Factors influencing bacterial adhesion to contact lenses

Bacterial characteristics-- Gram negative bacteria are the predominant causative agents in contact lens-related MK, with Pseudomonas species being the most commonly isolated organism [12,14,19]. Serratia marcescens, coagulase-negative staphylococci and Staphylococcus aureus are often the next most commonly identified causative organisms [19-21]. The range of organisms associated with contact lens MK may show regional variation [12], with Gram negative bacteria being more common in tropical climates. Here we will review the characteristics of P. aeruginosa, S. aureus and coagulase-negative Staphylococci that affect their adhesion to contact lenses. In general, the bacterial attachment process to any surface can be divided into two stages. The first stage is one of temporary adhesion, in which bacteria can break away from the surface. This stage is largely mediated by London-Van der Waals forces [22]. In the second stage, defined by Marshall et al. [22] as one of irreversible adhesion, a time-dependent firm union occurs in which bacteria no longer exhibit Brownian motion and cannot be removed by washing. After these stages of initial adhesion, the adherent bacteria can then progress to form a biofilm which further contributes to the anchoring of the bacteria to the surface [23].

P. aeruginosa is a ubiquitous environmental Gram negative bacterium, with a complex genetic makeup enabling its survival in a wide variety of nutritional environments. These characteristics contribute to the mechanisms by which it adheres to contact lenses, although these are not yet fully understood. While cell surface appendages termed pili and flagella can participate in the adhesion processes of P. aeruginosa [24,25], non-piliated Pseudomonas can also adhere to contact lenses [26]. The cell surface hydrophobicity of P. aeruginosa also contributes to its adhesion to contact lenses [27]. The prominent adhesive nature of P. aeruginosa is the result of its unique physio-chemical characteristics. The strength of bacterial attachment is often influenced by their surface hydrophobicity. Organisms with greater surface hydrophobicity adhere in greater numbers than hydrophilic organisms. This phenomenon could explain the greater adhesive nature of P. aeruginosa than Staphylococcus. P. aeruginosa GSU#3 is reported to be highly hydrophobic with a surface water contact angle of 132° compared to that of various S.aureus strains which ranges from 20° to 36° [28].

Adhesion of P. aeruginosa varies considerably between strains [27,29-34]. P. aeruginosa strains isolated from human corneas during keratitis adhere to soft contact lenses in significantly greater numbers than isolates from other body parts. Even then there are differences between clinical isolates in their ability to adhere to surfaces [31,35]. However, P. aeruginosa strains can be classified as invasive and cytotoxic based on the presence of several transcriptional genes, and while one report showed no direct correlation between the cytotoxic or invasive properties and degree of their adhesion to hydrogel and silicon hydrogel lenses [29], other reports have shown that cytotoxic strains (those carrying the Exotoxin U gene) are more frequently resistant to hydrogel contact lens disinfections [36] and generally strong biofilm producers (on a polystyrene surface) [37].

Since the last decade, there have been several reports of the growing incidence of keratitis caused by microorganisms other than P. aeruginosa such as coagulase negative Staphylococci [38,39]. Some studies reported that Staphylococcus epidermidis accounts for as many as 45% of all cases bacterial keratitis [40,41]. Similar to Pseudomonas, adhesion of this organism is also reported to be strain specific [42-45]. S. epidermidis can adhere to a variety of different surfaces using a polysaccharide adhesin (PS/A) contained within an exopolysaccharide (slime) capsule [45]. Variation of expression of PS/A affects the degree of S. epidermidis adhesion to biomaterials [43,45,46]. These differences may also be related to the absence of the intracellular adhesin operon in the biofilm negative strains [47]. This operon is involved in the production of the polysaccharide intercellular adhesin (PIA) that is functionally necessary for cell-to-cell adhesion and biofilm formation [48]. The expression of exopolysaccharide and adhesins differs for each S. epidermidis strain leading to differences in their adhesion [49].

Staphylococcus aureus is another microorganism commonly isolated from MK and inflammatory events including contact lens peripheral ulcer (CLPU) [3]. Thakur et al. [50] determined that different strains of S. aureus adhered to a hydroxyethylmethacrylate (HEMA)-based contact lens in different numbers. However, there have been no publications on the bacterial factors that are associated with adhesion of S. aureus to lenses.

Overall, P. aeruginosa usually shows significantly greater adhesion to unworn silicone hydrogel or hydrogel lenses compared to Staphylococcus species [28,29,34,42-44,51] and these findings are summarized in Table 1 and Table 2. Similar findings are reported for other bacteria such as Streptococcus pneumoniae [44], Hemophilus influenzae [44], Micrococcus luteus [52], and Serratia marcescens [29] in comparison to P. aeruginosa. Results from Bandara et al. [44] (Table 1) show an order of magnitude higher adhesion ratio between P. aeruginosa and S. epidermidis than most other studies. This is most likely a result of the strains tested; in this case P. aeruginosa 6294, which shows greater level of adhesion than other Pseudomonas strains and S. epidermidis 5 which is less adhesive compared to other strains (unpublished data). Interestingly, results reported by Kodjikian et al. [42] consistently showed higher adhesion ratios between P. aeruginosa and S. epidermidis compared to results from Henriques et al. [43] to various lens materials. This is possibly due to the nutritionally limiting phosphate buffered saline (PBS) used by Kodjikian et al. [42] compared to nutritionally rich artificial tears used by Henriques et al. [43] In nutritionally poor media S. epidermidis can show reduced cell viability especially if incubated for longer periods of time. Similarly, Borazjani et al. [29] (Table 2), using PBS as media for both bacteria, showed a very high comparative adhesion ratio between P. aeruginosa and S. aureus. The comparable adhesion between these two species in data from Zang et al. [52], Willcox et al. [53], and Vermeltfoort et al. [54] (Table 2) under similar assay conditions appears to be mainly strain driven, but still showing higher adhesion by P.aeruginosa.

In most of the studies PBS has been used as bacterial suspension media [29,31,32,42,51,55-57]. This buffer solution helps to maintain a constant pH and osmolarity. Highest bacterial adhesion with PBS has been observed at pH 7 [42]. Some studies have used instead sterile saline [52,58] but as both PBS and saline are nutritionally inert and thus longer incubation with PBS and saline may underestimate total cell numbers (especially for the more fastidious microbes such as Staphylococci which may die upon prolonged exposure). Some studies have used a complex and nutritionally rich media such as trypticase soya broth as bacterial suspension media, while other studies have used dilutions of this media in PBS [44,59]. In an attempt to more closely replicate adhesion that may occur during lens wear, several other studies have used artificial tears as the bacterial suspension fluid [43,60]. Any differences seen between adhesion levels between studies using these different media might be related to affects of electrolyte concentration and ionic charge of suspending media, as well as the nutritional fastidiousness of the bacteria. Furthermore, nitrogen or carbon limitation which can alter the ability of P. aeruginosa to adhere to Etafilcon A lenses [61].

Interestingly, George et al. [51] observed that the presence of S. epidermidis on hydrogel lens surfaces significantly reduces the adhesion of P. aeruginosa, but the presence of P. aeruginosa does not largely alter the adhesion of S. epidermidis [51]. This phenomenon is yet to be understood though it further supports the general premise that the normal ocular microbiota (of which coagulase-negative staphylococci such as S. epidermidis are predominant members) may be protective to eye. The strength of adhesion to contact lenses is also strain dependant. A rinsing step after bacteria adhesion can remove roughly 90% more of S. epidermidis compared to P. aeruginosa and passage through a air/liquid interface can detach more S. aureus than P. aeruginosa [42,51]. In contrast, a recent investigation of adhesion forces between bacteria and contact lenses, lens storage cases or the cornea (measured using atomic force microscopy) suggested that staphylococci and Serratia liquifaciens adhered significantly stronger than P. aeruginosa, and this might result in slightly higher transmission rates of P. aeruginosa to the cornea [62].

Using five different P. aeruginosa strains, Williams et al. [31] sought to determine whether an increasing bacterial concentration (1×10⁷ to 1×10⁹) results in greater viable bacterial counts on contact lenses. Maximum adhesion was seen when 1×10⁹ CFU/ml bacteria were added to lenses, except in the case of P. aeruginosa 6294, which reached maximum adhesion to worn lenses at 1×10⁸ CFU/ml. Interestingly George et al. [51] found that re-exposure of P. aeruginosa to fresh lenses after the bacterial inoculum had been previously allowed to adhere to a contact lens resulted in lower adhesion than to the initial lens [51]. This phenomenon suggests a limited number of bacterial cells in a standard inoculum are responsible for adhesion, suggesting a certain phenotype of bacterial cells within a population (approximately 10% of the cells) is responsible for most of the adhesion seen. This could be due to the use of cells grown in static culture to stationary phase, where cells of differing phenotype could be present. Indeed, Williams et al. [31], found that P. aeruginosa strains grown to stationary phase adhered in higher levels that those grown to exponential phase.

The time required for irreversible bacterial attachment and biofilm formation on the lens surface is a crucial factor that can differ between bacterial types. The adhesion of S. epidermidis on hydrogel lenses tends to be slowly incremental over 2 h while P. aeruginosa adhesion is as quick as 5 min [51]. Miller et al. [35] reported adhesion of P. aeruginosa nonmucoid isolate number 3 increased with time, peaking after 3 h and then remaining constant. Duran et al. [56] reported a steady increase in adhesion of an MK isolate of P. aeruginosa to Polymacon and Lidofilcon A lenses from 2 min to 1 h. Subsequent studies by Stapleton et al. [59] supported these data and showed rapid attachment up to 10⁷ cells per lens, and adhesion reached maximum after 45 min. Glycocalyx formation (i.e., biofilm formation) occurred after 30 min incubation with a bacterial inoculum of 10⁷ organisms per ml [59]. Andrews et al. [63], using an ATP based bioluminescent assay and image analysis, reported that the adhesion of P. aeruginosa, S. epidermidis, and S. marcescens was maximal at 4–6 h but this was followed by a metabolic decline after 18 h [63]. The decrease in metabolic activity is characteristic of a biofilm mode of growth [64]. However, others have shown that viable cell numbers on several lens materials (hydrogel and silicone hydrogel) significantly increased up to 16 to 24 h after incubation [33,60]. Randler et al. [60] noticed a decrease in viable bacterial numbers on silicone hydrogel lenses exposed to an artificial tear fluid within a few hours. This observed decrease may be due to the antimicrobial components such as lysozyme in the artificial tear fluid.

Overall the main bacterial factors that influence adhesion are their cell surface hydrophobicity, use of different strains, and the suspension media. Most studies show that the adhesion of P. aeruginosa is higher than other bacterial types to most lens types, perhaps being one reason for the predominance of this bacterium in microbial keratitis. Adhesion of P. aeruginosa to lenses is rapid, usually occurring within 1 h, and biofilm formation can occur with 24 h of initial adhesion.

Contact lens material characteristics-- Since the introduction of soft contact lenses to the contact lens market by Bausch and Lomb in 1971, exhaustive research has been conducted to investigate and improve soft lens materials. Susceptibility of this and all biomaterials to bacterial colonization has been a major concern of manufacturers, researchers and practitioners. Low oxygen transmissibility of hydrogel lenses was overcome by the introduction of silicone hydrogel lenses in 1999. These lenses have successfully reduced various clinical signs of corneal hypoxia. It was anticipated that contact lens related infiltrative events would also be reduced by these lenses but this is not evident [65]. Bacterial adhesion to lens surfaces is heavily influenced by lens material characteristics. The following are the distinguishing lens material characteristics and their effect on bacterial adhesion.

Several authors have observed greater levels of adhesion of various strains of P. aeruginosa to lenses composed of non-ionic polymers compared to those with ionic polymers [35,59]. Initial adhesion of S. aureus was higher to an ionic hydrogel compared to a non-ionic hydrogel [66]. The Federal Drug Administration (FDA) classifies soft contact lens materials as being of high water content (more than 50%) or low water content (less than 50%). Results of research into bacterial adhesion based on water content have been remarkably consistent with multiple studies concluding that bacterial adhesion increases inversely to the water content [30,34,35,42,46,67]. Garcia-Saenz et al. [46] showed increased adhesion to low water content hydrogel lenses of two strains of S. epidermidis. Kodjikian et al. [42], Miller et al. [30,35], and Cook et al. [67,68] found the same effect for strains of P. aeruginosa, although Miller et al. [35] reported no strict correlation between water content and adherence. However, two studies could not establish any relation between water content and adhesion [69,70] and there appears to be no relationship between the numbers of bacteria isolated from lenses after wear and water content [70]. Silicone hydrogel lenses are all of relatively low water content and so water content is likely to be a minor factor in the bacterial adhesion to this lens type. Polymer composition and surface hydrophobicity can mask the effects of water content, reducing the influence of this characteristic on the degree of bacterial adhesion [35].

Hydrophobicity is a crucial contact lens surface property. Common adhesive patterns of most bacterial isolates to various substrata indicate that hydrophobic surfaces attract greater numbers of bacteria than hydrophilic surfaces [58,71-74]. Bacteria attach with higher affinity to low-energy, hydrophobic surfaces than to high energy, hydrophilic surfaces [71]. Hydrophobicity of silicone hydrogels such as Lotrafilcon A, Balafilcon A, or Lotrafilcon B has been reported to be higher compared to hydrogel lenses such as Etafilcon A. In vitro P.aeruginosa, S.aureus, or S. epidermidis adhere in greater numbers to the hydrophobic silicone hydrogel lenses compared to hydrophilic hydrogel lenses [28,43,51,74]. Also, Lotrafilcon A and Balafilcon A are often reported as the most hydrophobic silicone hydrogels lenses and more bacteria adhere to these compared other silicone hydrogels [43]. It is likely that hydrophobic bacteria prefer adhering to hydrophobic lenses whereas hydrophilic bacteria adhere well to hydrophilic lenses [54]. However, Santos et al. [75] found the same amount of S. epidermidis adhesion to unworn silicone hydrogel and hydrogel lenses.

Detailed information regarding contact lens topography and roughness can be determined by atomic force microscopy. Giraldez et al. [58] reported that Comfilcon A and Omafilcon A lens materials had relatively smooth surfaces compared to Senofilcon A, Nelfilcon A, and Ocufilcon B. S. epidermidis showed stronger adhesion to the materials with higher surface roughness [58]. Similarly, with S. epidermidis, when adhesion to surfaces with similar hydrophobicity but differing surface roughness was examined, a higher surface roughness resulted in an increase in bacterial adhesion [76]. There are no reports investigating a link between adhesion of P. aeruginosa and lens surface roughness.

Contact lens wear can have significant effects on the surface properties of lenses, due to the deposition of tear film components (and possibly components of multipurpose disinfection solutions) during wear. After wear silicone hydrogel lenses show reduced surface hydrophobicity [54,75,77]. Worn hydrogel and silicone hydrogel lenses usually exhibit higher degrees of roughness than their unworn counterparts [75,78,79]. Therefore, these changes in surface characteristics of lenses during/after wear may influence bacterial adhesion. Worn Galyfilcon A and Lotrafilcon A adhere more S. epidermidis than unworn lenses [75]. The presence of sorbed protein can increase adhesion of S. epidermidis by 45% [67]. P. aeruginosa colonize the surface of worn extended wear contact lenses in direct proportion to lens surface deposits, preferentially adhering on areas of lens deposits [80]. In vitro Pseudomonas adhesion is highly correlated with the number of large (more than 150 µm) focal deposits on the lens after wear [81]. In contrast, though enzymatic cleaning is recommended for protein deposition, it does not appear to significantly reduce the adhesion of P. aeruginosa to lens surfaces [80].

Borazjani et al. [29] and Boles et al. [55] reported that 1 week extended wear of Balafilcon A and Etafilcon A had no major effect on the adhesion of P. aeruginosa. However, continuous wear of lotrafilcon A lenses reduced adhesion of the hydrophilic S. aureus strain 835, whereas continuous wear of balafilcon A lens significantly increased adhesion of this same strain [54]. Adhesion of the hydrophobic P. aeruginosa strain #3 to lenses after continuous wear was generally less than to unworn lenses, regardless of the type of lens [54]. Interestingly, P. aeruginosa 6294 adhered to a greater extent to the unworn Etafilcon A lenses than to 30 nights continuously worn lenses [82]. In vitro adhesion of P. aeruginosa significantly varies when lenses are worn by different individuals [30]. Butrus et al. [83] demonstrated that P. aeruginosa adhere in greater numbers to worn extended wear soft contact lenses compared to unworn lenses. In another study, worn Balafilcon A lenses have been shown to increase bacterial adhesion [84]. These differences in the effect of lens wear on adhesion may be related to bacterial strain differences, how the adhesion was measured (examining live bacteria (CFU/lens) or total bacterial cells) or to the difference between individuals [29].

It may be that it is the presence of specific adsorbed tear products on worn contact lenses that affect bacterial adhesion. Stern et al. [85] and Miller et al. [35] reported that adsorbed mucin, IgA, BSA, lysozyme, and lactoferrin enhanced the adhesion of P. aeruginosa GSU#3 to contact lenses. Adhesion of P. aeruginosa strain RT-1, a corneal ulcer isolate, was increased when albumin was adsorbed to a lens surface [86]. However, Williams et al. [31,32] and Lakshman et al. [87] showed that the presence of lactoferrin increased the total numbers of P. aeruginosa adhering to lenses but reduced their viability, killing the attached bacteria. In vitro, the presence of lysozyme on a lens surface has a variable impact on adhesion of P. aeruginosa [31,50] and S. aureus [52] but markedly reduces the viability of Micrococcus luteus [52]. The number of viable cells of S. aureus that adhere to contact lenses is reduced if those lenses are coated with secretory phospholipase A2 [87]. The ability of the tear film lipids cholesterol or phospholipids to modulate bacterial adhesion has been measured [88]. The lipids once adsorbed to lenses appear to have no effect on bacterial adhesion.

Overall, it is the lens surface hydrophobicity and roughness, as well as polymer characteristics such as water content and iconicity (at least for HEMA-based lenses) that appear to modulate bacterial adhesion. The effect of lens wear is not constant and may be affected by the individual who has worn a lens, and the types of proteins deposited on the lens surface.

Concluding remarks-- It is important to note that reported adhesion rates are sensitive to study methodology and every study has its unique features. Despite this, it is still instructive to attempt some comparisons. Most of the ocular pathogens showed in vitro variation in adhesion between species and strains. P. aeruginosa can adhere to contact lenses the most of any bacteria test thus far, and this may be a reason it is the most predominant microorganism that causes contact lens-associated MK.