Adhesion is crucial for the infective lifestyles of bacterial pathogens. Adhesion to non-living surfaces, other microbial cells, and components of the biofilm extracellular matrix are crucial for biofilm formation and integrity, plus adherence to host factors constitutes a first step leading to an infection. Adhesion is, therefore, at the core of pathogens’ ability to contaminate, transmit, establish residency within a host, and cause an infection. Several mycobacterial species cause diseases in humans and animals with diverse clinical manifestations. Mycobacterium tuberculosis, which enters through the respiratory tract, first adheres to alveolar macrophages and epithelial cells leading up to transmigration across the alveolar epithelium and containment within granulomas. Later, when dissemination occurs, the bacilli need to adhere to extracellular matrix components to infect extrapulmonary sites. Mycobacteria causing zoonotic infections and emerging nontuberculous mycobacterial pathogens follow divergent routes of infection that probably require adapted adhesion mechanisms. New evidence also points to the occurrence of mycobacterial biofilms during infection, emphasizing a need to better understand the adhesive factors required for their formation.
1. Introduction
Adhesion is central to microbial proliferation
[1]. It drives formation of biofilms, in which individual microbial cells have greater access to nutrients and protection from environmental stresses
[2][3]; it is a prerequisite for the colonization of environmental niches
[4], and, for pathogens, adhesion to host tissues and cells constitutes one of the first steps in the establishment of an infection
[5][6].
The mechanism of bacterial adhesion to a substrate may involve non-specific macroscopic surface properties, such as surface free energy, charge, or hydrophobicity
[7], or, as is often the case in pathogens, specialized surface-localized molecules, called adhesins, may act as effectors of adhesion through interactions with specific host molecules
[5]. These adhesins often participate in more elaborate processes than the mere act of binding to a specific substrate or ligand. For example, the adhesin function of type IV pili (T4P) is critical for their role in motility
[8]. In staphylococcal septicemia, a set of adhesins employing variations of the “dock, lock, and latch” binding mechanism form exceptionally stable, stress-enhanced bonds, allowing the bacteria to remain adhered to blood vessel walls under high flow rates
[9]. The adhesion mechanisms expressed and employed by a particular bacterial species, therefore, appear to be tailored to particular mechanobiological as well as broader physiological needs
[10].
Although Mycobacterium tuberculosis, the best studied mycobacterial pathogen, expresses a repertoire of adhesins, most of these do not appear to function like the bona fide adhesin virulence factors seen in other pathogenic bacteria that offer a means of adherence under high mechanical stresses. Central to the pathogenicity of mycobacterial pathogens, such as M. tuberculosis and Mycobacterium leprae, is their ability to invade and proliferate inside host cells, so the emphasis is on targeted host cell entry. However, this may not necessarily be the case for all mycobacteria, in particular an emerging class of nontuberculous mycobacteria (NTM).
2. Where and When Is Adhesion Important in Mycobacterial Pathogenesis?
2.1. Tuberculosis and Leprosy
An overview of mycobacterial adhesion in the clinical context is given in
Figure 1. The lifecycle of
M. tuberculosis starts with the inhalation of small aerosol droplets that were propelled into the air through the cough of a person with active tuberculosis (Tb). Due to their small size, some of these droplets pass the upper respiratory tract and carry tubercle bacilli straight to the alveolar spaces of the lung. These bacteria, thus, get to bypass the competition of the commensal microbial flora and primed microbicidal immunity of the upper respiratory tract
[11]. In the alveoli, they are phagocytosed by their preferred host macrophage cells, a process that initiates a complex inflammatory cascade that drives formation of a multicellular structure, called a granuloma, wherein bacilli are contained in a latent infectious phase
[12]. Up until here, the adhesion factors used by the tubercle bacilli would seem to mainly consist of surface-exposed lipids and glycoconjugates that bind to a range of receptors expressed on the macrophage surface
[13][14]. However, it was discovered that
Mycobacterium bovis BCG requires the heparin-binding haemagglutinin adhesin (HBHA), which it shares with
M. tuberculosis and which binds to heparin sulfate-containing receptors on the surface of epithelial cells, for extrapulmonary dissemination in a mouse infection model
[15]. This finding suggested that the adhesin may be used by tubercle bacilli to cross the alveolar epithelium, although this is also likely accomplished via diapedesis in the transmigration of infected alveolar macrophages
[16]. In certain individuals, especially young children and those with a suppressed immune system, intense intracellular bacillary multiplication during the initial phase of infection or lesion of granulomas, upon reactivation of a long latent infection, leads to lymphatic or hematogenous metastasis
[17]. During this metastatic spread to distant sites in the body (nervous system, bones, genitourinary system, skin), the tubercle bacilli probably require adherence to host extracellular matrix components to overcome colonization-hindering mechanical shear forces
[18].
Figure 1. Where and when mycobacterial adhesion occurs during infection. Pulmonary infections involve initial adhesion to alveolar macrophages (AM) or epithelial cells (AEC). After transmigration across the alveolar epithelium bacilli adhere to monocytic cells or dendritic cells (DC). The immune response can contain bacteria in granulomas, where it was very recently demonstrated that M. tuberculosis biofilms occur. Upon disintegration of granulomas, bacteria may adhere to extracellular matrix (ECM) components. During hematogenous dissemination, tubercle bacilli may bind plasma fibronectin and, ultimately, to ECM proteins to invade new tissues. Some nontuberculous mycobacteria (NTM) adhere to gut epithelium to cause gastrointestinal infections. NTM that cause nosocomial infections adhere to fomites or surgical equipment. Mycobacteria also self-adhere to form cords, which is a form of immune evasion employed by some NTM. EndoC, endothelial cells.
Although
M. leprae is an obligate intracellular pathogen affecting mostly peripheral zones of the body, one of its primary routes of infection is the nose
[19], and its ortholog of HBHA has also been implicated in its ability to attach to airway epithelial cells
[20]. In addition, the ability of this mycobacterial pathogen to invade the peripheral nervous system has been attributed, at least in part, to the interaction between a yet unidentified adhesin and the G domain of the laminin-α2 chain (LN-α2G)
[21]. This would facilitate attachment of
M. leprae to Schwann cells via a ternary interaction, where LN-α2G forms a bridge between the bacterial adhesin and β4 integrin on the Schwann cell.
2.2. Zoonotic and Opportunistic Infections and Emerging Mycobacterial Pathogens
M. tuberculosis and
M. leprae have historically been the most important mycobacterial pathogens, and their intracellular lifestyle may be seen as a paradigm of mycobacterial pathogenicity. Yet, several less important mycobacteria, causing mainly animal disease, and emerging pathogenic nontuberculous mycobacteria (NTM) appear to follow infective lifestyles that deviate from that of tuberculous and leprous bacilli, and, wherein, the mechanisms of adhesion are probably different
[22][23].
Mycobacterium bovis, the causative agent of bovine tuberculosis, which can also cause zoonoses, can, in addition to an infection resembling pulmonary tuberculosis, cause gastrointestinal infections in humans after consumption of contaminated unpasteurized dairy products
[24]. In the NTM
Mycobacterium avium subsp. paratuberculosis, which causes gastrointestinal infections in ruminants known as Johne’s disease and which has been theorized to be associated with human inflammatory bowel diseases such as Crohn’s disease
[25], a fibronectin-binding adhesin has been implicated in its ability to bind and penetrate intestinal mucosal epithelium
[26][27]. This was found to involve high-density integrin-displaying and, therefore, fibronectin-binding M cells
[28]. The role of adhesion in NTM physiopathology, often associated with environmental sources of contamination and distinct extracellular phases, is underexplored, and should be given more attention. These include the skin pathogens
Mycobacterium ulcerans and
Mycobacterium marinum, although for the former the role played by adhesins during cutaneous infection is disputed
[29], as well as for a number of NTM that are associated with surgical procedure–related infections, including
Mycobacterium chelonae,
Mycobacterium fortuitum, and members of the
Mycobacterium abscessus complex (MAC)
[30]. The nosocomial nature of these infections involves adhesion in the contamination of fomites, including surgical equipment, and subsequent transmission onto host tissues
[31][32][33]. It is worth noting that MAC also causes human transmissible pulmonary infections for which cystic fibrosis sufferers show a heightened vulnerability
[22][34][35].
2.3. Adhesive Interactions in Mycobacterial Biofilms
The capacity of various environmental mycobacteria (including species that can cause human infections) to form robust biofilms in sources such as domestic water distributions systems
[36][37][38] and medical equipment
[39][40] has been known for several decades. Evidence also exists of the occurrence of mycobacterial biofilms in vivo during the course of infection for some NTM species
[41][42][43][44][45]. However, the direct involvement of
M. tuberculosis biofilms during the course of tuberculosis was largely considered to be non-existent until very recently. Reports of such biofilms in animal models of infection, including nonhuman primates, as well as in histological lung sections from human tuberculosis patients, implicate the formation of these structures in pathogenesis
[46][47]. These biofilms were further demonstrated to contribute in the resistance of resident tubercle bacilli to both antitubercular treatments and the human immune response
[47]. Understanding how these mycobacterial biofilms are formed in vivo may pave the way to the discovery of new and better therapeutic strategies to treat mycobacterial infections. We point out that adhesion plays at least three key roles in biofilm formation: In the first instance, adhesion of planktonic bacteria to a substratum serves as the point of nucleation and, thus, as the very first step in biofilm development. Secondly, intercellular adhesion is very likely required during early growth of the biofilm. Third, the biofilm extracellular matrix, which holds individual cells together and which constitutes the major biomass component in mature biofilms, relies on adhesive interactions to provide mechanical stability to the mature biofilm
[48].
In routine axenic broth cultures, mycobacteria generally grow as pellicle biofilms, types of biofilms that form at the air-liquid interface and that are distinct from classical biofilms, which are surface-attached. Most of what is known for
M. tuberculosis biofilms is based on data for the pellicular form (
Figure 2). Principal components of the mycobacterial pellicle biofilm extracellular matrix are extracellular DNA
[49][50][51] and free mycolic acids
[52][53]. In addition, a range of lipids including short chain mycolic acids, monomeromycolyl diacylglycerol, mycolate ester wax, glycopeptidolipids (GPLs), phthiocerol dimycoserosates (PDIM), phenolic glycolipids (PGL), and keto mycolic acids have been directly implicated in the ability of mycobacteria to form pellicle biofilms
[53][54][55][56][57][58]. Although some of these lipids were also reported to be constituents of the biofilm extracellular matrix, it has been proposed that their major participation in biofilm formation is more likely to be in direct interbacterial adhesion and involves their hydrophobicity (at least in the case of the less amphiphilic ones)
[59]. While exopolysaccharides are abundant in the extracellular matrices of many microbes
[60][61][62][63][64], evidence of these complex carbohydrates in mycobacterial biofilms have been lacking. Very recently, a rapidly inducible surface-attached in vitro biofilm model was devised for
M. tuberculosis [47][65]. Interestingly, it was discovered that the principal components of these biofilms were exopolysaccharides (
Figure 2), among which cellulose was identified
[65]. The presence of cellulose in the biofilm extracellular matrix was also reported for
M. avium,
Mycobacterium fortuitum, and
M. smegmatis [65][66]. The detection of mannose and galactose in the composition of
M. ulcerans [43] and
M. smegmatis [67] biofilm extracellular matrices, respectively, in addition to glucose, may hint towards the presence of yet unidentified exopolysaccharides. Importantly, cellulose was also detected in mycobacterial biofilms occurring during infection
[65], pointing to a role for biofilm extracellular matrix exopolysaccharides in pathophysiology. It needs to be addressed how the exopolysaccharide components of mycobacterial biofilms contribute to structural integrity of the latter. Another important question to be addressed is what are the mycobacterial surface components that bind to these exopolysaccharides. A number of recent reports of bacterial lectins binding biofilm extracellular matrix exopolysaccharides
[68][69][70][71][72][73][74] intimate that such lectins may also be present in mycobacteria.
Figure 2. The composition of the biofilm extracellular matrix is different in pellicle biofilms and surface-attached biofilms. While the former is rich in free mycolic acids and monomeromycolyl diacylglycerol (MMDAG), the latter is lipid-poor and contains large amounts of exopolysaccharides. Both types contain extracellular DNA (eDNA) and proteins.
3. Non-Specific Adhesion: The Hydrophobic Mycobacterial Surface
Recent reports have highlighted the role of surface hydrophobicity in mycobacterial pathogenicity, more specifically how
M. tuberculosis’s evolution from a non-pathogenic environmental ancestor to an obligate and highly successful human pathogen positively correlates with surface hydrophobicity
[75][76][77][78]. According to the model given by the authors of these studies, the evolutionary loss of relatively hydrophilic lipids such as lipooligosaccharides (LOSs) and acquisition of highly hydrophobic lipids such as PDIMs, pentaacyl trehaloses (PATs), and sulfoglycolipids (SGLs) in modern
M. tuberculosis strains improved their aerosolization and, hence, their transmission
[75][77].
A clue for another way in which a very hydrophobic cell surface may contribute to mycobacterial virulence is brought by the rapidly growing NTM
M. abscessus, for which concern is mounting over the human transmissible infections that it is causing, which are particularly difficult to manage in cystic fibrosis sufferers
[34][35][79][80].
M. abscessus normally produces a class of polar lipids, called GPLs, that accounts for a comparatively hydrophilic surface. In mutants lacking a component of the GPL biosynthetic and transport machinery, the cell surface is as hydrophobic as that of
M. tuberculosis, most likely because of the surface exposure of hydrophobic lipids such as trehalose mycolates and trehalose polyphleates
[81][82][83]. These lipids have been implicated in the ability of mycobacteria to form cords, which is strongly associated with mycobacterial virulence
[84][85][86]. For
M. abscessus, the transition from being GPL
+ to being GPL
− has been reported to occur during the course of infection and is associated with more severe disease
[87][88]. Studies using zebrafish embryos as an infection model have attributed the increased virulence of GPL
− M. abscessus strains to their ability to form cords and the inability of macrophages to engulf these cords
[89][90]. A biophysical explanation for cording lies in the dehydrating capacity of the hydrophobic cell surfaces that removes the vicinal water film
[91]. This results in a greatly increased density of direct contacts between closely apposed cell surfaces and, hence, strong interbacterial adhesion. The adhesive effect of hydrophobic surfaces even counts for contacts with relatively hydrophilic surfaces and the role of bacterial surface hydrophobicity in adhesion to host tissues as well as in phagocytic ingestion has been known for a long time
[92][93]. We point out, based on our own laboratory experience, that even abundant mycobacteria-producing polar lipids such as LOSs and GPLs are, nevertheless, considerably more hydrophobic than other Gram-positive and Negative species, such as
Staphylococcus aureus,
Pseudomonas aeruginosa, and
Escherichia coli. Even non-mycobacterial mycolic-acid-containing species, such as
Corynebacterium and
Gordona, exhibit considerable surface hydrophobicity
[94]; in all fairness, hydrophobic mycobacteria, such as
M. tuberculosis and GPL
− M. abscessus, are to be considered extremely hydrophobic.
Hydrophobicity, therefore, plays a critical role in inter-mycobacterial adhesion, which in turn is a requirement for cord formation, a form of immune evasion in some pathogenic species. The hydrophobic cell envelope also probably contributes to the contacts that are instated between mycobacterium and macrophage during phagocytic engulfment by likely facilitating specific interactions between cell receptors and their cognate bacterial-surface ligands. Examples of mycobacterial factors that underlie adhesion to specific host factors are given in Table 1.
Table 1. Major known mycobacterial adhesive molecules and their host factor targets.
Molecule Class (Examples) |
Host Factor(s) |
Key References |
Lipids/glycoconjugates: |
|
|
mannose-capped lipoarabinomannan (ManLAM) |
Pattern recognition receptors/C-type lectins (Mannose receptor, DC-SIGN, Dectin-2) |
[95][96][97][98] |
α-glucan |
DC-SIGN |
[99] |
Adhesins: |
|
|
heparin-binding haemagglutinin adhesin (HBHA) |
heparan sulfate |
[15][100][101][102] |
fibronectin attachment protein (Fap) |
fibronectin |
[103][104][105] |
antigen 85 (Ag85) complex |
fibronectin |
[106][107] |
Lectin adhesins: |
|
|
β-prism II fold lectin |
Unknown |
[108][109] |
13 kDa ricin-like lectin (sMTL-13) |
Unknown |
[110] |
Appendages: |
|
|
M. tuberculosis pilus (Mtp) |
lamanin |
[111] |