1. Introduction
Microbial adhesion is the initial step in colonization and the formation of a biofilm—an accumulated biomass of microorganisms and extracellular materials on a solid surface.
Biofilms can be detrimental to both human life and industrial processes, causing infection associated with medical implants [1], pathogen interaction with host cells [2,3], periodontitis or dental caries [4,5], contamination of food from processing equipment [6,7], enhancement of metal corrosion [8], formation of marine biofilms on ships’ hulls [9], and so on.
However, microbial adhesion can be also beneficial, for example, in the degradation of environmental hazardous chemicals in soil[10,11] or in bioreactors for wastewater treatment [12] or off-gas treatment [13], in agricultural uses of root nodule bacteria in the rhizosphere [14], in the degradation of biopolymers such as cellulose [15], and in bioflocculants used for the separation of coal particles [16].
Therefore, there are two opposite goals for the control of microbial adhesion and biofilm formation: one is the prevention and inhibition of biofilms and the other is their enhancement and promotion. Development of materials and technologies for surface modification and treatment is desired to control microbial adhesion and biofilm formation. For this purpose, elucidation of the mechanisms underlying microbial adhesion is necessary.
In this paper, the theories of bacterial adhesion are outlined on the basis of two physicochemical approaches, the DerjaguinLandau-Verwey-Overbeek (DLVO) theory and the thermodynamic approach, and then the complexity of the adhesion process of actual bacteria is described. Next, bacterial cell surface structures are presented for better understanding of the molecular mechanism of bacterial adhesion since these play important roles in the adhesion process at the nanometer scale. Finally, state-of-the-art technologies for controlling microbial adhesion and/or biofilm formation are presented.
2. Theory of bacterial adhesion
2.1. Conventional model based on the DLVO theory
Among microorganisms, bacteria are major constituents of biofilms and are about 0.5–2 m in size, that is, nearly in the range of colloidal particles. Therefore, bacterial adhesion has been described by the DLVO theory [17], which originally described the interaction of a colloidal particle with a surface.
According to this theory, the total interaction between a surface and a particle is the summation of their van der Waals and Coulomb interactions [18].
Since the van der Waals attractive force is dominant in the vicinity of a surface, particles cannot separate from the surface by Brownianmotion, and therefore adhere irreversibly.
In contrast, the Coulomb interaction becomes dominant at a distance away from the surface because the van der Waals force decreases sharply with distance.
In the presence of a charged particle in an aqueous solution, counter ions against the surface charge are attracted by the particle, forming an electric double layer.
As bacteria and natural surfaces in aqueous solution are usually negatively charged [19], repulsive electrostatic energy is caused by overlap of the electrical double layers of bacterial cells and the substratum [17,19–22].
This repulsive energy increases as the ionic strength of an aqueous solution decreases because shielding of the surface charges by the ions in the electrical double layers lessens. At low ionic strengths, when a bacterial cell approaches a surface, there is an energy barrier which bacterial cells cannot surmount by swimming or Brownian motion (Fig. 1) [5,17,23,24].
Fig. 1. Total interaction energy between a bacterial cell and a surface depending on ionic strength.
In these conditions, there is a shallow secondary energy minimum outside of the energy barrier. The distance from the surface to the secondary energy minimum is usually within several nanometers, depending on the ionic strength. In the first step of cell adhesion, a bacterial cell comes to this position by its motility or Brownian motion, and adheres to the surface reversibly (Fig. 2A).
Fig. 2. Schematics of bacterial adhesion process. (A) Usual two-step adhesion process. (B) One-step adhesion model mediated by long nanofibers as seen in Acinetobacter sp. Tol 5.
In the following step, the bacterial cell uses nanofibers, such as pili and flagella, or produces exopolymeric substances (EPS), which can pierce the energy barrier due to their small radii, for bridging between the cell and surface. When the energy barrier becomes higher and farther from the substratum at lower ionic strengths, however, it becomes difficult for the nanofibers and EPS to reach the substratum, and bacterial cells become unable to adhere.
On the contrary, at high ionic strengths, the energy barrier disappears and bacterial cells easily and rapidly attain irreversible adhesion. Many researchers have found a link between decreasing bacterial adhesion and decreasing ionic strength, which is consistent with the DLVO theory [19,25–32].
2.2. Thermodynamic approach and extended DLVO theory
The thermodynamic approach is based on the surface free energies of the interacting surfaces [33,34], which are calculated using the following equation:
where γ(sm), γ(sl), and γ(ml) are the solid–microorganism, solid–liquid, and microorganism–liquid interfacial free energies, respectively. Adhesion is favored if the free energy is negative as a result of adhesion. However, it should be noted that thermodynamics principally assumes that the process is reversible, which is often not the case [35]. Based on the DLVO theory, bacterial adhesion has been described as a two-phase process including an initial, instantaneous and reversible physical phase (phase one) and a time-dependent and irreversible molecular and cellular phase (phase two) [36]. However, the thermodynamic approach cannot be applied to the adhesion in phase one at the secondary energy minimum, where a new cell–substratum interface is not formed [35,37]. Thus, the thermodynamic approach, in which the formation of a new cell–substratum interface at the expense of the substratum–medium and cell–medium interfaces is calculated, contrasts with the DLVO theory, in which the interaction energy is distance dependent. However, the thermodynamic approach helps to explain a common observation: bacteria with a hydrophobic cell surface prefer hydrophobic material surfaces; those with a hydrophilic cell surface prefer hydrophilic surfaces [36].
ffects bacterial adhesion and the self-agglutination of bacterial cells [36–42]. Hydrophobic interaction between two apolar moieties immersed in water is the sole consequence of the hydrogen-bonding energy of cohesion of the water molecules surrounding these moieties [43]. Hydrogen-bonding can be viewed as a form of more general electron-donor/electron-accepter interactions, namely, Lewis acid–base interactions [44]. According to van Oss, the surface tension( γ ) consists of the Lifshitz-van der Waals component γ(LW) and the Lewis acid–base component γ(AB) [45].
γ (LW) comprises the London dispersion force, the Keesom dipole–dipole force, and the Debye dipole-induced dipole force. van Oss developed the extended DLVO theory, in which the hydrophobic/hydrophilic interactions and osmotic interaction also are included [46,47]. Since the osmotic interaction is negligibly small in bacterial adhesion, the total adhesion energy can be expressed as:
where △GvdW is the Lifshitz-van der Waals interaction, △Gdl is the electric double layer interaction, and △GAB relates to acid–base interactions. The latter component introduces a component that, in principle, describes attractive hydrophobic interactions and repulsive hydration effects. The influence of the acid–base interactions is enormous compared with electrostatic and Lifshitz-van der Waals interactions. In some cases, the extended DLVO theory seems to qualitatively predict experimental adhesion results better than the classical DLVO theory; the extended DLVO theory predicts very strong interaction due to acid–base interactions leading to an extremely deep minimum without an energy barrier, whereas adhesion is not expected to occur according to the classical DLVO theory [20,35,48]. However, the distance dependence of acid–base interactions is also relatively short-ranged. Acid–base interactions and the electric double layer interaction decay exponentially from the value at close contact [47], and calculations have shown that a distance between the interacting surfaces of less than 5 nm is required before acid–base interactions can become operative [20].
2.3. Estimation of surface potential of bacterial cells
To estimate the height of the energy barrier, ζ -potentials of bacterial cells have been calculated from the electrophoretic mobility (EPM) by using the Smoluchowski equation. However, it has been found that in many bacterial strains, the EPM approaches a nonzero value with increasing ionic strength [49,50]. This suggests that the ζ -potential is inappropriate for accurately measuring the surface potential of bacterial cells. Ohshima et al.[51–55] have pointed out that the Smoluchowski equation can be applied to only rigid particles having no polymers, and have developed a model that describes the EPM of particles having a soft polymer surface layer in which diffuse double layer ions and watermolecules can freely penetrate. Morisaki et al. [49,50] used the Ohshima model to estimate accurate surface potentials of actual bacterial cells. In the Ohshima model, EPM ( μ ) is described by the following formula and the EPM change approaching a non-zero value along with ionic strength can be explained:
where εr is the relative permittivity of the medium in which the cells suspended, ε0 is the permittivity of a vacuum, is the viscosity of the medium, Ψ 0 is the surface potential of the particles, Ψ DON is the Donnan potential of the polymer layer, κ m is the Debye-Hückel parameter of the polymer layer, z is the valence of the charged groups in the polymers, e is the electron charge, N is the density of the charged groups, and γ is the softness parameter, which has dimensions of reciprocal length. It is reasonable that bacterial cells are recognized as soft particles that are covered with or encapsulated by polymers, for example, polysaccharides and proteins. By applying the Ohshima model to bacterial cells, it has been revealed that the actual surface potential of bacterial cells is much smaller than the ζ -potential calculated from the Smoluchowski equation and that the energy barrier disappears or is sufficiently low for bacterial cells to surmount it even at low ionic strengths, where, by applying the ζ -potential to the DLVO theory, a high energy barrier had been thought to prevent direct interaction of a cell and a surface [49]. However, the high energy barrier preventing the cells from irreversibly adhering to a surface does seem to be present under very low ionic strength conditions even for a ‘soft’ particle [49].
2.4. Complexity of actual bacterial adhesion
Actual bacterial adhesion is an extremely complicated process and frequently deviates from the adhesion models described above. Usually, solid materials in various environments do not expose bare surfaces, and various organic and inorganic matter adsorbs to the surfaces before microorganisms adhere, forming layers called conditioning films [2,20,56,57].
The physicochemical properties of conditioning films are quite different from the original bare surfaces, and interactions of microorganisms with the conditioning films also differ accordingly [58]. This should be taken into consideration when materials that prevent biofilm formation are developed. Unlike simple colloidal particles, a bacterial surface is structurally and chemically heterogeneous. For example, Gram-negative bacteria have an outer membrane (OM) consisting of a lipid bilayer containing lipopolysaccharide (LPS) at the outer layer of the OM with significant variations in the coverage density and local distribution [59–61].
The contribution of LPS to cell adhesion will be described later. Various kinds of OM proteins are heterogeneously embedded in the OM and many of them protrude out of cells, forming into cell appendages. Pili and flagella are typical cell appendages, have lengths from hundreds of nanometers to several micrometers, forming long fibrous structures with diameters of several nanometers to tens of nanometers.
These bacterial nanofibers not only pierce the energy barrier described by the DLVO theory and/or tether a cell body to surfaces but also cause deviation of cell adhesion behavior from that predicted by the DLVO theory. In addition, although different structures on the same cell individually contribute to the net cell character, they are often described in terms of their combined contribution to the overall cellular properties. A hydrophobic part localized at a bacterial nanofiber against the hydrophilic main cell surface specifically orients the cell at the interface [38].
Hori et al. [13] reported the highly adhesive bacterium Acinetobacter sp. Tol 5, which was isolated as a toluene-degrading bacterium from a bioreactor for treatment of off-gas containing volatile organic carbons. The adhesiveness of this toluene degrader is noteworthy. The inner walls of plastic tips and pipettes become coated with cells immediately just by cell sampling. Although two morphologically different nanofibers, namely, anchor-like and peritrichate pilus-like nanofibers, were initially found [62], at least three types of peritrichate nanofibers as well as the anchor on Tol 5 cells have been identified using state-of-the-art electron microcopy techniques. The less adhesive mutant T1 of this bacterium is defective in these nanofibers [63]. This mutant exhibits decreased adhesion with decreasing ionic strength and does not adhere at all at 0.015 mM, whereas adhesion of the wild type of Tol 5 is fully independent of ionic strength. Even the wild-type strain exhibits ionic strength-dependent adhesion when the cells are grown under culture condition, in which those nanofibers are scarcely produced [64]. These observations imply that adhesion of Tol 5 cells deviates from the DLVO theory due to these nanofibers. In addition, the wild-type cells expressing these nanofibers attain irreversible adhesion immediately under conditions in which adhesion of T1 cells is still reversible [63]. Tol 5 cells are considered to be able to attain irreversible adhesion rapidly without approaching the vicinity of the substratum through the long distant interaction mediated by the nanofibers (Fig. 2B). On the other hand, the mutant T1 has also lost the self-agglutinating property of the wild type, resulting in monolayer adsorption to hydrocarbon surfaces [65]. Thus, bacterial nanofibers contribute not only to cell adhesion to solid surfaces but also to cell agglutination.