Article pubs.acs.org/est
Sulfhydryl Binding Sites within Bacterial Extracellular Polymeric Substances Qiang Yu* and Jeremy B. Fein Department of Civil & Environmental Engineering & Earth Sciences, University of Notre Dame, Notre Dame, Indiana 46556, United States S Supporting Information *
ABSTRACT: In this study, the concentration of sulfhydryl sites on bacterial biomass samples with and without extracellular polymeric substances (EPS) was measured in order to determine the distribution of sulfhydryl sites on bacteria. Three different approaches were employed for EPS removal from Pseudomonas putida, and the measured sulfhydryl concentrations on bacterial EPS molecules are independent of the EPS removal protocols used. Prior to EPS removal, the measured sulfhydryl sites within P. putida samples was 34.9 ± 9.5 μmol/g, and no sulfhydryl sites were detected after EPS removal, indicating that virtually all of the sulfhydryl sites are located on the EPS molecules produced by P. putida. In contrast, the sulfhydryl sites within the S. oneidensis samples increased from 32.6 ± 3.6 μmol/g to 51.9 ± 7.2 μmol/g after EPS removal, indicating that the EPS produced by S. oneidensis contained fewer sulfhydryl sites than those present on the untreated cells. This study suggests that the sulfhydryl concentrations on EPS molecules may vary significantly from one bacterial species to another, thus it is crucial to quantify the concentration of sulfhydryl sites on EPS molecules of other bacterial species in order to determine the effect of bacterial EPS on metal cycling in the environment. adsorption in the presence of EPS,16,17,21,22 while some report negligible18,19,23 or even negative effects20 by the EPS on the extent of metal adsorption onto bacteria. A number of factors may cause these contradictory results. First, very few of these previous studies tested whether significant cell lysis occurred during EPS removal, and cell lysis may affect the concentration of binding sites within the biomass. Cell lysis causes the cell membrane to rupture, thereby exposing internal functional groups for additional metal binding. Conversely, the degradation of the cell wall during cell lysis may destroy binding sites and thereby decrease their concentration in the biomass. For example, several studies have claimed that EPS enhances Cu and Cd adsorption onto Pseudomonas putida based on the observation that adsorption was inhibited after EPS removal by exposure of the biomass to cation exchange resins for 24 h.17,22 However, significant cell lysis can occur after 2 h of cation exchange treatment,24,25 suggesting that the decreased adsorption may have been due to a combination of the effects of cell lysis and EPS removal. Second, different EPS removal approaches, including genetic modification of cells to induce minimal EPS production,26 cation exchange resin treatment,17,18,22,27 heating,21 enzyme cleavage,19,23 and ultrasonication,16 were applied in previous studies, which could
1. INTRODUCTION Microbial biofilms are present in soils, sediments, natural waters and engineered systems, and they consist of cells surrounded by a matrix of biopolymers, termed extracellular polymeric substances (EPS). EPS molecules are mainly long biopolymers of proteins, polysaccharides, DNA and lipids,1 and they can extend tens of microns or more from the cell surface.2 Although it is possible to subdivide the EPS into distinct structural species,2 for the purpose of this study and to simplify this initial attempt at thermodynamic modeling, we consider them as a single entity and study their behavior as a whole. Metal binding by EPS can reduce the stress of toxic metals on bacterial cells3 and can enhance the bioavailability of nutrient metals.4 Both effects tend to enable bacterial cells to survive under extreme conditions. The binding of metals onto EPS also strongly affects many natural and engineered processes associated with bacteria, such as biomineralization,5 bioflocculation,6 metal immobilization,7,8 biofouling of reverse osmosis membranes9 and corrosion of water pipes.10,11 As a result, it is of particular importance to understand metal adsorption onto EPS, which is largely controlled by metal binding sites on EPS molecules, such as carboxyl, phosphoryl, sulfhydryl, and amino groups.3 Numerous studies have characterized the proton and metal binding capacity of EPS that is extracted from either planktonic cells, biofilms, or sludge flocs, demonstrating that significant concentrations of binding sites are present on EPS molecules.4,12−15 However, contrasting results on the effects of EPS on metal adsorption onto bacterial biomass have been reported.16−20 Some of these studies report enhanced metal © XXXX American Chemical Society
Received: January 21, 2016 Revised: May 9, 2016 Accepted: May 13, 2016
A
DOI: 10.1021/acs.est.6b00347 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
different types of interactions that these species have with metals in their environment.
result in different yields and compositions of the removed EPS.28,29 In general, the metal-binding capacity of EPS is directly related to the measured concentration of proton-binding sites on EPS molecules as measured by potentiometric titrations.17,19,26 However, some previous studies have reported the opposite effects of EPS molecules on proton-binding and metal-binding by bacterial biomass. For example, Ha et al.26 measured the concentration of proton binding sites of wild-type (EPS-producing) and EPS-inhibited Shewanella oneidensis strains in 0.1 M NaNO3, and obtained similar total site concentrations, but found that the extent of Pb adsorption onto the wild-type cells was significantly lower than onto the EPSinhibited cells. Similarly, Gonzalez et al.20 reported that EPSrich Pseudomonas aureofaciens samples contain a higher concentration of metal-binding sites, but adsorb less Cu than EPS-poor samples that have a lower concentration of binding sites. These results suggest that further characterization of the binding sites on EPS molecules is needed. Among the various types of binding sites on EPS molecules, sulfhydryl sites (-SH) are of particular importance due to their high affinities for chalcophile metals such as Hg, As, Cd, Cu, and Pb, and an earlier study12 suggested the presence of sulfhydryl sites on EPS molecules from some biofilms. Despite the low abundance of sulfhydryl sites within cell envelopes, Extended X-ray Fine Structure (EXAFS) spectroscopy and batch adsorption measurements show that the adsorption of Cd, Hg, Au, and Cu onto bacterial cells is dominated by metal-sulfhydryl binding when metal:bacteria ratios are below several μmol/g (wet weight).30−35 However, currently there have been no direct measurements of the concentration of sulfhydryl binding sites on EPS molecules, making it impossible to determine the distribution of sulfhydryl sites within biofilms or to understand the interactions between metals and EPS-hosted sulfhydryl binding sites. In a previous study,36 we developed an approach for the detection of bacterial sulfhydryl sites by coupling a selective site-blocking technique37 with potentiometric titrations and surface complexation modeling. We used this method to determine the total concentrations of sulfhydryl sites on a variety of bacterial species. In this study, we focus on the concentration of sulfhydryl sites within the EPS molecules that are produced by bacteria. In order to determine the concentration of sulfhydryl binding sites on EPS molecules, we measured the sulfhydryl concentrations within bacterial biomass samples before and after removal of the EPS. The experiments involved a prolific EPS-producing species Pseudomonas putida, and three different EPS-removal approaches were tested and compared, including a cation exchange resin treatment, enzyme cleavage, and ethylenediaminetetraacetic acid (EDTA) treatment. We used Live/Dead staining tests to estimate the extent of cell lysis that occurred during each type of EPS removal procedure, and we used SEM imaging to monitor the efficiency of EPS removal by the different approaches. In addition to studying the EPS from P. putida, for comparison we also measured the sulfhydryl concentrations before and after EPS removal on biomass samples of Shewanella oneidensis, a Gram-negative bacterium with similar concentrations per mass of biomass of sulfhydryl binding sites to P. putida. This study is the first to quantify the concentrations of sulfhydryl sites on EPS molecules. We observed different distributions of sulfhydryl sites for S. oneidensis and P. putida, and these differences likely reflect the
2. MATERIALS AND METHODS 2.1. Preparation of Bacterial Cells with and without EPS. The procedures for growth and purification of Shewanella oneidensis MR-1 (ATCC#: 700550) and Pseudomonas putida (ATCC#: 33015) cells followed those described previously.36 Briefly, bacteria were first cultured aerobically in 3 mL of trypticase soy broth with 0.5% yeast extract at 32 °C for 24 h and then transferred to 2 L of growth medium of the same composition at 32 °C for the desired incubation time. To study the effect of growth phase on sulfhydryl concentration, the incubation time of P. putida was set to 6, 24, and 120 h, representing early exponential, early stationary and late stationary phases of bacterial cells, as determined by growth curve measurements (data not shown). In all other experiments, 24 h was used as the incubation time in order to yield bacterial cells in early stationary phase. After incubation, bacterial cells were harvested by centrifugation at 10 970g for 5 min. The obtained biomass pellets were rinsed with 0.1 M NaCl three times, with each rinse followed by centrifugation at 8100g for 5 min. Finally, the biomass pellets were transferred into preweighed test tubes and centrifuged for two 30 min intervals at 8100g, and the wet weight of the cells was recorded after decanting the supernatant. In this study, all bacterial concentrations are reported in terms of this wet weight of the biomass. To calculate the wet weight to dry weight conversion rates for these bacterial species, approximately 2 g (wet weight) of washed biomass was dried at 100 °C for 24 h until the weight did not change with time. The resulting ratios of wet weight to dry weight are 4.52 for P. putida and 4.75 for S. oneidensis. EPS molecules were removed from some biomass samples in order to compare the sulfhydryl concentration of the EPSremoved biomass with that of the biomass samples with EPS. To remove EPS, the freshly harvested and rinsed cell pellets were immediately resuspended in 0.1 M NaCl with either 2% EDTA38 (Aldrich Chemical, disodium salt, 0.6 g of EDTA/g of biomass in wet weight), a cation exchange resin25 (Dowex Marathon C sodium form, 20−50 mesh, 30 g of resin/g of biomass in wet weight) or the enzyme glucoamylase23 (MP Biomedicals, 100 units of glucoamylase/g of biomass in wet weight), and allowed to react for 2 h at room temperature (∼20 °C) with slow stirring in order to maintain homogeneous suspensions. The EPS on P. putida cells was removed using each of the three removal methods separately in order to compare their extraction efficiencies. Because we did not observe significant differences in EPS sulfhydryl site concentrations between the different removal methods (see Results and Discussion), for the S. oneidensis experiments, we only used the cation exchange resin treatment to remove EPS for all subsequent experiments. After reaction, bacterial cells with EPS removed were collected by centrifugation at 8100g for 5 min, and the obtained biomass pellets were rinsed with 0.1 M NaCl three times, each followed by centrifugation at 8,100 g for 5 min. Finally, the biomass pellets were transferred into preweighed test tubes and centrifuged for two 30 min intervals at 8100g, and the wet weight of the cells was recorded after decanting the supernatant. 2.2. Characterization of Bacterial Cells with and without EPS. P. putida cells with and without EPS were stained using a LIVE/DEAD BacLight bacterial viability kit, which contains SYTO9 stain and propidium iodide, and then B
DOI: 10.1021/acs.est.6b00347 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology observed using fluorescence microscopy in order to test the integrity of cell membranes in each sample. While SYTO9 alone labels both dead and live cells and causes them to fluoresce green, propidium iodide penetrates only those cells with damaged membranes and reduces SYTO9 fluorescence, causing dead cells to fluoresce red. The biomass samples for scanning electron microscope (SEM) imaging were prepared by fixing the cells and EPS (if any) in 2% glutaraldehyde and then in 2% osmium tetraoxide in 0.1 M NaCl. The fixed cell pellets were then washed sequentially using 50%, 70%, 80%, 95%, and 100% ethanol, followed by critical point drying. Finally, the obtained samples were mounted onto adhesive carbon tape that attached to sample holders, sputter-coated with Au−Pd, and imaged at an acceleration voltage of 5−6 kV under high vacuum using a LEO EVO 50 SEM. 2.3. Determination of Sulfhydryl Concentrations. The approach used to determine the sulfhydryl concentrations of bacterial cell envelopes with and without EPS followed the same procedure that we developed in a previous study.36 Monobromo(trimethylammonio)bimane bromide (qBBr) effectively blocks the protonation ability of sulfhydryl sites,36,37 but does not react with carboxyl or phosphoryl sites and is not proton-active itself.36 After selectively blocking cell envelope and/or EPS sulfhydryl sites using qBBr, we conducted potentiometric titrations of suspensions of cells with and without EPS removed, coupled with surface complexation modeling of the resulting data. The decrease in the total site concentration that occurred in response to EPS removal was taken to be the sulfhydryl concentration within the EPS. In the present study, qBBr, purchased from Santa Cruz Biotechnology, Inc., was used to create a pretreatment solution which was exposed to the suspended biomass in order to block cell envelope and EPS sulfhydryl sites to proton binding. The bacterial pellets were suspended in a freshly prepared qBBr solution in 0.1 M NaCl with pH buffered to 7.0 ± 0.1 using a 1.8 mM Na2HPO4/18.2 mM NaH2PO4 buffer, with a qBBr:biomass ratio of approximately 70 μmol/g, and the mixture was allowed to react for 2 h at room temperature under continuous shaking on a rotating plate at 60 rpm. Our previous study demonstrated that the qBBr adsorption reaction onto the cell envelope sulfhydryl sites is complete after 2 h of reaction.36 Potentiometric titrations were conducted using a T70 autotitrator from Mettler Toledo, Inc. and using 1 M HCl or NaOH standards with predetermined concentrations purchased from Fluka Chemical Corp. Bacterial cells with and without EPS, and with and without qBBr treatment, were suspended in the 0.1 M NaCl solution to achieve a homogeneous bacteria suspension with a concentration of approximately 30 g/L, and 10−11 mL of suspension was titrated in each titration. Before each titration, the 0.1 M NaCl solution was purged with N2 for at least 1 h in order to remove dissolved CO2. All the titrations were conducted under a N2 atmosphere and each suspension was stirred continuously with a magnetic stir bar. Each titration consisted of two steps: (1) acidifying the bacterial suspension to pH 3.0 by adding HCl standard; and (2) a forward titration from pH 3.0 to pH 9.7 by adding NaOH standard. Only these forward titration data were used for calculating the total sulfhydryl site concentrations. The titrator was set to operate using a method in which the equilibration time for each step of the titration was controlled, and the volume of acid or base added at each step was recorded, with a minimum addition volume of 0.25 μL. New titrant was added after the signal drift reached a minimum stability of 0.01 mV/s, or after a maximum
waiting time of 60 s was achieved. In preliminary experiments, backward titrations from pH 9.7 to pH 3.0 immediately followed by the forward titrations were conducted and the obtained titration curves matched well with their corresponding forward titrations, suggesting rapid reversibility of the protonation reactions and that no significant damage occurred to the cells during the forward titrations. In order to compare titration results from different experiments, the results were plotted in terms of a mass normalized net concentration of protons added to the system: [H+]net added = (Ca − C b)/m
(1)
where Ca and Cb are the total concentrations of acid and base added to the system during the titration, respectively, with units of mmol/L wet biomass, and m is the bacterial concentration in the suspension, with units of g/L. For each set of bacterial cells (with or without EPS and with or without qBBr treatment), the titration experiments were repeated at least three times. We model the proton-active functional groups on bacterial cell envelopes as discrete monoprotic acids, whose deprotonation reactions can be expressed using the following reaction:39,40 R − Ai H ° ↔ R − Ai− + H +
(2)
where R denotes the bacterial envelope macromolecule to which the ith organic acid functional group, Ai, is attached. Note that although we write Reaction 2 for an organic acid functional group, the same approach applies to protonation of amino-type sites as long as it is proton-active. The acidity constant (Ka,i) and the concentration (Ci) of the ith site can be expressed as Ka , i =
[R − Ai−]a H + [R − Ai H °]
(3)
Ci = [R − Ai−] + [R − Ai H °] −
(4)
where [R − Ai ] and [R − AiH ] represent the concentrations of the deprotonated and protonated ith organic acid functional group on the bacterial cell envelope, respectively, and aH+ is the activity of H+ in bulk solution. Based on proton mass balance, the proton added to the system at any point of the titration can be described as 0
Ca − C b = [H+] − TH 0 − [OH−] −
∑ [R − Ai−]
(5)
0
where TH represents the initial proton concentration at the commencement of the titration, [X] represents the concentration of species X in the experimental system, including H+, OH− and all the deprotonated organic acid functional groups within the bacterial cell envelope. FITEQL41 was used as a modeling tool for optimization of TH0, Ka,i, and Ci in eqs 3, 4, and 5 to best fit the titration data and to solve for these unknown parameters, following the approach described by Fein et al.40 In our previous study,36 we found that a 4-site nonelectrostatic model yielded a good fit to the titration data for the five bacterial species considered, including S. oneidensis. The same model successfully described the P. putida data in this study, and therefore was used for determining the concentration of each bacterial envelope site and the total concentration of binding sites. In order to calculate the sulfhydryl site concentrations, at least three titration experiments were conducted for each biomass with or without qBBr treatment. We used the Student’s t test to determine if the total concentration of binding sites within cell C
DOI: 10.1021/acs.est.6b00347 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 1. SEM images of: (A) untreated, (B) resin-treated, (C) EDTA-treated, and (D) enzyme-treated P. putida cells.
envelopes decreased significantly after qBBr treatment, and a P value greater than 0.05 was taken to indicate no significant difference between the signals from the biomass samples with and without qBBr treatment. In these cases, the concentration of sulfhydryl sites in the biomass was too low to be detectable using our approach.
change in biomass binding site concentration after EPS removal using each of the three methods considered can be attributed to the loss of EPS. 3.2. Sulfhydryl Sites on EPS Molecules of P. putida. Figure 2 depicts representative potentiometric titration curves
3. RESULTS AND DISCUSSION 3.1. Characterization of Bacterial Cells Before and After EPS Removal. We used SEM imaging to qualitatively evaluate the EPS removal efficiency of the three removal approaches. As shown in Figure 1A, untreated P. putida cells are surrounded by a thick layer of EPS coating, confirming that EPS production by P. putida is extensive and that the EPS coating remains extensive even after our biomass washing procedure. In contrast, in the treated biomass samples (Figure 1B−D), only cells are visible with little to no EPS present in any of the samples, demonstrating that essentially all of the EPS materials are removed by each of the three removal methods. After EPS removal, the mass loss of P. putida samples treated by cation exchange resin, EDTA and enzyme was 40 ± 8%, 31 ± 0%, and 28 ± 6%, respectively, while a control test treating the biomass samples with only 0.1 M NaCl, which should remove minimal EPS from the biomass, yielded a mass loss of only 7%. These results confirm that P. putida produced abundant EPS under the growth conditions used in this study, and that all three of the EPS removal methods were effective at removing significant quantities of EPS from the biomass samples. Prior to EPS removal, most of the P. putida cells fluoresced green when exposed to the live/dead stain, with only a limited number of red cells present, indicating that the cell membranes of virtually all of the cells are intact (Supporting Information Figure S1). Similar results were observed for P. putida cells with EPS removed by each of the three methods tested, demonstrating that none of the EPS removal methods induce significant damage to cell membranes. As a result, we conclude that any
Figure 2. Representative titration curves of: (A) untreated, (B) resintreated, (C) EDTA-treated, and (D) enzyme-treated P. putida cells before (Δ) and after (○) qBBr treatment of the biomass to block sulfhydryl sites.
of suspensions with untreated and qBBr-treated P. putida cells in the absence and presence of EPS. Potentiometric titrations yield high precision measurements of the total concentration of proton-active sites exposed to solution, with the total concentration equal to the difference between the highest and lowest values measured for the biomass suspension, minus that D
DOI: 10.1021/acs.est.6b00347 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology for a blank electrolyte. The representative titration curve for the qBBr-treated P. putida suspension with EPS intact is lower in the acidic pH region and higher in the basic region than that for the untreated P. putida suspension in the presence of EPS, indicating that the buffering capacity of qBBr-treated biomass is weaker than that of the untreated biomass, due to the blocking of sulfhydryl sites by the qBBr treatment. In contrast, the qBBr treatment shows nearly negligible effect on the titrations curves for the P. putida suspensions with EPS removed by any of the three approaches, demonstrating a lower sulfhydryl concentration on P. putida without EPS than that with EPS. In order to quantify sulfhydryl site concentrations on P. putida with or without EPS, we conducted the titration experiments for a specific treatment combination at least three times and use surface complexation modeling to calculate the total site concentrations of each sample with and without qBBr treatment. In all cases a 4-site nonelectrostatic surface complexation model40 yields an excellent fit to the titration data, with pKa values for P. putida of 3.6, 5.3, 7.1, and 9.5 (1σ uncertainties ≤0.1, Table S1). Although virtually all of the EPS was removed using any of the three removal approaches employed in the present study, we observed no significant change in the calculated pKa values and the concentrations of individual sites between the P. putida biomass samples with and without EPS (Table S1). Previous studies also noted similar binding capacities for biomass with and without EPS removed using either enzyme treatment19,23 or genetic modification26 for EPS removal. However, some studies17,18,22,27 have reported decreased binding site concentrations after EPS was removed by cation exchange resin treatment. As Frolund et al.25 found that significant cell lysis occurs after 2 h of exposure of biomass to a cationic exchange resin, one possible explanation for the disagreement is that these studies used either overnight or 24 h as a reaction time for the EPS removal treatment, compared to the 2 h that was used in this study. The effect of qBBr treatment on the measured total concentration of proton-active binding sites within P. putida cell envelopes depends strongly on whether EPS is present or not. In the presence of EPS, the Student’s t test indicates that qBBr treatment results in a statistically significant decrease of 35.1 ± 9.5 μmol/g in total site concentration (Table 1, Figure 3), a relatively high concentration of sulfhydryl sites compared to the concentration found on other bacterial species.36 In contrast, after EPS was removed by any of the three methods considered, the qBBr treatment did not affect the calculated total site concentration relative to that obtained for the untreated biomass, a result corroborated by the Student t test (P > 0.57, Table 1). These results strongly suggest that virtually all of the sulfhydryl sites within P. putida cell envelopes are located on the EPS molecules, and that the concentration of sulfhydryl sites in other components of the cell envelope, such as the outer membrane and the peptidoglycan layer, is extremely low. Growth phase and growth temperature both exert strong effects on the concentration of sulfhydryl sites on P. putida, and these observations are consistent with the primary location of the sulfhydryl sites being on the EPS molecules. As shown in Table 2, sulfhydryl sites were not detected using our approach for bacterial cells collected in early exponential phase (6 h of growth), but bacterial cells harvested in early (24 h of growth) and late stationary phase (120 h of growth) exhibited sulfhydryl concentrations of 35.1 ± 9.5 and 37.7 ± 9.4 μmol/g, respectively. After EPS was removed from the bacterial cells
Table 1. Calculated Total Binding Sites within P. putida and S. oneidensis Cell Envelopes in the Presence or Absence of EPS and with or without qBBr Treatmenta bacteria
EPS removalb
P. putida
with EPS
resin
EDTA
enzyme
S.
with EPS oneidensis resin
qBBr treatment
replicates
total binding sites (μmol/g)
6 6
340.2 ± 12.7 305.1 ± 19.4
0.004
3 3
324.8 ± 24.3 322.8 ± 22.4
0.922
3 3
341.9 ± 16.5 331.7 ± 22.9
0.565
3 3
314.2 ± 32.7 311.9 ± 27.9
0.931
5 5
383.5 ± 7.7 350.9 ± 2.7
