|Home | About | Journals | Submit | Contact Us | Français|
Biosorption of metal ions may take place by different passive metal-sequestering processes such as ion exchange, complexation, physical entrapment, and inorganic microprecipitation or by a combination of these. To improve the biosorption capacity of the potential yeast biosorbent, short metal-binding NP peptides (harboring the CXXEE metal fixation motif of the bacterial Pb2+-transporting P1-type ATPases) were efficiently displayed and covalently anchored to the cell wall of Saccharomyces cerevisiae. These were fusions to the carboxyl-terminal part of the sexual adhesion glycoprotein α-agglutinin (AGα1Cp). Compared to yeast cells displaying the anchoring domain only, those having a surface display of NP peptides multiplied their Pb2+ biosorption capacity from solutions containing a 75 to 300 μM concentration of the metal ion up to 5-fold. The S-type Pb2+ biosorption isotherms, plus the presence of electron-dense deposits (with an average size of 80 by 240 nm, observed by transmission electron microscopy) strongly suggested that the improved biosorption potential of NP-displaying cells is due to the onset of microprecipitation of Pb species on the modified cell wall. The power of an improved capacity for Pb biosorption was also retained by the isolated cell walls containing NP peptides. Their Pb2+ biosorption property was insensitive to the presence of a 3-fold molar excess of either Cd2+ or Zn2+. These results suggest that the biosorption mechanism can be specifically upgraded with microprecipitation by the engineering of the biosorbent with an eligible metal-binding peptide.
Once released, nondegradable heavy metal species tend to persist indefinitely in the environment, circulating in the ecosystem and eventually accumulating through the food chain. It is well known that each metal has its tolerated limit above which it becomes toxic or hazardous (8, 25). In the context of increased public awareness (even in developing countries) of the issue of the environmental toxicity of heavy metals, wastewater treatment is of the utmost importance. While most of the remediation methods in current use rely on physico-chemical processes with man-made synthetic materials, the use of microorganisms and plants as low-cost and eco-friendly alternatives of high efficiency is gaining increasing attention. Consequently, various bioremediation concepts are being proposed (4, 18, 23, 29, 34, 38, 41). Among them, the biosorption of metal ions with different types of biomass as biosorbents has proven an ideal bioremediation technology for metal-containing effluents and occupies the position of a “traditional” bioremediation approach, with several attempts at its commercialization (38, 41).
Biosorption of metal ions is a metabolism-independent metal accumulation event that occurs at the cell wall through the action of polysaccharides, associated molecules, and functional groups. It involves mainly ion exchange, chemisorption, adsorption, and, in some cases, also inorganic microprecipitation of certain heavy metal species (29, 38, 41). In the search for strategies to enhance the biosorption capacity for a specific metal ion, the anchoring of particular amino acid sequences to the microbial cell wall has proved to be a promising approach. Surface displays of metal-binding oligopeptides, metallothioneins (MTs), or metalloproteins with the capacity to form coordination centers for the metal ions have been shown to improve the natural metallosorption ability of cells of Escherichia coli, Staphylococcus xylosus, and Staphylococcus carnosus. This approach was successfully extended to other environmentally robust bacteria and yeasts (reviewed in reference 30). For example, the engineering of mouse MT on the cell surface of Cupriavidus metallidurans (formerly Ralstonia eutropha and Ralstonia metallidurans) strain CH34 (7, 29) resulted in the MTB strain, which showed a markedly improved capacity to immobilize Cd2+ in soil and to protect plants from the biological toxicity of the heavy metal (36). However, the power of surface-display-enhanced biosorption of metal ions was demonstrated with intact living microbes, whereas biosorbents for wastewater treatment should preferably be formulated from nonliving biomass (29, 38, 41).
In the present paper we describe the engineering of the carboxyl-terminal part of the sexual adhesion glycoprotein α-agglutinin (AGα1Cp) to anchor the metal fixation motif (CXXEE of the bacterial P1-type ATPases) onto the cell wall of Saccharomyces cerevisiae. In many bacteria, certain P1-type ATPases act as heavy metal ion-specific efflux pumps, protecting the cell interior from metal toxicity (27, 33). A characteristic feature of the metal transporting P1-type ATPases is the presence of one or more heavy metal binding sites at the cytosolic amino-terminal end. Specifically, the CXXEE motif of PbrT of C. metallidurans CH34 is expected to be involved in the fixation of intracellular Pb2+ prior to its export (3, 21). We show that the display of CXXEE on the surface of S. cerevisiae aided the natural Pb2+ biosorption mechanism, with attendant microprecipitation events. Microprecipitation resulted in a substantial increase in the amount of cell-surface-bound Pb. The acquired property was specific for Pb2+ and also remained effective with the isolated cell walls.
The S. cerevisiae strain W303 (MATa leu2-3,112 ura3-1 trp1-1 his3-11,15 ade2-1 can1-100 mal10 GAL SUC2), harboring constructs (Fig. (Fig.1)1) based on the centromeric expression vector p1V5-AG (37), was used for the surface display of fusions to AGα1Cp. The W303 strain was aerobically grown with shaking (200 rpm; 25-mm orbit) in YPD medium (1% [wt/vol] Difco yeast extract, 2% [wt/vol] Difco Bacto peptone, 2% [wt/vol] glucose, 0.003% [wt/vol] adenine hemisulfate) at 30°C. The surface-engineered cells harboring p1V5-AG variants were cultured in URA+ selective dextrose (SD) medium (0.7% [wt/vol] Difco yeast nitrogen base, 2% [wt/vol] glucose, and 0.005% adenine hemisulfate plus l-histidine, l-tryptophan, and l-leucine, each at concentrations of 0.003% [wt/vol]). Yeast cells from cultures at early stationary phase (optical density at 590 nm [OD590] of 3 to 3.5) obtained after 20 h of growth in YPD medium and inoculated with 5% of saturated culture in the same medium were used in all experiments. E. coli TG1 [supE hsdΔ5 thi Δ(lac-proAB) F′ (traD36 proAB+ lacIq lacZΔM15)], used to multiply the plasmids, was grown in Luria broth (31) with 150 μg of ampicillin ml−1.
The DNA manipulations were performed according to standard protocols (31). The lithium acetate method (10) was used to transform host S. cerevisiae. The identity of S. cerevisiae transformants was verified by sequencing of the plasmid DNA, extracted using a Zymoprep kit (Zymo Research Corp.).
Two complementary oligonucleotides (plus strand, 5′-GATCCCTAACTATCTCCAACATGGACTGTCCAACTGAAGAAGCTTTGATCAGAGACAAGTTGGCTGGTA-3′), used to construct fusions of NP peptides (Fig. (Fig.1)1) to the N terminus of V5-AGα1Cp, were designed to produce a double-stranded NP gene fragment flanked at the 5′ and 3′ ends by BamHI and BglII cohesive termini, respectively. This DNA fragment was inserted into the unique BamHI site between the MFα1 and V5 gene sequences of p1V5-AG, giving rise to the plasmid p1AG-NP1 (Fig. (Fig.1).1). Only one BamHI site (at the 5′end of the NP gene) was then reconstituted within the resulting MFα1-NP-V5-AGα1C gene fusion. This allowed for successive insertions of the second and the third NP gene sequences in tandem, resulting in plasmids p1AG-NP2 and p1AG-NP3, respectively (Fig. (Fig.11).
Immunofluorescence microscopy was performed at room temperature without the fixation of yeast cells. Approximately 106 cells from early stationary cultures were washed twice with TBS (20 mM Tris-Cl, pH 7.4, 150 mM NaCl, 3 mM KCl) and incubated for 1 h in 200 μl of TBS with 1% (wt/vol) bovine serum albumin. Next, the anti-V5-fluorescein isothiocyanate (FITC) monoclonal antibody (Invitrogen Corp.) was diluted 1:300 in the same buffer and was applied for 1 h. Cells were washed three times with 500 μl of TBS buffer prior to microscopy. Fluorescence was detected using an Olympus IX81 microscope, equipped with a U-MWIBA/green fluorescent protein (GFP) filter (excitation, 460 to 490 nm; emission, 510 to 550 nm). The images were recorded using a model DP71 cooled digital camera and CellB software (Olympus Corp., Japan).
TEM of unstained cells attached to the carbon-coated grids was conducted using a model JEOL JEM-1010, at magnifications ranging from ×10,000 to ×400,000 at 80 kV.
Cells of S. cerevisiae, grown to the early stationary phase in SD medium, were harvested by centrifugation for 5 min at 4,000 × g and 25°C and then washed with ice-cold isolation buffer (10 mM Tris-HCl, pH 7.8, 1 mM phenylmethylsulfonyl fluoride [PMSF]). The cells were resuspended at a density of 0.25 g (wet weight) per 1 ml of the isolation buffer. Subsequently, 0.5 g of glass beads (0.5 mm diameter) was added for cell disruption in a Mini-Beadbeater device (BioSpec Products Inc.). This was conducted for 3 min at the maximum speed and 0°C. The glass beads were separated from the disintegrated cells by gravity. The supernatant and three successive washes of the beads with 1 ml of isolation buffer were pooled, and the cell walls were recovered by centrifugation for 5 min at 1,000 × g and 4°C.
Cell wall proteins were fractionated as described elsewhere (32, 37). Briefly, the cell walls obtained from 1 g of cells (wet weight) were washed three times with 10 ml of ice-cold 1 mM PMSF. Extraction of the cell walls was done with hot 2% SDS solution under reducing conditions and subsequent incubation with laminarinase (Sigma-Aldrich Corp.). This was followed by treatment of the released cell wall proteins with endoglycosidase H (EndoH; Sigma-Aldrich Corp.). The extracted proteins were resolved by using 8% (wt/vol) acrylamide sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions and then electroblotted on nitrocellulose membranes. These were then blocked with 10% skim milk in TBS at 4°C for 16 h. The monoclonal anti-V5 antibody was conjugated with horseradish peroxidase (anti-V5-HRP; Invitrogen Corp.). This was applied at a 1:5,000 dilution in TBS-0.1% Tween 20 (TBST) with 2% skim milk for 2 h. Membranes were washed with TBST and developed using a West Femto detection kit (Pierce Biotechnology, Inc.). The signal was scanned using an LAS 1000 luminescent image analyzer (Fuji Photo Film Co., Japan).
The nitrogen content of isolated cell walls was determined in triplicate in 2 mg of freeze-dried material by using an Elementar vario EL III analyzer (Elementar Americas, Inc.); 4-amino-benzenesulfonic acid was used as a standard.
Fresh cells, harvested from S. cerevisiae cultures at an early stationary phase by centrifugation (at 4,000 × g for 5 min at 25°C), or the isolated cell walls (see above) were washed twice with 25 mM morpholineethanesulfonic acid (MES; pH 5.5). The cells were resuspended to a final density of approximately 1.5 mg of dry cell wt ml−1 (OD590 of 2.5) and then with the cell wall material to a corresponding density of 0.45 mg of dry weight ml−1 of the same buffer. Metal binding at pH 3 and 4 was performed in 25 mM MES-acetate, at pH 5 in 25 mM MES, and at pH 6 and 7 in 25 mM HEPES. The desired metal ions were added as Pb(NO3)2, Cd(NO3)2, and ZnCl2 (Sigma-Aldrich Corp.) to the intended final concentrations. No precipitation of heavy metals in the test solutions was detected prior to the addition of the biosorbent materials. Following agitation at 130 rpm and 25°C for the intended period of time, the biosorbent materials were separated by centrifugation for 2 min at 4,000 × g and 25°C. The metal content in the supernatant was determined by atomic absorption spectrometry ([AAS] model Spectr AA300; Varian, Inc.). The separated biosorbent was washed with 25 mM HEPES (pH 6.5), and its metal content was determined by AAS after digestion with 70% nitric acid overnight under atmospheric pressure and at room temperature. Alternatively, washed cells were incubated for 15 min with 1/10 of the sorption solution volume of 10 mM EDTA (pH 7.0) in order to remove the surface-bound metal. The cells were then pelleted and treated as described above. Dry weights of the biosorbent were determined by drying the batch biosorption assay aliquots at 80°C.
The NP peptide, harboring a putative metal fixation motif CXXEE, was designed such that the conserved sequence MDCPTEEALIR, along with the flanking residues, should mimic corresponding secondary structures within the P1-type ATPases PbrT and CadA (Fig. (Fig.1).1). The secondary structure patterns were predicted using a hierarchical neural network prediction method (6) (data not shown). To display up to three tandem NP peptide sequences on the cell wall of S. cerevisiae W303, the DNA fragments encoding NP were successively engineered to the MFα1-V5-AGα1C gene of the centromeric expression vector p1V5-AG (Fig. (Fig.1).1). This p416GPD-based (22) vector was obtained by fusion of the sequence encoding the leader peptide of α-factor precursor (MFα1) for surface targeting, the V5 epitope tag for immunochemical detection, and the C-terminal anchoring domain AGα1Cp of α-agglutinin (37). The expression vectors for the surface display of one, two, and three NP peptides were named p1AG-NP1, p1AG-NP2, and p1AG-NP3, respectively (Fig. (Fig.1).1). We previously noted that use of centromeric constructions allows uniform expression of fusions to AGα1Cp in all culture cells (37). When analogous 2μm vectors based on p426GPD (22) were used, corresponding fusions were displayed by only 20 to 75% of the culture cells (unpublished data). A surface display efficiency of 25.1% with a 2μm-based vector was also observed by Kuroda et al. (13), who attributed this phenomenon to mitotic instability of the vector.
Immunofluorescence analysis showed that virtually all cells of S. cerevisiae expressing the MFα1-NPn-V5-AGα1C gene fusion (n represents the number of NP repeats) showed bright fluorescent labeling with the anti-V5-FITC antibody, confirming the constitutive expression and surface display of the corresponding proteins (Fig. (Fig.2A).2A). Expression had no adverse effect on culture growth, as determined in URA+ selective SD medium and by comparison with the transformant harboring p416GPD.
To investigate the covalent attachment of NPn-V5-AGα1Cp to the cell surface, the cell wall proteins were fractionated and examined by immunoblotting. The noncovalently bound proteins were released from isolated cell wall materials by hot SDS treatment, and the covalently bound proteins were subsequently liberated by hydrolyzing the cell walls with endo-1,3(4)-β-glucanase laminarinase. The released proteins were then treated with N-acetylglucosaminidase H (EndoH) to reduce size heterogeneity, due to their N-linked polysaccharides. As shown in Fig. Fig.2B,2B, substantial portions of the NPn-V5-AGα1Cp and V5-AGα1Cp fusions were found in the EndoH-treated laminarinase extracts, thereby confirming the covalent modification of the cell wall glucan. The immunoblot analysis further revealed that NP peptide extensions had no negative effect on the anchoring of V5-AGα1Cp variants to the cell wall. The 100-kDa band corresponding to the unbound V5-AGα1Cp variants, entrapped in the glucan network, was detected in all SDS extracts (Fig. (Fig.2B).2B). A significant increase in the apparent molecular weights of the extracted fusion proteins could be attributed to glycosylations of V5-AGα1Cp variants with predicted apoprotein sizes of 43.1 to 35.3 kDa. Such modifications involve O glycosylations at abundant Ser and Thr residues, a remnant of the cell wall β(1→6)-glucan, and potential EndoH-resistant N glycosylations (17, 32).
To estimate the number of displayed V5-AGα1Cp fusions, we determined the nitrogen content of cell walls by elemental analysis. While cell walls of wild-type S. cerevisiae W303 contained 2.41% ± 0.02% N, its content was found to be increased to 3.40% ± 0.03% in those walls containing V5-AGα1Cp and to 3.32% ± 0.02% in those walls with NP1-V5-AGα1Cp or NP3-V5-AGα1Cp. Assuming that the increase in N content comes from the proteinaceous part of V5-AGα1Cp variants, the number of protein molecules per cell was calculated as 4.1 × 106 for V5-AGα1Cp and as 3.8 × 106 for NPn-V5-AGα1Cp. Similar surface density (1.6 ×106 copies per cell) was reported as natural for the protein CWP2 of S. cerevisiae (9). However, other changes in cell wall composition due to the presence of V5-AGα1Cp variants (e.g., increased chitin deposition to the cell wall) could not be excluded.
The effect of the NP display on Cd2+, Zn2+, and Pb2+ binding capacity was initially tested using yeast cells expressing NP1-, NP2-, NP3-V5-AGα1Cp and V5-AGα1Cp grown to the early stationary phase. The cells (1.5 mg of cell dry weight ml−1) were exposed to 150 μM metal solutions (i.e., to a 400-fold excess of metal ions over the calculated number of displays in solution) at pH 5.5 and 7.0 for 4 h. While the display NP peptides had no effect on the sequestration of Cd2+ and Zn2+, they enhanced the capability of the cells to bind Pb2+. The amounts of accumulated Pb2+ were the same at pH 5.5 and 7.0 but were reduced by 8% at pH 5.0 and by more than 75% at pH 4 and 3 (data not shown). The detailed biosorption isotherms were thus evaluated at pH 5.5 also to eliminate formation of PbOH+, which may occur at pH 7 at higher Pb concentrations. Compared to the control (displaying merely an anchoring domain), the cells displaying the NP peptides showed a sharp increase in their Pb2+-binding capacity (Fig. (Fig.3A).3A). This was most pronounced when initial concentrations of Pb2+ were 75 to 300 μM. The acquired property was essentially independent of the number of displayed NP peptide repeats per fusion protein (Fig. (Fig.3A3A shows data concerning the display of NP1 and NP3). The promoted metal sequestration capacity appeared solely to be due to a metabolism-independent biosorption of Pb2+ on the cell surface, as virtually all accumulated metal was released by washing with 10 mM EDTA from the cells pretreated for 4 h with 50 and 150 μM Pb2+ (Fig. (Fig.3B3B).
To obtain insight into the kinetics of the Pb2+ binding process, additional biosorption isotherms with the cells expressing NP1-V5-AGα1Cp, NP3-V5-AGα1Cp, and V5-AGα1Cp were evaluated under the same conditions over contact time periods of 45 min, 2 h, and 8 h. While only a slight contribution of the displayed NP peptides to biosorption of Pb2+ occurred during 45 min (Fig. (Fig.4A),4A), a gradual increase in Pb2+ accumulation was observed with the cells displaying NP peptides after 2 h and 8 h of contact time (Fig. (Fig.3A3A and 4B and C). The most pronounced contribution of displayed NP peptides was observed at initial Pb2+ concentrations of 100 to 150 μM. At these concentrations the biosorption capacity of the modified cells increased by factors of 3.5 to 5.2, compared to the V5-AGα1Cp-displaying control. The NP-displaying cells removed 90 and 95% of the metal after 4 and 8 h of contact time, respectively.
To investigate whether the yeast cell walls, indeed, gained the increased biosorption capacity due to the NP display, they were incubated in Pb2+-containing solutions at pH 5.5 for 2 and 4 h. The cell wall density used corresponded to the cell wall mass proportion determined in the experiments with the intact cells. The cell walls enriched with NP peptides sequestered during 2 and 4 h of contact time up to 2.5- and 3-fold more Pb2+, respectively, compared to the control walls containing V5-AGα1Cp protein alone (Fig. (Fig.5).5). The biosorption capacity of the modified walls was again independent of the number of NP repeats per fusion and was highly effective at the initial Pb2+ concentrations of 100 to 150 μM, when 90 to 95% of the metal was immobilized.
A sharp increase in the amount of cell wall-associated metal, which occurs after a certain threshold concentration of metal ions in solution is exceeded, often signalizes the precipitation of metal species on the surface of the biosorbent (1, 20). Thus, transmission electron microscopy (TEM) was employed to check for the presence of precipitated Pb species within the cell walls (Fig. (Fig.6).6). While there was no precipitate detected in the controls (Fig. 6B and D), electron-dense particles with an average size of 80 by 240 nm were observed on the cells displaying NP peptides and treated with 150 μM Pb2+ for 4 h (Fig. 6A and C).
As indicated above, we did not observe any contribution of NP peptides to the Cd2+ and Zn2+ binding capacity of S. cerevisiae cells incubated in the presence of 150 μM concentrations of the respective metal ion. This prompted us to investigate the specificity of NP display for the biosorption of Pb2+ in two metal systems. Therefore, the isolated cell walls were incubated with 100 μM Pb2+ in the presence of 300 μM Cd2+ or Zn2+. This experiment showed that none of the competing metal ions affected the biosorption of Pb2+ (Fig. (Fig.7).7). In contrast, the biosorption of Cd2+ and Zn2+ was reduced by 40 to 50% compared with the biosorption data obtained with the single-metal systems. The isolated cell walls containing NP peptides also showed a slightly improved capacity for biosorption of Zn2+ and Cd2+ from both the single and the double metal solutions (Fig. (Fig.7).7). The same metal binding properties were observed with the intact cells treated under identical conditions except that the NP peptides did not contribute to biosorption of Zn2+ and Cd2+ (not shown).
To test whether competing metal chelating molecules may inhibit biosorption by modified cell walls, we studied the effects of some commonly known Pb2+ chelators. Specifically, EDTA, citrate, and glutathione were added in biosorption experiments with 150 μM Pb2+. The effects of these competitors were generally independent of the presence of NP peptides. EDTA exerted complete inhibition when added at 150 μM (data not shown). While glutathione showed only a slight inhibition, 5 mM citrate reduced the amount of biosorbed Pb by 75 to 87% (Fig. (Fig.7B7B).
The covalent attachment of NP peptides harboring the CXXEE metal fixation motif of the bacterial metal transporting P1-type ATPases on the surface of S. cerevisiae (Fig. (Fig.2)2) was expected to promote the biosorption of metal ions on the modified yeast cell wall. The rationale behind the choice of the CXXEE motif, rather than the (designed) high-affinity metal-binding center, was its expected propensity to exchange the metal ion, an event which should occur during metal export by the transporter ATPase. It has been well documented that displays of metal-binding peptides enhance biosorption of metal ions on the bacterial cell walls (reviewed in reference 30). This effect was due to the capacity of the displayed peptides to bind metal ions, as well as to their competence to exchange the metal ion with the cell wall structures, which otherwise cannot react with the metal ion in solution (12, 35, 36). Display of the hexahistidine or yeast MT has also been reported to multiply the natural biosorption capacity of S. cerevisiae for Cd2+ and Cu2+ (14, 15, 16). However, the display of the peptide GHHPHG, named HP peptide, increased the amount of adsorbed Zn2+ by only 20% (37), which was consistent with the HP/Zn2+ stoichiometry of 1:1 (with an apparent dissociation constant of 120 nM). Such properties of the HP peptide indicate that the display of a high-affinity peptide does not support “funneling” of the metal ion to the natural metal-binding sites of the S. cerevisiae cell wall.
The CXXEE motifs naturally localize at the cytosolic amino-terminal end of the PbrA transporter of the Pb2+-specific resistance system (3) and also of the Pb2+- and Cd2+-transporting CadA of C. metallidurans CH34 (21). It occurs within the conserved sequence MDCPTEEALIR (Fig. (Fig.1).1). When this sequence was displayed on the surface of S. cerevisiae as a part of the NP peptide, it markedly promoted the biosorption of Pb2+ (Fig. (Fig.3A3A and and4).4). An improved capacity to sequester Pb2+ from solution was also retained by the isolated NP-containing cell walls of S. cerevisiae (Fig. (Fig.5).5). It confirmed that the functionality of NP peptide is independent of intracellular metabolism and that the accumulation of Pb2+ can be solely attributed to the biosorption process. Here, it should be noted that we had previously observed a substantial reduction in the contribution of displayed peptides to the biosorption capacities of engineered cell walls of E. coli and S. cerevisiae compared to the biosorption capacities of intact cells (12, 37), demonstrating that the fully native cell wall architecture played a critical role in these particular cases.
The cell wall of S. cerevisiae is a multilaminate, microfibrillar structure composed mainly of β(1→3)-d-glucan, β(1→6)-d-glucan, chitin, (phospho)mannoproteins, and a minor proportion of lipids and pigments (17). Functional groups such as carboxylate, phosphate, hydroxyl, amine, sulfhydryl, and imidazole contained in these biomolecules participate in sequestering of the metal ions via bonds of either an ionic or covalent nature (2, 40, 41). The biosorption mechanism of heavy metal ions involving ion exchange and complexation is characterized by L-type biosorption isotherms with an initially higher slope that decreases with an increase in the equilibrium metal concentration (1, 11, 38). The L-type Pb2+-biosorption isotherms were obtained for S. cerevisiae, producing the anchoring domain only (Fig. (Fig.3A,3A, ,4,4, and and5).5). The biosorption isotherms observed for the cell walls engineered with NP peptides (Fig. (Fig.3A,3A, 4B and C, and and5)5) can be regarded as the S type. The S-type isotherms are generally characterized by a lower initial slope, which sharply increases as the equilibrium concentration increases (1, 11, 38). Such biosorption isotherms are indicative of the inorganic microprecipitation of metals on biosorbent, initiated by heterogeneous nucleation (11, 20). Nucleation and microprecipitation of metal hydroxides or phosphates may result from locally increased concentration of surface-bound metal ion and increased pH or presence of biogenic phosphates (11, 20, 24, 38). Given the S shape of biosorption isotherms (Fig. (Fig.3A,3A, 4B and C, and and5),5), as well as the presence of electron dense particles (Fig. 6A and C), we deduce that the microprecipitation of Pb species is responsible for the enhanced biosorption property of the NP-peptide-containing cell walls. However, the chemical form of the microprecipitate remains to be investigated. A microprecipitation phenomenon had been described in S. cerevisiae only for uranium biosorption (39). The microprecipitation of Pb on a natural biomass was observed with the alga Sargassum vulgaris, in which the metallic Pb0 became deposited (26). Microprecipitation of Pb, presumably as hydroxides and phosphates, was observed with Rhizopus arrhizus (24). The metal-based titration of the R. arrhizus biomass at pH 5 allowed Naja et al. (24) to show that the formation of insoluble Pb species occurred after the initial binding of Pb2+ via ion exchange at a lower saturation of the biomass with the metal. Microprecipitation became evident at a higher saturation when the equilibrium Pb2+ concentrations in solution exceeded micromolar levels.
It appears reasonable to assume that the contribution of the NP peptide involves a gradual increase in the local Pb2+ concentration to a certain level that triggers the microprecipitation. Intriguingly, the overall biosorption capacity of modified cells was essentially independent of the number of NP peptide repeats displayed per fusion protein (Fig. (Fig.3A,3A, ,4,4, and and5).5). In our model we assume that the single CXXEE motif has sufficient power to promote formation of the nucleation sites for microprecipitation. However, it remains unclear why the effect of the display of NP peptides is lost in the intact cells when the initial Pb2+ concentrations exceed 300 μM (Fig. (Fig.3A3A and and4),4), whereas the isolated NP-engineered cell walls remained effective even at 500 μM Pb2+ (Fig. (Fig.55).
A number of biosorption studies have shown that the Pb2+ biosorption capacity of the natural biomass of S. cerevisiae exceeds that for Cd2+ and Zn2+ by a factor of 4 (5, 40), making this yeast a suitable candidate for a biosorbent to be utilized in the Pb2+ remediation process. Current efforts in biosorption research are mainly dedicated to algal biosorbents, which show both a high efficiency at low metal concentrations and high biosorption capacities (38, 41). The best performing are biosorbents made of brown marine algae, which show average maximum biosorption capacities of 1,240 nmol of Pb mg−1 at high equilibrium metal concentrations of about 1 mM (40). The maximum biosorption capacity of NP-containing cell walls of S. cerevisiae was apparently 300 nmol of Pb mg−1 at an equilibrium Pb2+ concentration of 350 μM. Based upon data obtained by Romero et al. (28) with representative algal biosorbents and under similar experimental conditions (biosorbent density of 0.5 mg ml−1, pH 5.0), the biosorption capacity at an equilibrium concentration of 350 μM Pb2+ (given in parentheses as nmol mg−1) decreased in the following order: Fucus spiralis (885; brown algae), Ascophyllum nodosum (747; brown algae), Chondrus crispus (410; red algae), modified S. cerevisiae (291; NP1-containing walls), Codium vermilara (267; green algae), Spirogira insignis (244; green algae), Asparagpsis armata (235; red algae), and S. cerevisiae (130; cell walls with AGα1Cp only). However, the contribution of NP peptides was most pronounced at lower Pb2+ concentrations (Fig. (Fig.5).5). Indeed, the order of the biosorption capacity at equilibrium of 9 μM Pb2+ changes in favor of NP-containing cell walls (in descending order): NP1-modified S. cerevisiae walls (193), F. spiralis (189), S. insignis (128), A. nodosum (125), C. vermilara (52.1), S. cerevisiae walls with AGα1Cp only (36.8), A. armata (23.4), and C. crispus (17.9).
In summary, we have shown that the Pb2+ biosorption capacity of S. cerevisiae can be substantially improved by the surface display of the specific metal-binding peptide NP. The contribution of NP peptides was not impaired by an excess of potentially competing Cd2+ and Zn2+ ions (Fig. (Fig.7A).7A). Moreover, virtually all of the Pb can be recovered from the cell surface by EDTA treatment (Fig. (Fig.3B),3B), and the contribution of the NP peptides also remained effective in the isolated cell walls. Such properties of modified cell walls could be of practical importance since the biosorbents are based on nonliving biomass. The feasibility of using a biosorption column packed with silica-immobilized cell walls of S. cerevisiae has already been demonstrated (19).
We thank Pavel Ulbrich for his generous help with TEM microscopy and Nick J. Russell for a careful reading of the manuscript.
This work was funded by research projects 1M6837805002 and MSM 6046137305 supported by Czech Ministry of Education.
Published ahead of print on 19 February 2010.