Effect of flagella expression on adhesion of Achromobacter piechaudii to chalk surfaces


  • A. Nejidat,

    1.  Department of Environmental Hydrology and Microbiology, Zukerberg Institute for Water Research, The Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus, Midreshet Ben-Gurion, Israel
    Search for more papers by this author
  • I. Saadi,

    1.  Agricultural Research Organization, Institute of Soil, Water and Environmental Sciences, Newe Ya’ar Research Center, Ramat Yishay, Israel
    Search for more papers by this author
  • Z. Ronen

    1.  Department of Environmental Hydrology and Microbiology, Zukerberg Institute for Water Research, The Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus, Midreshet Ben-Gurion, Israel
    Search for more papers by this author

Zeev Ronen, Department of Environmental Hydrology and Microbiology, Zukerberg Institute for Water Research, The Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus, 84990 Midreshet Ben-Gurion, Israel.
E-mail: zeevrone@bgu.ac.il


Aims:  To examine flagella role and cell motility in adhesion of Achromobacter piechaudii to chalk.

Methods and Results:  Transmission electron microscopy revealed that stationary cells have thicker and longer flagella than logarithmic cells. SDS-PAGE analysis showed that flagellin was more abundant in stationary cells than logarithmic ones. Sonication or inhibition of flagellin synthesis caused a 30% reduction in adhesion to chalk. Preincubation of chalk with flagella extracts reduced adhesion, by 50%. Three motility mutants were isolated. Mutants 94 and 153 were nonmotile, expressed normal levels of flagellin, have regular flagella and exhibited reduced adhesion. Mutant 208 expressed low levels of flagellin, no flagella and a spherical cell shape but with normal adhesion capacity.

Conclusions:  Multiple cell surface factors affect the adhesion efficiency to chalk. Flagella per se through physical interaction and through cell motility contribute to the adhesion process. The adhesion behaviour of mutant 208 suggests that cell shape can compensate for flagellar removal and motility.

Significance and Impact of the Study:  Physiological status affects bacterial cell surface properties and hence adhesion efficiency to chalk. This interaction is essential to sustain biodegradation activities and thus, remediation of contaminated chalk aquifers.


Natural bacterial communities are generally found as biofilm structures that are essential for microbial activity and survival. Many factors affect the adhesion of bacteria to solid surfaces and the development of a biofilm. Among these are the physicochemical properties of the matrix, bacterial cell surface structures and solution composition (Lawrence et al. 1995; Bos et al. 1999). As a part of our effort to design a practical scheme for the remediation of contaminated fractured chalk aquifers, we have been studying bacterial cell adhesion kinetics on a chalk matrix and the activity of the adhered cells (Nejidat et al. 2004).

Chalk is a unique active matrix as the dissolution of calcite (CaCO3), the major constituent of chalk, leads to continuous changes in the solid’s surface charge and nanostructure, and in the micro-environment at the interface (pH, calcium concentration). The zero point of charge for calcite is between pH 8 and 9·5. In this pH range, the surface is electrically neutral, whereas at higher or lower pH, the surface is negatively or positively charged, respectively (Somasundaran and Agar 1967). Previously, we demonstrated that the adhesion of Achromobacter piechaudii TBPZ-N61 to chalk follows a Langmuir isotherm and at saturation, cells in the stationary growth phase adhere more efficiently to chalk particles than cells in the logarithmic phase (Nejidat et al. 2004). In addition, stationary cells are much more hydrophobic than logarithmic ones, which might explain their advantage in the adhesion process (van Loosdrecht et al. 1990). Cells in the stationary phase were not noticeably active, either in suspension or when attached to chalk particles. Their activity was restored only after amendment with yeast extract. At saturation, it was calculated that the bacteria cover only 0·04% of the chalk surface, suggesting that bacteria adhere to specific sites on the chalk particles and that this limits the number of cells that can adhere (Nejidat et al. 2004). However, neither the nature of the chalk components that serve as the sites for bacterial adhesion nor the moiety of the bacterial structure which interacts with those sites are known. Nevertheless, as the efficacy of the adhesion process is dependent on physiological state, one could assume that structural changes at the surface of the stationary cells account for their higher adherence efficacy. In preliminary experiments, we examined both logarithmic and stationary cells under transmission electron microscopy (TEM) and the stationary cells were seen to have thicker and longer flagellar filaments. It is well known that flagella are responsible for motility in many bacteria, and that the flagellation pattern can be prone to environmental adaptation. Flagella are also known to play an important role in bacterial adhesion and pathogen virulence (Moens and Vanderleyden 1996). In this study, we examined the possible role of flagella and motility in the initial adhesion of A. piechaudii to the unique active chalk matrix.

Materials and methods

Cultures and medium

The bacterial strain A. piechaudii TBPZ-N61 was used throughout this study. This strain is capable of degrading 2,4,6-tribromophenol (TBP) and is resistant to the selective marker kanamycin. The isolation and characterization of this bacterial strain have been previously described (Ronen et al. 2000). TBPZ-N61 cells were cultured in a mineral medium containing as described by Nejidat et al. (2004). After autoclaving, the medium was supplemented with 100 mg l−1 TBP (Sigma) and 50 mg l−1 kanamycin. In some experiments, cells were grown in PTYG medium (Nejidat et al. 2004). Cells were collected by centrifugation at 5000 g for 20 min and washed twice in phosphate buffer (10 mmol l−1, pH 7·2) prior to adhesion studies.

Adhesion of TBPZ cells to crushed chalk

Adhesion experiments were conducted at 25°C in 20-ml glass vials. The incubation medium (10 ml) contained 10 mmol l−1 phosphate buffer (pH 7·2), chalk particles and washed bacterial cells grown on mineral medium at 25°C. The vials were placed on a rotary shaker (150 rev min−1) for 15 min followed by 30-min settling time. Adhesion was measured as difference in CFU (on PTYG agar amended with 50 mg l−1 kanamycin) between the initial and final suspensions (Mills et al. 1994). White and gray chalk samples were excavated from the surface and at a depth of 85 m, respectively, from a contaminated site in Israel (Wefer-Roehl et al. 2001). The different chalk types have been previously characterized and the specific surface areas of the chalk particles measured by the N2 BET method were 25·2 and 11·7 m2 g−1 for the white and the gray chalk, respectively (Wefer-Roehl et al. 2001).

Flagellin extraction

Cells were grown to logarithmic (16 h) or stationary (90 h) phase in mineral medium at 25°C, harvested by centrifugation, washed in 10 mmol l−1 sodium phosphate buffer (pH 7), and sonicated (Branson Digital Sonifier®) 10 times (5 s each, amplitude = 53% at 4°C). Flagella fraction was isolated as described by (Cohen et al. 1988). Samples (4 μl) were subjected to 10% SDS-polyacrylamide gel electrophoresis (PAGE) for protein analysis. After electrophoresis, gels were washed three times with deionized water (15 min) and rinsed with GelCode® Blue Stain Reagent (Pierce) overnight. Gels for protein sequencing were processed according to an established method (Matsudaira 1987). The gel, sandwiched between a PVDF membrane and several sheets of blotting paper, was assembled in a semi-dry blotting system (Poly Blot™ Transfer System Model SBD-1000; Bio-Rad) and electroeluted for 30 min at 0·5 A. The protein was sequenced using the automated Edman degradation procedure (Proteomics Center, Department of Biology, Technion, Haifa, Israel), and identified using the Blast program from the National Center for Biotechnology Information website (http://www.ncbi.nlm/nih.gov/blast/Blast.cgi). To examine the effect of free flagellin on adhesion, protein extracts of TBPZ cell outer membranes (100 μg ml−1, 90% w/v flagellin) were mixed with 1% (w/v) crushed gray chalk in phosphate buffer (pH 7·2, 10 mmol l−1) for 1 h prior to the addition of TBPZ cells (1 × 109 CFU ml−1).

TEM observation of flagella

TBPZ cells grown to logarithmic (16 h) or stationary (90 h) phases at 25°C in mineral medium were prepared for TEM observation (Grossart et al. 2000). The specimens were examined with a T-12 FEI TEM by using an accelerating voltage of 120 kV and images were obtained with Megaview-II (Soft Imaging Systems; Olympus, Munster, Germany).

Isolation of Achromobacter piechaudii TBPZ motility-defective mutants

Stationary phase A. piechaudii TBPZ cells were exposed to UV light to a 0·01% survival level (Kawagishi et al. 1995). The irradiated cells were spread on 1·5% (w/v) agar-PTYG plates and incubated at 30°C for 48 h. Mutants were identified based on their colony size: 800 small colonies were each cultured in 2-ml mineral medium containing 100 mg l−1 TBP for 90 h (stationary phase). The motility of these colonies was tested by spotting 3 μl of each strain on 0·3% agar-PTYG plates and incubating at 30°C for 20 h. Each swarming rings of each colony were measured and motility-defective mutants were identified.


Identification of flagella and flagellin levels in logarithmic and stationary cells

In a previous study, we found that stationary-phase TBPZ cells adhere to chalk more efficiently than logarithmic-phase cells (Nejidat et al. 2004). TEM examination revealed that stationary cells have thicker flagella than logarithmic ones (Fig. 1). To confirm this observation, outer membrane proteins were extracted from logarithmic (16 h) and stationary (90 h) A. piechaudii TBPZ cells. SDS-PAGE analysis showed a major protein band (65 kDa) that is much more abundant in stationary cells than logarithmic ones (Fig. 2). This protein’s N-terminal amino-acid sequence (MAQVINTLYLSLVAQN) shares identity with flagellin, the monomer of the flagellar filament (Soutourina and Bertin 2003), from many bacterial sources. The highest matching of 87% was to FlaA from Proteus mirabilis (accession P42272) and its molecular weight was similar to that of Escherichia coli (Li et al. 1993).

Figure 1.

 Transmission electron micrograph of flagella of TBPZ cells. TBPZ cells were grown for 16 or 90 h in TBP medium. Bacteria were placed onto Formvar-coated copper grids and negatively stained with 0·5% uranyl acetate. Scale bar, 2000 nm. Arrows show location of flagella.

Figure 2.

 SDS-PAGE analysis of flagellar extracts from TBPZ cells. Lane 1 extract of stationary cells (90 h). Lanes 2: extracts of logarithmic cells (16 h). Lane 3: molecular weight standards. Arrow shows location of flagellin band.

Involvement of flagella in adhesion to chalk

The possible role of flagella in the adhesion process was examined by physical and biochemical means. TBPZ cells were sheared by sonication without affecting viability (no differences in CFU compared with nonsonicated cells). The adhesion of sonicated stationary TBPZ cells to crushed gray chalk particles was reduced by 32% under conditions of cell saturation (Table 1).

Table 1.   Effect of sonication, flagellin-extract pretreatment and growth in the presence of azithromycin on adhesion of stationary TPBZ cells (1 × 109 CFU ml−1) on 1% (w/v) gray chalk
Treatment% Adhered cells*% Inhibition
  1. *Mean ± SD of four replicates from two separate experiments.

None80·1 ± 1·7
Sonication54·1 ± 2·532·4
3 mg l−1 Azithromycin55·0 ± 6·031·3
100 μg ml−1 Flagellin extract31·6 ± 4·060·5
100 μg ml−1 BSA53·2 ± 3·033·6

Flagellin expression was markedly repressed by 3 mg l−1 azithromycin (Fig. 3), resulting in a 31% reduction in TBPZ cell adhesion to the chalk at saturation (Table 1). Furthermore, preincubation of chalk particles with protein extracts of TBPZ cell outer membranes (100 μg ml−1) caused a 60% reduction in TBPZ adhesion to the chalk (Table 1), suggesting that free flagellin can occupy the adhesion sites on the chalk matrix. The same concentration of bovine serum albumin (BSA) caused a 33% reduction in TBPZ adhesion to the chalk, indicating nonspecific protein attachment to the adhesion sites.

Figure 3.

 SDS-PAGE analysis of flagellar extracts from TBPZ cells grown in TBP medium containing azithromycin. Lane 1: molecular weight standards. Lane 2: control. Lane 3: 1 mg l−1 azithromycin. Lane 4: 3 mg l−1 azithromycin. A representative experiment out of three.

Three motility-defective mutants were isolated after exposure to UV (Soutourina and Bertin 2003) and assigned the numbers 94, 153 and 208. The properties of the wild-type and mutant strains are summarized in Table 2. Analysis of flagellar extracts by SDS-PAGE showed that mutant strains 94 and 153 express comparable amounts of flagellin relative to the wild type (Fig. 4a). Furthermore, their cell shape and flagellar dimensions were similar to those of the wild type, as observed by TEM (data not shown). In contrast, mutant 208 expressed low amounts of flagellin (Table 2, Fig. 4a), had no flagella, and a spherical cell shape instead of the rod shape of wild-type cells (Fig. 4b).

Table 2.   Properties of the UV-induced motility mutants. Motility was tested on 0·3% agar plates after 20 h incubation
TBPZ strainsColony diameter (cm)% Adhered cells*FlagellaCell shape
Gray chalkWhite chalk
  1. *Mean ± sd of four replicates from two separate experiments.

WT1·05 ± 0·2192·3 ± 1·195·6 ± 0·4+Rod
940·5 ± 0·2867·3 ± 0·187·6 ± 1·4+Rod
1530·5 ± 0·1418·1 ± 0·345·9 ± 1·9+Rod
2080·35 ± 0·0196·8 ± 1·798·5 ± 0·2Spherical
Figure 4.

 (a) SDS-PAGE analysis of flagellar extracts from TBPZ motility-defective mutants. Lane 1: molecular weight standards. Lane 2: extracts of wild-type TBPZ cells. Lanes 3, 4, 5: extracts of TBPZ motility-defective mutant cells 153, 94 and 208, respectively. (b) Transmission electron micrograph of cells of TBPZ motility-defective mutant strain 208 (left) and TBPZ wild-type strain (right). TBPZ cells were grown for 90 h in TBP medium. Scale bar, 5 μm (left) and 10 μm (right). A representative experiment out of four.

TBPZ mutant strains 94 and 153 exhibited reduced adhesion ability to gray chalk and, to a lesser extent, to white chalk. Mutant 208, on the contrary, showed adhesion properties comparable to those of the wild type (Table 2). Mutant strains degraded TBP in the logarithmic growth phase as reported for the wild type and showed kinetic behaviour similar to that of the wild type (data not shown) (Nejidat et al. 2004).


The observation that TBPZ adhesion to chalk follows a Langmuir isotherm (Nejidat et al. 2004) suggests an interaction between the bacterial cells and specific sites on the chalk (McEachran and Irvin 1985; Staddon et al. 1990). Cell-wall structures such as flagella, pili and EPS can provide the necessary attachment structures for bacteria to bridge the repulsion barrier and adhere firmly to surfaces (Genevaux et al. 1999; Gomez-Suarez et al. 2002; Gusils et al. 2002). Stationary TBPZ cells possess longer and thicker flagella than logarithmic cells (Fig. 1) and are superior in their ability to attach to chalk surfaces (Nejidat et al. 2004). This led us to examine the possible contribution of the flagella to the bacteria’s site-specific adhesion process on chalk surfaces.

Several lines of evidence support flagella contribution to the adhesion of bacterial cells to chalk, although it is not the only factor. This includes reduced adhesion after sonication (32%), azithromycin (31%) treatment and pretreatment of chalk particles with extracts with outer membrane proteins (60%) (Table 1). The pretreatment of chalk particles with extracts of outer membrane proteins was the most effective at reducing the efficiency of the adhesion process to chalk possibly due to the blockage of sites on the chalk particles to which the flagella adhere. Similarly, it has been reported that the adhesion of Pseudomonas aeruginosa to Muc1 mucin on the epithelial cell surface is blocked by pretreatment of the bacteria with an antibody to flagellin or pretreatment of Muc1 cells with purified flagellin (Lillehoj et al. 2004).

The results suggest that flagella may contribute to the adhesion process in part through a physical interaction between the flagella and specific structures in the chalk. It has been suggested that interactions between chalk and organic compounds are site- and force-specific and that the specific adsorption sites are organic in nature (Wefer-Roehl et al. 2001; Graber and Bonisover 2003) Thus, these sites may also be involved in the attachment of TBPZ to the chalk.

Aside from their physical effect on the attachment to chalk particles, flagella can contribute to adhesion by affecting bacterial motility. Both chalk and bacteria repel each other because of their negative charges (Nejidat et al. 2004). The kinetic energy released by the movement of flagella may enable bacterial cells to overcome this repletion (Tomich et al. 2002; Young 2006). The observed behaviour of the motility-defective mutants indicates that adhesion to the chalk surface is a complex process. Although the TBPZ mutant strains 153 and 94 expressed flagellin (Fig. 4) and have flagella (data not shown), similar to the wild type, they were still nonmotile, perhaps as a result of a defective flagellar motor (Tomich et al. 2002). On the contrary, both mutant strains exhibited reduced adhesion capacity on chalk (Table 2), supporting the importance of motility to the adhesion process. Thus, flagella can be involved in bacterial adhesion to chalk by direct physical interactions and through motility. However, the properties of mutant strain 208 suggest the involvement of factors other than flagella in TBPZ’s adhesion to chalk. This mutant expresses low amounts of flagellin, is nonmotile and has no flagella. Nevertheless, mutant 208 exhibited adhesion capabilities identical to those of the wild type. These results indicate the presence of mechanisms that can compensate for the lack of flagella and motility in initial adhesion of bacterial cells to chalk surfaces. In the case of mutant 208, the spherical shape of this mutant may enable it to better fit the adhesion site. The spherical shape allows bacterial cells to settle into molecular grooves or canyons, which increases the number of possible cell-to-surface contacts (Young 2006).

In summary, the results of this study indicate that cell surface properties flagella, cell shape and motility affect TBPZ cell adhesion to chalk. Those characteristics are physiological status dependent (Young 2006) and sensitive to environmental conditions (Li et al. 1993). Our previous work demonstrated that TBPZ cells were more active when attached to chalk (Nejidat et al. 2004) and TBP degradation occurred in biofilms developed within natural fractured chalk (Arnon et al. 2005). Thus, the effect of the on-site prevailing conditions on cell surface properties have to be carefully considered in order to ensure efficient cell adhesion in practices aimed at producing an active biobarrier inside chalk fractures for the prevention of contaminants spreading in chalk aquifers.


This research was supported in part by a grant from the Israel Science Foundation (251/98).