Mechanisms of Cation Exchange by Pseudomonas aeruginosa PAO1 and PAO1 wbpL, a Strain with a Truncated Lipopolysaccharide

Abstract

The ability of bacterial cells to sequester cations is well recognized, despite the fact that the specific binding sites and mechanistic details of the process are not well understood. To address these questions, the cation-exchange behavior of Pseudomonas aeruginosa PAO1 cells with a truncated lipopolysaccharide (LPS) (PAO1 wbpL) and cells further modified by growth in a magnesium-deficient medium (PAO1 wbpL − Mg²⁺) were compared with that of wild-type P. aeruginosa PAO1 cells. P. aeruginosa PAO1 cells had a negative surface charge (zeta potential) between pH 11 and 2.2, due to carboxylate groups present in the B-band LPS. The net charge on PAO1 wbpL cells was increasingly positive below pH 3.5, due to the influence of NH₃⁺ groups in the core LPS. The zeta potentials of these cells were also measured in Na⁺, Ca²⁺, and La³⁺ electrolytes. Cells in the La³⁺ electrolyte had a positive zeta potential at all pH values tested. Growing P. aeruginosa PAO1 wbpL in magnesium-deficient medium (PAO1 wbpL − Mg²⁺) resulted in an increase in its zeta potential in the pH range from 3.0 to 6.5. In cation-exchange experiments carried out at neutral pH with either P. aeruginosa PAO1 or PAO1 wbpL, the concentration of bound Ca²⁺ was found to decrease as the pH was reduced from 7.0 to 3.5. At pH 3.5, the bound Mg²⁺ concentration decreased sharply, revealing the activity of surface sites for cation exchange and their pH dependence. Infrared spectroscopy of attached biofilms suggested that carboxylate and phosphomonoester functional groups within the core LPS are involved in cation exchange.

The ability of microbial cells to sequester metal ions from aqueous environments is important for microbial growth and biogeochemical processes, such as mineral formation/dissolution, metal transport, and the bioremediation of metal-contaminated sites (1). The external surfaces of gram-negative bacteria are comprised of phospholipids, lipoproteins, lipopolysaccharides (LPS), and proteins. These polymers contain carboxylic acids and phosphate esters, which are primarily responsible for the cells having a net negative charge (at neutral pH), and associated cations, which are retained even after thorough washing with water due to the electroneutrality condition (6, 12, 31). These cations may undergo cation exchange with the introduction of electrolyte solutions containing different cations. While many studies have focused on the amount of metal bound/adsorbed by cells and the development of bulk partitioning models (5), questions such as the mechanism of adsorption (20), the specific binding sites (9), and the impacts of different environmental conditions on adsorption (31) have received less attention. Such information is crucial for the development of models to predict the behavior of metals under a range of conditions and to determine the potential uses of these organisms for environmental remediation (11, 15).

Pseudomonas aeruginosa is a ubiquitous, well-characterized microorganism used in many metal-binding investigations (10). Suggested metal ion binding sites in P. aeruginosa include the phosphate and carboxyl groups in the peptidoglycan (29) and charged groups in both the core oligosaccharide region and the O-antigenic side chains that make up its LPS (14, 28). The LPS of P. aeruginosa extends up to 40 nm from the cell wall (13) and is comprised of an inner lipid section connected to a core oligosaccharide section, which has O-antigen side chains as outward extensions. These O-antigen side chains occur in two chemically and antigenically distinct forms (Fig. 1), termed A-band LPS and B-band LPS (26), whose relative expression appears to vary depending on the bacterial strain and environmental conditions (16).

FIG. 1.

The structure of the LPS component of the cell wall of P. aeruginosa PAO1 (a) and PAO1 wbpL (b) based on published literature (4, 7, 27). The LPS gene deletion separation points (cleavage sites) for the mutant strain PAO1 wbpL are indicated in panel a. The positions of the molecular insertions due to modified growth medium are indicated in panel b. The insets in panel a show an enlarged view of the sugar units making up the A- and B-band LPS. For the purpose of the figure, the ionizable groups are presented in their uncharged forms.

The core region of the LPS is primarily comprised of neutral sugars with a few ionizable sites associated with 2-keto-3-deoxyoctulosonic acid, phosphate, and amine groups. A-band LPS contains an O antigen with up to 20 trisaccharide repeating units of d-rhamnose and is electroneutral at physiological pH. B-band LPS contains an O antigen that is relatively long compared to A-band O antigen and is comprised of 30 to 50 repeat trisaccharide units of 2 residues of an amino derivative of manuronic acid and 1 residue of N-acetyl-d-fucosamine. The uronic acid residues result in B-band LPS having numerous negatively charged sites at neutral pH (14).

A number of isogenic mutants of P. aeruginosa have been isolated, and strains deficient in either one (A⁻ B⁺ or A⁺ B⁻) or both (A⁻ B⁻) of the LPS types have been used to investigate the role of the O-antigen side chains in metal binding (14, 17). As all four strains bound similar amounts of copper at the cell surface, the authors concluded that the major surface metal-binding sites must occur in portions of the LPS common to all strains. Interestingly, significant differences were observed in the abilities of some of the strains to bind iron or lanthanum and to form metal precipitates, suggesting that while the negatively charged sites located in the O-antigen side chains may not be directly responsible for the binding of metallic cations as a whole, they may contribute to overall cell surface properties that favor the precipitation of distinct metal-rich mineral phases (14). Makin and Beveridge (17) postulated that phenotypic variation in the relative expression of A- and B-band LPS may be a mechanism by which P. aeruginosa can alter its overall surface characteristics in such a way as to influence adhesion and favor survival.

In order to further elucidate the relative roles of the core regions and the A- and B-band LPS in metal binding, we have chosen to work with P. aeruginosa PAO1 and P. aeruginosa strain PAO1 wbpL, which does not contain the gene wbpL encoding the glycosyltransferase enzyme involved in the transport of the component sugars of the A- and B-band polymer strands. The LPS of this strain is therefore truncated, and it does not possess either A- or B-band O antigen (Fig. 1). The wild-type P. aeruginosa strain (PAO1) and PAO1 wbpL were further modified by growth in magnesium-deficient medium, which results in the incorporation of 4-amino-4-deoxyarabinose into the LPS core and palmitate into its lipid component (Fig. 1) (4, 7, 27). The addition of 4-amino-4-deoxyarabinose effectively reduces the number of negatively charged phosphate monoester groups and adds an amino moiety, which is positively charged at neutral pH.

This paper examines the cation-exchange properties of P. aeruginosa PAO1 and PAO1 wbpL cells grown in the presence or absence of magnesium and subsequently exposed to sodium, calcium, or lanthanum. The effect of cation exchange on the surface charge of the cell was assessed by zeta potential measurements, and information on the functional groups involved in cation exchange was obtained using infrared (IR) spectroscopy.

MATERIALS AND METHODS

Metal salts and solutions.

Aqueous solutions of 0.03 M NaCl (BDH AnalaR analytical reagent), 0.01 M CaCl₂ (BDH AnalaR analytical reagent), and 0.005 M LaCl₃ (BDH AnalaR analytical reagent) having identical ionic strengths were prepared by dissolving the metal salts in ultrapure deionized water (E-pure; Barnstead).

Growth and harvest of bacteria.

The P. aeruginosa wild-type strain (PAO1) and PAO1 wbpL, obtained from J. S. Lam (University of Guelph, Guelph, Ontario, Canada), were maintained as stock cultures in the Microbank cryovial system (Prolab Diagnostics) at −80°C. The cells were cultured on Vogels medium (30) comprised of 2 g citric acid (Pierce), 3.5 g NaNH₄HPO₄·7H₂O (BDH; 99%), 0.2 g MgSO₄·7H₂O (Ajax; analytical reagent), 10 g K₂HPO₄ (Riedel de Haen; analytical reagent), and 5 g of glucose (May and Baker) per liter of water. A glucose stock solution was filter (0.45 μm) sterilized and added to the other components previously sterilized by autoclaving. To induce LPS core modification, a magnesium-deficient medium was used based on Vogels medium, except that the 0.2 g MgSO₄·7H₂O was replaced by 0.005 g MgSO₄·7H₂O and 0.2 g NaSO₄ (BDH; analytical reagent) (4, 23). Cell counts and purity checks were carried out using tryptic soy agar (TSA) (Difco). To prepare a bacterial suspension, the medium (250 ml) was inoculated with a single colony from a 24-h streak plate and agitated on a shaking table at 170 rpm for 48 h at 25°C. The culture was centrifuged (8 min at 8,000 × g and 4°C), and the pellet was resuspended in 0.03 M NaCl, shaken, and washed by centrifugation two additional times.

Cation-exchange studies.

Cultures of P. aeruginosa PAO1 and PAO1 wbpL were washed as described above, resuspended in 250 ml of 0.03 M NaCl at a pH of about 5.4, and immediately used in cation-exchange trials. The cells were centrifuged and resuspended in 250 ml of 0.01 M CaCl₂. The pH was then adjusted to 7.5 using concentrated NaOH, and the solution was stirred for 1 h at room temperature. Following this opportunity for ion exchange, the cells were separated by centrifugation and the pellet was resuspended in 0.03 M NaCl (pH 8.5). The pH was recorded, and a 2-ml subsample was removed for metal analysis. The pH was then reduced using 2 ml of HCl (the acid solution strength increased as the pH decreased), and the solution was stirred for 15 min before a subsample was removed and the process was repeated. The acidification and sampling of the bacterial suspension was repeated 12 times until a final pH of approximately 1.8 was achieved. A similar methodology was previously used by Borrok et al. (2) to examine the effect of acid on the surface-bound concentrations of cations. Each subsample was filtered (0.45 μm) and analyzed for dissolved cations by atomic absorption spectroscopy using a Perkin Elmer (Norwalk, CT) model 3100 atomic absorption spectrometer. To determine the cell dry weight and the concentration of metal ions remaining associated with the cells, the cells were pelleted by centrifugation, washed (250 ml 0.03 M NaCl at pH 2.5), recovered by centrifugation, dried (60°C for 16 h), and reduced to ash (600°C for 5 h). The ash was dissolved in 10 ml of acidified water (1% HCl), and triplicate samples were analyzed for Ca²⁺ and Mg²⁺ (CaCl₂ and MgCl₂ standards; BDH AnalaR analytical reagent). The numbers of bacteria used in the cation-exchange experiments were assessed by diluting (0.1%; peptone) the samples and plating them onto TSA.

Zeta potential measurement.

For surface charge (zeta potential) determinations, P. aeruginosa PAO1 and PAO1 wbpL cells (250 ml) were grown in magnesium-rich or magnesium-deficient medium, washed as described above, and resuspended in 200 ml of 0.03 M NaCl at a pH of about 5.4. A 1-ml subsample of the resulting bacterial suspension was added to 9-ml volumes of either 0.03 M HCl or 0.03 M NaOH and 0.03 M NaCl. In this manner, an almost constant ionic strength was maintained over a range of pH values. The suspensions were held on ice until electrophoretic mobility measurements were taken (Zetasizer Nano ZS; Malvern Instruments UK). The complete experiment with independently grown cultures was carried out twice.

In a further series of experiments, P. aeruginosa PAO1 wbpL cells (250 ml) grown in a magnesium-rich or a magnesium-deficient medium were washed as described above and resuspended in 200 ml of 0.03 M NaCl at a pH of about 5.4. A 1-ml subsample of the resulting bacterial suspension was added to 9-ml volumes of either 0.03 M HCl or 0.03 M NaOH and either 0.01 M CaCl₂ or 0.005 M LaCl₃. In this manner, an almost constant ionic strength was maintained over a range of pH values. Zeta potential measurements were taken as described above. The complete experiment with independently grown cultures was carried out twice.

IR spectroscopic measurements.

Spectra were recorded using a Digilab FTS 4000 spectrometer (Cambridge, MA) equipped with a Harrick Horizon attenuated-total-reflection (ATR) accessory having a ZnSe 13-internal-reflection element (IRE). Total absorption of IR radiation by water can occur with such multiple reflection elements, giving rise to noisy features in mid-IR spectra at about 1,640 cm⁻¹.

For the spectroscopic studies with P. aeruginosa PAO1 and PAO1 wbpL, cell suspensions (250 ml) were grown in magnesium-rich medium, washed as outlined above, resuspended in 0.03 M NaCl at a pH of about 5.4, and divided into two 125-ml volumes. To form a P. aeruginosa PAO1 biofilm on the surface of the ZnSe IRE, a modification of the method outlined by Kang et al. (10) was used. A bacterial suspension (125 ml) containing approximately 10¹⁰ CFU ml⁻¹ was pumped across the surface of a ZnSe IRE for 2 h at a flow rate of 5 ml min⁻¹ in a continuous-flow loop system. Once a maximum absorbance had been reached, indicating the formation of a P. aeruginosa PAO1 monolayer, the pH of the cell suspension flowing across the biofilm-coated IRE was reduced to 2.6 by the addition of small volumes of HCl solution and directed to waste. This procedure was used to remove Mg²⁺ ions from the cells in the biofilm. Removal of Mg²⁺ was confirmed for P. aeruginosa PAO1 and PAO1 wbpL in the cation-exchange section of this work.

A monolayer of P. aeruginosa PAO1 wbpL cells was formed in a similar way. As the rate of cell attachment of this strain to the ZnSe was lower, the suspension was pumped across the IRE for 4 h prior to the acidification step. In addition, as the reduction in pH caused the P. aeruginosa PAO1 wbpL cells in suspension to flocculate, sonication (three times for 10 s each time) was used to deflocculate the cells.

To compare the effects of pH and the different electrolytes on the absorption spectra of the attached cells, the remaining 125-ml volume of P. aeruginosa PAO1 or PAO1 wbpL cell suspension was acidified to pH 2.6 for 20 min using a small volume of HCl solution. The suspension was then centrifuged and resuspended in 0.03 M NaCl, pH 7.0 (acid-washed cells). The first spectra (a, PAO1, or e, PAO1 wbpL) of the attached biofilm were recorded while the acid-washed bacterial suspension in 0.03 M sodium chloride electrolyte at pH 7.0 was circulated. The pH of the flowing solution was decreased to 2.6 with a small volume of HCl solution, and the spectra (b, PAO1, or f, PAO1 wbpL) were recorded. CaCl₂ was then added to the bacterial suspension to give a concentration in the cell suspension of 0.03 M NaCl and 0.01 M CaCl₂. The pH was increased to 7.0 using a small volume of NaOH. The spectra (c, PAO1, or g, PAO1 wbpL) were recorded after 10 min. The pH was then reduced again to 2.6 using a small volume of HCl solution, and LaCl₃ was added to the bacterial suspension to give concentrations in the flowing cell suspension of 0.03 M NaCl, 0.01 M CaCl₂, and 0.005 M LaCl₃. The pH was readjusted to 7.0 using a small volume of NaOH. The spectra (d, PAO1, or h, PAO1 wbpL) were recorded after 10 min. The purpose of circulating the bacterial suspension over the attached cells was to take advantage of the buffering capacity of bacteria in controlling the solution pH.

RESULTS AND DISCUSSION

Characterization of the two PAO1 strains.

P. aeruginosa PAO1 and the mutant strain, PAO1 wbpL, had mean generation times (doubling rates) and standard deviations from three independently grown cultures in Vogels medium (magnesium rich) at 25°C of 128 ± 7 and 164 ± 31 min, respectively. In magnesium-deficient medium, the generation times were 150 ± 10 and 183 ± 21 min, respectively. On TSA plates, P. aeruginosa PAO1 had typical smooth margins while P. aeruginosa PAO1 wbpL colonies displayed rough margins, characteristic of A- and B-band LPS-minus strains.

The surface charge (zeta potential) data for P. aeruginosa PAO1 or PAO1 wbpL grown in magnesium-rich medium and for P. aeruginosa PAO1 wbpL grown in magnesium-deficient medium (PAO1 wbpL − Mg²⁺) over the pH range of 2.2 to 11.0 are shown in Fig. 2. For P. aeruginosa PAO1 cells, the surface charge was negative at all measured pH values down to pH 2.2 due to the high number of carboxylate groups present in the B-band LPS. In contrast, P. aeruginosa PAO1 wbpL exhibited an increasingly positive surface charge below pH 3.5, believed to be due to positively charged —NH₃⁺ groups in the core LPS. Growing P. aeruginosa PAO1 wbpL in magnesium-deficient medium (PAO1 wbpL − Mg²⁺) resulted in an increase in its zeta potential in the pH range from 3.0 to 6.5. This was believed to be due to the insertion of 4-amino-4-deoxyarabinose into the core LPS, resulting in a change in the terminal charge-carrying group from a phosphomonoester to a phosphodiester, which contains, at neutral pH, a positively charged terminal amine. At pH 11.0, there was no difference in surface charge between the cells, as the charge-carrying groups were all deprotonated (amine groups, pK_a = 9.0 to 11.0; carboxylic acid groups, pK_a = 2.0 to 6.0 [mean, 4.5]; phosphodiesters, pK_a = 3.2 to 3.5; and phosphomonoester, pK_a1 = 2.0 to 4.0 and pK_a2 = 5.6 to 7.2) (3, 18, 19).

FIG. 2.

Zeta potential measurements of P. aeruginosa PAO1 grown in a magnesium-rich medium (•) and P. aeruginosa PAO1 wbpL grown in magnesium-rich (PAO1 wbpL) (▵) or magnesium-deficient (PAO1 wbpL − Mg²⁺) (□) medium.

Interestingly, it was observed that P. aeruginosa PAO1 wbpL cells in suspension had a tendency to aggregate at pH values between 3.0 and 5.0, a characteristic not observed with P. aeruginosa PAO1 cells, the suspensions of which were stabilized by electrostatic repulsion over the entire pH range tested and by their greater hydrophilicity. The aggregation of P. aeruginosa PAO1 wbpL cells over this pH range is due to their close-to-net-neutral surface charge and hydrophobicity.

IR absorption spectra were obtained from monolayers of P. aeruginosa PAO1 or PAO1 wbpL cells attached to the surface of a ZnSe IRE (Fig. 3). During the attachment process, the bacterial absorption intensity increased with time (data not shown), indicating that the number of cells attached to the ZnSe surface was increasing. Microscopic studies in our laboratory have previously shown that P. aeruginosa PAO1 readily forms a bacterial monolayer on ZnSe under similar flow conditions (10). The ATR-IR spectra exhibit distinct bands, around 1,646 cm⁻¹ (amide I, mainly the C=O stretch), 1,548 cm⁻¹ (amide II, mainly the N—H bend), 1,454 cm⁻¹ (partly C—H deformation), 1,402 cm⁻¹ (partly symmetric carboxylate stretch), and 1,235 cm⁻¹ (amide III, mainly the P=O, C—O—C antisymmetric stretch of ester groups) and in the 1,100- to 1,000-cm⁻¹ region (P—O and C—OH stretch) (31). The spectrum of P. aeruginosa PAO1 (Fig. 3) was more intense than the spectrum of P. aeruginosa PAO1 wbpL due to the P. aeruginosa PAO1 cells attaching more readily than the LPS-truncated P. aeruginosa PAO1 wbpL cells. This is believed to be due to the difference in hydrophobicity between the bacterial surfaces. P. aeruginosa PAO1 has a more hydrophilic surface due to the presence of B-band O antigen, whereas the sugars at the surface of P. aeruginosa PAO1 wbpL are similar to those found in the more hydrophobic A-band O antigen. Also noticeable is that attached P. aeruginosa PAO1 wbpL cells have a less pronounced symmetric-stretch carboxylate peak at about 1,400 cm⁻¹ and differences in band shape between 1,200 and 900 cm⁻¹ compared to P. aeruginosa PAO1 cells. This is believed to be due to the different carbohydrates present on the surfaces of the two strains.

FIG. 3.

ATR-IR spectra of P. aeruginosa PAO1 (a) and P. aeruginosa PAO1 wbpL (b) cells attached to the surface of a ZnSe IRE at pH 7.0.

Cation exchange.

P. aeruginosa PAO1 or PAO1 wbpL cells cultured in magnesium-rich medium and suspended in 0.03 M NaCl were resuspended in a 0.01 M CaCl₂ solution at pH 7.5 for 1 h with mixing to facilitate Ca²⁺ for Mg²⁺ exchange. The cells were subsequently washed and resuspended in a 0.03 M NaCl solution at pH 8.5, which gave a final suspension pH of about 7.5. It was important to carry out the experiment in this manner to prevent a decrease in pH causing the metal to be lost from the cell before it could be measured.

As the pH of the suspension was decreased to pH 1.8 by the addition of small volumes of HCl, the concentrations of Mg²⁺ and Ca²⁺ ions in suspension were measured and the amounts of Mg²⁺ and Ca²⁺ associated with the cells were calculated and plotted against the pH (Fig. 4). For P. aeruginosa PAO1 or PAO1 wbpL cells, the decrease in pH from 7.0 to 3.5 resulted in a decrease in the concentration of bound Ca²⁺. The bound Mg²⁺ concentration remained relatively constant until there was a sharp decrease at around pH 3.5. Interestingly, the decrease in bound Mg²⁺ corresponded to a slight increase in the concentration of bound Ca²⁺ at pH values from 3.5.to 2.5. This was attributed to Ca²⁺ being exchanged for Mg²⁺ due to protonation at the metal-binding site (exchange of H⁺ for Mg²⁺, followed by exchange of Ca²⁺ for H⁺). There was a further loss of bound Ca²⁺ and Mg²⁺ as the pH decreased to 1.8. At the conclusion of all the pH titration experiments, the concentration of Mg²⁺ and Ca²⁺ remaining bound to the cells was less than 0.0016 mmol g⁻¹ dry weight, confirming that the majority of the Ca²⁺ and Mg²⁺ had been released during titration to pH 1.8. Viable-cell counts and optical-density measurements (at 600 nm) (data not shown) of PA01 and PA01 wbpL cell suspensions after 30 min at pH 3 confirmed that increases in the concentration of calcium and magnesium in solution were not due to lysis of the cells.

FIG. 4.

The Ca²⁺ (solid symbols) and Mg²⁺ (open symbols) concentration associated with PAO1 (squares) or PAO1 wbpL (triangles) cells as the pH is decreased from 7.0 to 1.8.

As the loss of the A- and B-band LPS did not significantly affect the total divalent-cation-exchange capacity of the cell, the negatively charged groups in the core LPS appeared to play a dominant role in cation exchange. Potential Ca²⁺ binding groups in the core LPS include phosphate monoesters and carboxylate groups. This result is in agreement with the divalent-cation-exchange results of other researchers, who stated that removal of A- and B-band O antigen has little impact on the binding of Cu²⁺ (14).

Impact of cation exchange on zeta potential.

P. aeruginosa PAO1 wbpL cells were cultured in magnesium-rich or -deficient medium, washed, and resuspended in either 0.03 NaCl, 0.01 CaCl₂, or 0.005 LaCl₃ over a pH range of 2.5 to 11.0, and their zeta potentials were measured (Fig. 5). By using P. aeruginosa PAO1 wbpL cells, the role of charge in the core LPS on metal binding could be examined independently of the charged groups of the B-band O antigen. Cells suspended in solutions of the different cations displayed strikingly different zeta potential profiles in the pH range between 3.0 and 11.0. As the charge on the cation present in the electrolyte solution increased, the zeta potential of the cell increased. We postulate that as the negatively charged groups on the cell surface become deprotonated with increasing pH, H⁺ ions are exchanged by cations present in the electrolyte solution, with the presence of di- and trivalent cations consequently increasing the absolute charge on the cell compared to cells in the presence of monovalent (Na⁺) cations. This result is in agreement with our cation-exchange experiments and further supports the postulate that cells can bind cations in the absence of A- and B-band O antigen. At pH values above 6.0, P. aeruginosa PAO1 wbpL cells suspended in LaCl₃ flocculated, with the formation of a white precipitate. The increased zeta potential in this pH range may be partly due to precipitation of a positively charged lanthanum species on the surfaces of the cells, as shown previously in scanning electron microscopy studies (14).

FIG. 5.

Zeta potential measurements of PAO1 wbpL grown in magnesium-rich (a) or -deficient (b) medium and resuspended in either 0.005 M LaCl₃ (▵), 0.01 M CaCl₂ (□), or 0.03 M NaCl (○).

P. aeruginosa PAO1 wbpL grown in magnesium-rich and magnesium-deficient media showed different zeta potential profiles when suspended in NaCl and CaCl₂ but a similar change in the zeta potential pH profile with the changing valency of cations present in solution. The increased isoelectric point is believed to be due to the insertion of 4-amino-4-deoxyarabinose into the inner phosphate moiety (Fig. 1). The similar increases in zeta potential upon cation valency increase suggests that the negatively charged groups being balanced by the di- and trivalent cations are not the inner phosphate groups that are blocked by the 4-amino-4-deoxyarabinose insertion in PAO1 wbpL − Mg²⁺ cells. It is likely that metal ions associated with these functional groups present in unmodified P. aeruginosa PAO1 wbpL cells undergo cation exchange only at pH values below 3.0. By careful examination of zeta potential profiles, it may be possible to further investigate ion-exchange processes occurring in the core LPS and to probe the affinities of a bacterial cell's surface for different cationic species.

IR spectroscopy.

The association of metal ions with anionic bacterial-surface functional groups is expected to give rise to perturbations of the IR spectra. To confirm the roles of these functional groups in metal binding and to understand the role of bacterial surface structures in lanthanum salt crystallization, P. aeruginosa PAO1 and PAO1 wbpL were attached to the surface of a ZnSe IRE by allowing a suspension of the cells in 0.03 M NaCl to flow over the IRE surface. Attached cells were subsequently exposed to metal ions by a sequence of steps summarized in Fig. 6.

FIG. 6.

Flow chart outlining the steps carried out prior to obtaining P. aeruginosa PAO1 or PAO1 wbpL ATR-IR spectra.

Spectra a and b in Fig. 7A show the absolute spectra of PAO1 at pH 7.0 and 2.6. Noticeable changes occur on pH reduction, with an increase in a band at about 1,720 cm⁻¹ corresponding to —COOH groups and a decrease in bands at 1,580 and 1,400 cm⁻¹ corresponding to carboxylate groups. There are also minor band shape alterations with pH in the broad band at about 1,100 cm⁻¹ containing phosphate absorptions. Apart from these specific changes, there are also fairly general band absorbance increases with pH reduction, which have been previously attributed to compression of P. aeruginosa PAO1 surface polymers with reduced repulsion between negatively charged carboxylate groups (21). Differences in this behavior from that previously reported (21) may be due to the different metal contents of these PAO1 bacteria. Spectra e and f in Fig. 7B show the corresponding spectra of P. aeruginosa PAO1 wbpL at pH 7.0 and 2.6. While these spectra are weaker, they show similar behavior with reduction in pH. There is a less pronounced spectral-intensity change and lesser intensity changes of the carboxylate bands with pH, which is consistent with the removal of the charged O-antigen polymers.

FIG. 7.

ATR-IR spectra of PAO1 (A) or PAO1 wbpL (B) at either pH 7.0 (solid line, a or e) or 2.6 (dashed line, b or f) and difference spectra showing the change that occurs when either 0.01 M calcium chloride (c or g) or 0.005 M lanthanum chloride (d or h) is added. The pairs of spectra a and b, c and d, e and f, and g and h share absorbance scale bars of 0.05, 0.01, 0.01, and 0.004, respectively.

Spectra c and g in Fig. 7 are difference spectra recorded at pH 7.0 after the addition of CaCl₂ to the circulating bacterial suspension at pH 2.6 and adjustment to pH 7.0. The addition of CaCl₂, NaOH, and HCl to the bacterial suspension increased its ionic strength. Polymer compression and a general increase in intensity occur as the ionic strength is increased. This is due to charge screening of negatively charged carboxylate groups in the B-band O antigen of P. aeruginosa PAO1 (21). To accommodate the general spectral absorbance increases with ionic strength (raw data not shown), a scaled subtraction, based on the amide II band at 1,550 cm⁻¹, which is unaffected by pH, was used to obtain the difference spectra (c and g). The prominent features in these spectra at about 1,640 cm⁻¹ arise from the total absorption of radiation by water at this wave number. The remaining spectral features in spectrum c are not very pronounced. However, there are minor changes in the regions of the carboxylate groups at about 1,580 and 1,400 cm⁻¹, as well as an absorption at 1,100 cm⁻¹ in the vicinity of phosphate absorptions. In spectrum g, the absorption at 1,100 cm⁻¹ is less pronounced due to the smaller number of attached cells, while there is little evidence for any spectral change associated with carboxylate groups.

Spectra d and h in Fig. 7 are scaled difference spectra recorded at pH 7.0 after the addition of LaCl₃ to the circulating bacterial suspension at pH 2.6 and adjustment to pH 7.0. Scaling accommodates the influence of ionic strength on spectral band intensities. Spectrum d for P. aeruginosa PAO1 has prominent features having the appearance of peak first-derivative spectra centered about 1,566, 1,409, and 1,096 cm⁻¹. Such bipolar peaks at 1,506 and 1,409 cm⁻¹, as well as at 1,096 cm⁻¹, correspond to spectral band shifts of carboxylate and phosphate groups, respectively (24). The wave number separation in bipolar peaks of difference spectra greatly exaggerate the actual band shifts and are thus sensitive indicators of small band shifts. These features directly indicate that La³⁺ ions interacted with negatively charged carboxylate and phosphate groups (22) and exchanged with existing cations in the available binding sites on the bacterial surface. The band at 1,203 cm⁻¹ is likely due to the P=O stretch in (—OPO₃La)⁺, as lanthanum binding resulted in the negative charge no longer being delocalized on multiple oxygen atoms (25). The remaining spectral features in spectrum d at about 1,137 and 1,485 cm⁻¹ are unassigned. Spectrum h for PAO1 wbpL shows less pronounced spectral changes, which are similar to those in spectrum d.

In summary, the IR spectra clearly show that La³⁺ is associated with carboxylate and phosphate groups in both P. aeruginosa PAO1 and PAO1 wbpL. Perturbations in the same spectral regions for PAO1 cells suspended in calcium chloride suggest that Ca²⁺ also interacts with these functional groups. As P. aeruginosa PAO1 wbpL cells, which do not possess carboxylate-containing B-band O antigen, gave similar spectral shifts, we postulate that these metal ions are bound by the 2-keto-3-deoxyoctulosonic acid of the LPS common to both strains and the negatively charged phosphate monoester groups in the LPS core region. The additional spectral feature in the region of the P=O stretch for cells immersed in La³⁺ at pH 7 and its absence when they are immersed in Ca²⁺ suggest that lanthanum precipitate formation on the surfaces of PA01 and PA01 wbpL cells involves inner-sphere coordination of La³⁺ to phosphomonoester groups.

The use of zeta potential pH profiles of O-antigen-truncated strains to examine the charge within the core section of the LPS may aid in future study of antimicrobial resistance mechanisms and their relationship to the affinity of a bacterial cell's surface for cationic antimicrobial peptides (8). In general, the results obtained support the postulate that the main metal-binding regions of PAO1 are in the core section of the LPS, suggesting, therefore, that the selection of O-antigen traits does not affect the metal-binding ability of the cell.