSA; Thermo Fisher Scientific). The recovery of A. radioresistens was calculated by assessing the difference in the number of bacteria recovered from MSA and NA plates (A. radioresistens + Staphylococcus or Micrococcus spp.), as A. radioresistens does not grow on MSA plates. The differentiation of M. luteus and the staphylococci on MSA or NA plates was performed by evaluating the morphologic appearance of each recovered bacterial colony. M. luteus and staphylococci colonies were differentiated by the pattern of pigmentation: golden-brown and bright yellow, respectively. This assay was repeated in duplicate on three different occasions. 

The ability of bacteria adhering or cohering to the lens cases to produce biofilms was also evaluated. Extracellular DNA (eDNA) in biofilms was quantified using a previously described protocol⁴⁴ using 1 μM SYTOX Green (Invitrogen Australia Pty Limited, Sydney, Australia) at excitation/emission wavelengths of 504/523 nm. The amount of eDNA produced by strains grown individually in wells was measured by incubation in SYTOX Green for 2 to 4 minutes, as was the amount of eDNA produced when strains were grown in the same well sequentially, as per the protocol for cohesion. These experiments were repeated in duplicate on three different occasions. 

Bacterial Growth

For this assay only, the bacterial pairs that cohered were tested. In the first instance, bacterial cells were grown overnight in BHI and were resuspended to 0.1 OD₆₆₀ in PBS and spread on NA plate using a sterile cotton swab. Subsequently, the other bacteria of the pair was prepared as above and spotted (10 μL) on the bacterial lawn.⁴² The plates were then dried and incubated at 37°C for 24 hours. Inhibition of growth was seen as an inhibitory zone around the spotted strain. The zone of inhibition was graded on 1–4 scale, as described previously.⁴² 

For bacterial growth in nutrient broth, the method of Qin et al. was used,⁴⁵ with modifications. Overnight cultures in BHI of A. radioresistens, M. luteus or Staphylococcus spp. were centrifuged, washed with PBS, and diluted to OD 0.1 at 660 nm (1 × 10⁸ CFU/mL) in BHI. One mL of one bacterial suspension was added to 50 mL of the other partner bacterial suspension and the cocultures were incubated at 37°C with shaking at 250 rpm. Every 2 hours, the bacterial suspension was diluted 10-fold in PBS and these dilutions were plated onto NA and MSA plates and the number of bacteria was counted after overnight incubation at 37°C. The bacteria on agar plates were identified by evaluating the morphological appearance of each recovered bacterial colony. The experiment was repeated in duplicate on three different occasions. 

Coaggregation Assay

Bacterial coaggregation was performed as previously described.⁴² In brief, cells were grown in BHI for 18 to 24 hours at 37°C followed by washing and adjusting the OD to 1.0 (1 × 10⁹ CFU/mL) in coaggregation buffer (1 mM Tris [hydroxymethyl] amino methane, 0.1 mM CaCl₂, 0.1 mM MgCl₂, 0.15 M NaCl and 3.1 mM NaN₃ at pH 8.0).⁴⁶ An equal volume (0.2 mL) of two cell suspensions was mixed (e.g., A. radioresistens plus S. epidermidis), vortexed, and the OD was measured after 24 hours of static incubation at ambient temperature³⁹ and the percentage coaggregation was assessed by comparing to single suspensions of each bacteria.⁴² A percentage decrease in OD of greater than 30% indicated that coaggregation had occurred.³⁸ The experiments were performed in duplicate and repeated three times. 

Inhibition of Bacterial Coaggregation

Lactose (0.06 M) and sucrose (0.06 M) were used to inhibit the coaggregation reactions as described previously.^46,47 Additionally, the inhibition of bacterial coaggregation was assessed after incubating bacterial strains in protease (from Streptomyces griseus, P-5130; Sigma-Aldrich Corp., Castle Hill, Australia).⁴² The bacterial strains were incubated for 2 hours in lactose, sucrose, or pronase at ambient temperature and then the cell suspensions were washed three times with coaggregation buffer to remove unbound inhibitors. Then, the treated bacteria were added to their nontreated bacterial partners and the percentage of coaggregation was recorded after 24 hours of incubation at ambient temperature as described previously.⁴² A solution of 0.05% (wt/vol) Tween-20, 0.2 M NaCl was used as a negative control, to control for nonspecific bacterial interactions such as those associated with hydrophobicity and ionicity. Only if there was a greater inhibition than this control was inhibition of coaggregation considered to have occurred.^37,46 The experiments were performed in duplicate and repeated three times. 

Statistical Analysis

Data analysis was performed using spreadsheet and statistical software (Excel 2010; Microsoft Corp, Redmond, WA, USA; and SPSS Statistics, version 20.0; SPSS, Inc., Chicago, IL, USA). The recovery of bacteria after cohesion or coculture was compared a 2-tailed Student's t-test and using repeated measures ANOVA (for different time points). The percentage bacterial coaggregation between different pairs and the inhibition of bacterial coaggregation with the treatment of different inhibitory substances were compared using ANOVA with Bonferroni correction for post-hoc multiple comparisons. Statistical significance level was set at P < 0.05. 

Results

There was cohesion between the pairs of S. epidermidis or S. haemolyticus and M. luteus or A. radioresistens. Primary adhesion of M. luteus or A. radioresistens significantly (P = 0.05) enhanced the subsequent adhesion of staphylococci (Table 2). The greatest effect was seen with the primary adhesion of A. radioresistens 22-1, which increased the secondary colonization of S. epidermidis 22-1 by 1.8 ± 0.3 Log₁₀ colony forming units (CFU) compared to the adhesion of S. epidermidis 22-1 in isolation (P = 0.002; Table 2). Additionally, the primary adhesion of S. epidermidis 22-1 increased the secondary colonization of A. radioresistens 22-1 by 1.1 ± 0.2 Log₁₀ CFU, compared to adhesion of A. radioresistens alone (Table 2). The secondary colonization by S. epidermidis 22-1 significantly (P = 0.008) increased the primary adhesion of A. radioresistens 22-1 (P = 0.002) and M. luteus 22-1 (P < 0.005). The primary adhesion of M. luteus 22-1 or M. luteus 14-1 significantly increased the secondary adhesion of A. radioresistens 22-1 or 14-1, respectively (Table 2). The secondary colonization by S. haemolyticus 14-1 significantly (P = 0.005) increased the primary adhesion of M. luteus 14-1 and the secondary colonization S. epidermidis 22-1 increased the primary adhesion of M. luteus 22-1 (P = 0.05; Table 2). 

Table 2

 

Bacterial Cohesion on Contact Lens Cases Between the Strains of Staphylococcus spp., Micrococcus spp., and Acinetobacter spp., That Were Isolated From the Same Contact Lens Case

 

In the eDNA assay, there was evidence of biofilm formation with the largest amount of biofilm for bacteria incubated alone being produced by S. haemolyticus 14-1 (Table 3). When these strains were allowed to cohere, the amount of biofilm appears to closely mirror the amount produced by the strains adhered alone, and the amounts produced when strains were in combination did not differ from the amounts produced alone by more than 2 SDs. 

Table 3

 

Estimation of the Amount of Biofilm Formation (eDNA) Produced by Strains Alone or in Combination

 

The ability of bacteria to inhibit one another's growth was first investigated on nutrient agar, but no inhibition with any bacterial pairs was found. The effect of coculturing two different types of bacteria was genera dependent. Incubating staphylococci with A. radioresistens 22-1 or 14-1 increased the numbers of both staphylococci (P < 0.005; Fig. 1), and A. radioresistens 22-1 or 14-1 (P < 0.005; Fig. 2) that grew. Conversely, while incubating S. epidermidis 22-1 with M. luteus 22-1 did not significantly (P = 0.59; Fig. 1) increase the numbers of S. epidermidis, incubating S. haemolyticus 14-1 with M. luteus 14-1 reduced the final numbers and the growth rate of S. haemolyticus 14-1 (P < 0.005; Fig. 1). The coculture of M. luteus 22-1 with S. epidermidis 22-1 significantly increased the growth of M. luteus (P = 0.007; Fig. 3). The coculture of M. luteus 14-1 with S. haemolyticus 14-1 significantly altered the growth kinetics of M. luteus (P < 0.005; Fig. 3), but approximately the same number of cells of M. luteus were produced after 24 hours incubation. 

Figure 1

 

Effect of incubation with A. radioresistans or M. luteus on the growth of (A) S. epidermidis or S. (B) haemolyticus 14-1. 50 mL of S. epidermidis was incubated for 24 hours in BHI alone or in presence of other bacteria.

Figure 2

 

Effect of incubation with (A) S. epidermidis or (B) S. haemolyticus on the growth of A. radioresistans strains. We incubated 50 mL of A. radioresistans for 24 hours in BHI alone or in presence of S. epidermidis or S. haemolyticus.

Figure 3

 

Effect of incubation with (A) S. epidermidis or (B) S. haemolyticus 14-1 on the growth of M. luteus strains. We incubated 50 mL of M. luteus for 24 hours in BHI alone or in presence of S. epidermidis.

Coaggregation was observed between certain strains of S. epidermidis or S. haemolyticus with M. luteus or A. radioresistens (Table 4). The highest coaggregation occurred between A. radioresistens 22-1 and S. epidermidis 22-1 (54% ± 5%; Table 4). The highest coaggregation between two Gram positive bacteria was 50% ± 3% for the pair M. luteus 22-1 and S. epidermidis 22-1 (Table 4). The positive control of A. naeslundii ATCC 12104 and S. sanguinis CR2B showed 92% ± 3% coaggregation after 24 hours. 

Table 4

 

Coaggregation Between Gram-Positive and Gram-Negative Bacteria After Overnight Incubation

 

The results of the inhibition of coaggregation with lactose, sucrose or pronase are shown in Table 5. The negative control solution of Tween-20 and NaCl (that was used to determine any coaggregation as the result of hydrophobic or charge interactions) inhibited up to 38% of coaggregation (data not shown), therefore a significant inhibition in coaggregation for any treatment was considered to be ≥40%.³⁸ Lactose treated S. epidermidis 22-1 showed reduced coaggregation with M. luteus 22-1 or A. radioresistens 22-1 (45% ± 4% and 41% ± 3%, respectively). Lactose treatment of S. haemolyticus 14-1 caused 42% ± 4% reduction in coaggregation with A. radioresistens 22-1. Sucrose treatment of S. epidermidis 22-1 reduced coaggregation by 40% ± 3% with A. radioresistens 22-1 (Table 5) only. Incubation of any strain with pronase did not inhibit coaggregation. 

Table 5

 

Inhibition of Coaggregation After Treatment With Lactose (0.06M), Sucrose (0.06M), and Pronase

 

Discussion

The current study was designed to understand how multispecies biofilms might form in contact lens cases. Biofilms may be produced by a number of mechanisms and cohesion, coaggregation and stimulation of growth may be important aspects. The present study demonstrated for the first time that A. radioresistens, M. luteus, and Staphylococcus spp. could cohere. Additionally, the presence of A. radioresistens increased the growth of Staphylococcus spp. Coaggregation could occur between certain strains of A. radioresistens, M. luteus, S. epidermidis, and S. haemolyticus that had been isolated from lens cases of asymptomatic wearers. 

No coaggregating pairs presented the high coaggregation scores of the positive controls of A. naeslundii ATCC 12104 plus S. sanguinis CR2B (92% ± 3%). This outcome resembles a previous study finding where the coaggregation between P. aeruginosa and S. aureus reached only (62% ± 3%).⁴² In general, there was a trend for bacteria isolated from the same contact lens cases to coaggregate with each other, which was similar to the coaggregating pair of P. aeruginosa and S. aureus in the previous study.⁴² 

The current study also investigated the inhibitory effect of lactose, sucrose, and pronase on coaggregation. Sato et al.⁴⁸ demonstrated that coaggregation between actinomycetes and streptococci occurred via lectin-like substances (i.e., substances similar or identical to proteins that bind sugars) on the surface of actinomycetes with carbohydrate(s) on the surface of streptococci. Pretreating Staphylococcus spp. with lactose or sucrose inhibited the coaggregation, which indicates the involvement of staphylococcal lectins in coaggregation, consistent with previous studies.^42,49,50 However, the selected inhibitory sugars were unable to stop coaggregation between Micrococcus spp. and Acinetobacter spp. Understanding how bacteria coaggregate and cohere may help to produce strategies to halt biofilm formation. For example, adding inhibitory substances such as sugars to multipurpose disinfecting solutions may be of benefit. Although, it is unlikely that sugars such as lactose and sucrose can be used as they can be the source of nutrition of many types of bacteria. 

Cohesion between the strains of M. luteus or A. radioresistens with S. epidermidis or S. haemolyticus provides support to the concept that the initial adhesion of bacteria may control the secondary colonization of other types of bacteria (cohesion). M. luteus or A. radioresistens may form a conditioning film to enhance the adhesion of staphylococci or vice versa. However, the amount of biofilm formed by strains, as measured by estimating the amount of eDNA produced, was not affected by them adhering alone or in pairs. The ability of small numbers of A. radioresistens 22-1 or staphylococci to stimulate the growth of the other partner in coculture demonstrates that this can be another factor involved in the cohesion of bacterial cells and hence multispecies biofilm formation. Identifying the factors that are involved in this phenomenon may lead to determining mechanisms to reduce their production or interfere with their mechanism of action, which again could be used to reduce biofilm formation. 

In summary, it appears that a complex series of events may take place between bacteria that are involved in multispecies biofilm formation in contact lens cases. Certain bacteria can cohere—that is, the presence of one bacterial type increases the ability of another to attach. This can be facilitated by coaggregation, the direct adhesion between cells of different bacteria. Coaggregation may be involved in the cohesion between stains of A. radioresistens or M. luteus and staphylococci. The ability of small numbers of A. radioresistens or staphylococci to promote each other's growth may also be involved in cohesion and hence biofilm formation. This research will provide a framework for future studies that examine how to reduce biofilm formation in contact lens cases (e.g., adding substances to contact lens disinfecting solutions that can prevent aspects of cohesion or coaggregation). This may then reduce the contamination of contact lens cases during use. The research may also have application in other areas. For example, both Acinetobacter and Staphylococcus can be found concurrently in bronco-alveolar lavages from people with chronic obstructive pulmonary disease.⁵¹ 

Acknowledgments

The authors thank Nancy Briggs, senior statistical consultant from Mark Wainwright Analytical Centre, for assistance with statistical analysis. 

Disclosure: A. Datta, None; F. Stapleton, Alcon (F), Allergan (F), Coopervision (F), Novartis (C); M.D.P. Willcox, Alcon (F), Allergan (C, F), Cochlear (F), Coopervision (C, F, R), Johnson and Johnson Vision Care (F, R), Ophtecs (C, F, R, S), P