1160 Ewelina Dziubakiewicz1,2 Bogusław Buszewski1,2 1 Chair
of Environmental Chemistry and Bioanalytics, Faculty of Chemistry, Nicolaus ´ Copernicus University, Torun, Poland 2 Centre for Modern Interdisciplinary Technologies, Nicolaus Copernicus University, ´ Poland Torun,
Received August 22, 2013 Revised December 4, 2013 Accepted December 9, 2013
Electrophoresis 2014, 35, 1160–1164
Research Article
Capillary electrophoresis of microbial aggregates Uncontrolled aggregation of bacterial cells is a significant disadvantage of electrophoretic separations. Various aspects of the electrophoretic behavior of different strains of Grampositive Bacillus cereus, Bacillus subtilis, Sarcina lutea, Staphylococcus aureus(1), and Micrococcus luteus bacteria and Gram-negative Escherichia coli bacteria were investigated in this study. Our findings indicate that bacteria can be rapidly analyzed by CZE with surface charge modification by calcium ions (Ca2+ ). Bound Ca2+ ions increase zeta potential to more than 2.0 mV and significantly reduce repulsive forces. Under the above conditions, bacterial cells create compact aggregates, and fewer high-intensity signals are observed in electropherograms. The above can be attributed to the bridging effect of Ca2+ between bacterial cells. CE was performed to analyze bacterial aggregates in an isotachophoretic mode. A single peak was observed in the electropherogram. Keywords: Bacterial aggregation / Capillary electrophoresis / Modified cell surface charge / Zeta potential DOI 10.1002/elps.201300588
Additional supporting information may be found in the online version of this article at the publisher’s web-site
1 Introduction CZE supports fast detection and identification of microorganisms, and it has a variety of practical applications in bioengineering, medical diagnosis, environmental protection, and food analysis [1–5]. Despite its popularity, the method has certain limitations, such as the aggregation and/or adhesion of bacteria to the surface of solids [6–14]. Every bacterial species has a complex and characteristic composition of the cell wall. Macromolecules that are present in the cell wall and bacterial membranes, including proteins, phospholipids, teichoic acid, teichuronic acid, and lipopolysaccharides, produce unique biochemical fingerprints. Those macromolecules also contribute to the surface charge of bacterial cells through the ionization of protonactive functional groups, such as carboxyl, phosphate, amino, and hydroxyl groups, and the adsorption of ions from the solution [15, 16]. Research studies have demonstrated that the discussed macromolecules play a significant role in bacterial aggregation and adhesion to solid and contribute to microbial differentiation [17, 18].
Correspondence: Professor Bogusław Buszewski, Department of Environmental Chemistry and Bioanalytics, Faculty of Chemistry, ´ Poland Gagarina 7, 87-100 Torun, E-mail: [email protected] Fax: +48-56-6114837
Abbreviation: PEO, poly(ethylene oxide) C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Bacteria may be divided into two major groups based on the Gram-staining reaction. Peptidoglycan is a significant component of the cell wall and is found in both bacterial groups. The cell wall of Gram-positive bacteria is composed of a rigid and relatively thick peptidoglycan layer with adjacent teichoic and teichuronic acids [19, 20]. In more complex envelopes of Gram-negative cells, peptidoglycan is found only in a thin layer between two membranes—the inner cytoplasmic membrane and outer membrane. The outer membrane consists of proteins, phospholipids, and lipopolysaccharides [21–23]. Bacterial adhesion was first explained by the DLVO (Derjaguin, Landau, Verwey, Overbeek) theory. Bacterial adhesion is mediated by an interplay between Lifshitz-van der Waals forces and electrostatic interactions originating from the overlap of electrical double layers and the solid surface. This theory has been extended to acid–base interactions [15, 24]. The leading theories explaining aggregation and adhesion effects during electrophoretic analysis were proposed by Armstrong and co-workers [12, 13], Zheng et al. [6], and Buszewski and co-workers [7, 10, 11] who characterized the electrophoretic behavior of bacteria based on analyses of electropherograms, zeta potential measurements, and microscopic observations. Despite those efforts, the described phenomena still lack a detailed theoretical explanation. The objective of this study was to investigate various aspects of the electrophoretic behavior of different strains of Gram-positive Bacillus cereus, Bacillus subtilis, Sarcina lutea, Staphylococcus aureus(1), and Micrococcus luteus bacteria, and Gram-negative Escherichia coli bacteria and to demonstrate www.electrophoresis-journal.com
CE and CEC
Electrophoresis 2014, 35, 1160–1164
that bacteria can be rapidly analyzed by CE with surface of bacteria charge modification by calcium ions (Ca2+ ) in an isotachophoretic mode.
1161
Between runs, the capillaries were washed with 1.0 M NaOH, deionized water for 2 min each, and running buffer for 4 min. A total of 0.5 mLstock bacterial suspension was used for electrophoretic measurements.
2 Materials and methods 3 Results and discussion 2.1 Preparation of bacterial suspensions Escherichia coli , B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacterial strains were obtained from the collection of the Department of Microbiology (Nicolaus Copernicus University, Poland). Bacterial cells were cultivated according to Dziubakiewicz et al. [25]. The final pellet of washed bacterial cells was suspended in 5 mL of 0.005 M NaNO3 and used for zeta potential measurements. The same procedure was applied to cultivate bacterial suspension for CE analysis, but the pellet with bacterial cells was washed and the final pellet of washed bacteria was suspended in a running buffer. The final pellet of washed bacterial cells was suspended in 5 mL of 0.005 M Ca(NO3 )2 to modify surface charges, and after 6 h, the pellet with bacterial cells was washed to remove free Ca2+ ions for future analysis. The OD (OD measured spectrophotometrically at 590 nm) of final bacterial suspensions varied between 0.097 and 0.320 (109 –1012 cells per mL).
Figure 1 shows the results of CZE analysis of E. coli, B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria in fused silica capillaries in TBE buffer. The obtained electropherograms indicate that the use of the TBE buffer in electrophoretic analyses of the examined bacteria is not effective. Similar results were reported for other buffers, including Tris and MES. Bacteria migrated close to electroosmotic flow (tEOF 1.02 min). The above conditions did not support a reproducible electrophoretic analysis of the examined bacteria. Bacterial adhesion to the inner surface of the capillary and the formation of bacterial aggregates with different size and surface charge, which produced several signals in the electropherograms (Fig. 1), were observed. The inner wall of the capillary was modified both dynamically and chemically to minimize surface adhesion and
2.2 Zeta potential The zeta potential of E. coli, B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacterial strains was measured in 0.005 M NaNO3 at 25°C using the Malvern Zetasizer Nano ZS system (Malvern Instrument, UK) equipped with a titrator (Malvern Instrument) and the Malvern polystyrene U-shaped cell. The suspension was acidified to pH ࣈ 2.0 with 0.1 M HNO3 (POCh, Gliwice, Poland) and was titrated to pH ࣈ 10.5 with 0.1 M NaOH (POCh, Gliwice). All titrations were performed using a titrator assembly consisting of a glass vessel with a lid. The electrode was calibrated over the entire pH range of 2.0–11.0. Zeta potential ( ) was calculated from electrophoretic mobility () with the use of Smoluchowski’s equation: =
, ε0 εr
(1)
where is the viscosity of the electrolyte solution, and ε r and ε o represent the relative permittivity of the electrolyte solution and vacuum, respectively. 2.3 Capillary electrophoresis CE experiments were performed in a HP3D CE system (Agilent Technologies, Waldbronn, Germany) equipped with a DAD with the use of fused silica capillaries (id = 75 m; Ltot = 33.5 cm; Leff = 25 cm; Composite Metal Services, Shipley, UK). New capillaries were rinsed with 1.0 M NaOH, deionized water, and BGE (outlet buffer) for 10 min each before use. C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. Electropherograms of E. coli, B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria. Conditions: TBE buffer, Leff = 25 cm, Ltot = 33.5 cm, detection at = 214 nm, V = 20 kV, injection – 100 mbar·s
www.electrophoresis-journal.com
1162
E. Dziubakiewicz and B. Buszewski
Electrophoresis 2014, 35, 1160–1164
Figure 2. The effect of calcium ions on the zeta potential of E. coli (A), B. cereus (B), B. subtilis (C), S. lutea (D), S. aureus(1) (E), and M. luteus (F) bacteria
slow down EOF. Dynamic modification involved the addition of poly(ethylene oxide) (PEO) to the buffer that increased buffer viscosity and decreased EOF. According to some authors [5,6,12], PEO contributes to the aggregation of bacterial cells. EOF was decreased, but reproducible analyses were not achieved, probably due to system instability. Casual interactions between PEO and bacterial cells, and the inner surface of the capillary (different film thickness) further deepened system instability. The use of capillaries modified with acrylamide eliminated EOF and enabled bacterial cells to migrate according to their own electrophoretic mobility. Despite the above, several signals were observed in the electropherograms (data not shown). The resulting curves of zeta potential of the examined bacteria as a function of pH (Fig. 2) indicate that the observed phenomenon can be attributed to the high content of peptides and polysaccharides in bacterial walls and membranes, which are responsible for uncontrolled aggregation and adhesion to the hydrophilic capillary surface. Gram-positive bacteria, excluding B. cereus, have the pI of ࣘ 2.2. Rijnaarts et al. [26] observed the presence of teichoic acids (R−HPO4 − /R−PO4 2− ) and anionic polysaccharides on the surface of bacterial cells with pKa ࣘ 2.8 (R−HPO4 − /R−PO4 2− ; R−COOH/ R−COO− ). The pI ࣙ 3.6 for B. cereus and E. coli bacteria cannot be unambiguously interpreted. All components of the bacterial wall and membrane should be taken into account. The pI can be determined by pI values of surface peptides and peptidoglycans (R−NH3 + /R−NH2 ; R−COOH/ R−COO− ), polymers, including amino groups (R−NH3 + /R−NH2 ) and polysaccharides (R−HPO4 − /R−PO4 2− ; R−COOH/ R−COO− ). The pI can also vary within a wide spectrum of values, subject to
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
differences in peptidoglycan composition in various bacterial species. The ratio of amino groups to carboxyl groups in peptidoglycan is minimum 1:3, which produces pI ⬎ 3.8. The interactions between amino and carboxyl groups can decrease pI values, whereas the amidation of carboxyl group can lead to an increase in pI. The pI of ࣙ 3.6 for B. cereus and E. coli bacteria suggests the presence of an additional protein layer. According to previous research, B. cereus, Bacillus antracis, and Bacillus brevis are surrounded by an S-layer composed of protein [27] or a mucous coating, which consists of extracellular polymeric substances (26–40% protein) [28, 29]. In E. coli, this phenomenon can be explained by the presence of lipoproteins, porins, and large numbers of protein appendages (flagella, fimbriae, pili) on the surface of the outer membrane. The modification of the surface charge of bacterial cells with Ca2+ facilitates cell aggregation and improves the selectivity of electrophoretic analysis. As demonstrated by Fig. 2, E. coli, B.cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria are charged negatively at pH ⬎ 4. Attractive electrostatic forces (van der Waals and acid–base forces) have to dominate repulsive forces for bacterial cells to aggregate. Deprotonated functional groups on the surface of bacterial cells support the formation of cationic bridges. Calcium ions bound to the surface of cells improve their hydrophobicity, and they evoke the participation of attractive acid–base forces. Calcium ions facilitate overcoming of the electrostatic energy barrier between cells, which in turn facilitates cell aggregation. Extracellular structures, including fimbriae, which bind Ca2+ ions at terminal sites also overcome repulsive forces. As demonstrated by Fig. 2, the modified surface charge of bacteria with calcium ions leads to changes in their
www.electrophoresis-journal.com
CE and CEC
Electrophoresis 2014, 35, 1160–1164
1163
Figure 3. Electropherograms of E. coli bacteria surface modification with Ca2+ (A) and without Ca2+ and application of CE in an isotachoretic mode (B). Conditions: Leff = 25 cm, Ltot = 33.5 cm, detection at = 214 nm, V = 15 kV, injection – 100 mbar·s
Table 1. Buffer compositions
Buffer composition
TRIS Boric acid HCl
Leading buffer
Terminating buffer
c [mM]
pH
c [mM]
pH
5.00 50 4.32
7.11
5.00 50 0
8.23
electrophoretic mobility. Bound Ca2+ ions increase zeta potential to more than 2.0 mV that, according to Bos et al. [30], significantly reduces repulsive forces. Under the above conditions, bacterial cells create compact aggregates [31–33], and fewer high-intensity signals are observed in electropherograms (Fig. 3A). Application of buffers with different ions mobilities (an isotachophoretic mode) and without bacterial surface modification with Ca2+ , according to Oukacine et al. [34,35] and Phung et al. [36] allowed focusing the zones. However, in our case we did not obtain single peak but multiplet (Fig. 3B). This phenomenon is probably due to high concentrations of bacterial cells (109 –1012 cells/mL). An optimal composition of buffers (leading buffer and terminating buffer) is presented in Table 1. Finally, CE in an isotachophoretic mode was performed to focus the migrating bacterial aggregates (after surface modifications with Ca2+ ) in a single zone, and as the result, a single signal was obtained in the electropherograms (Fig. 4). Sharpening of the peaks and an improvement in the shape of the basic line were observed. The high reproducibility of electrophoretic evaluations supported quantitative analyses of bacteria. The applied electrophoretic conditions resulted in good linearity. The correlation coefficient for the examined bacteria was determined at 0.98 (an example of quantification of E. coli bacteria is inserted in the Supporting Information Fig. 1). Application of developed technique did not allow for separation of bacteria and we observed their coelution. From this reason, this technique could be valuable and useful as a sterility test as previously described and published by Tong et al. [37]. C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 4. Electropherograms of E. coli (A), B. cereus (B), B. subtilis (C), S. lutea (D), S. aureus(1) (E), and M. luteus (F) bacteria. Conditions: buffer (Table 1), Leff = 25 cm, Ltot = 33.5 cm, detection at = 214 nm, V = 15 kV, injection – 100 mbar·s
4 Concluding remarks Various aspects of the electrophoretic behavior of different strains of Gram-positive B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria, and Gram-negative E. coli bacteria were investigated in this study. The following conclusions can be formulated based on our findings: (i) The modification of surface bacterial groups with Ca2+ ions contributed to the aggregation of bacterial cells and www.electrophoresis-journal.com
1164
E. Dziubakiewicz and B. Buszewski
decreased the number of signals in the electropherograms. (ii) CE in an isotachoretic mode was performed to focus bacterial aggregates in a single zone, and it supported quantitative analyses of bacterial cells (R2 = 0.98). (iii) The results of this study contribute to our knowledge of electrophoretic separation bacteria and supply new information about the surface charge of bacterial cells. Our findings proved to be useful for describing the fundamental mechanisms of bacterial aggregation.
Electrophoresis 2014, 35, 1160–1164
Wei, W., Zheng, J., Yeung, E. S., Anal. Chem. 2002, 74, 5523–5530. [14] Desai, M. J., Armstrong, D. W., Microbiol. Mol. Biol. Rev. 2003, 67, 38–51. [15] Poortinga, A. T., Bos, R., Norde, W., Busscher, H. J., Surf. Sci. Rep. 2002, 47, 1–32. [16] Beveridge, T. J., Murray, R. G., J. Bacteriol. 1980, 141, 876–887. [17] Tsuneda, S., Aikawa, H., Hayashi, H., Hirata, A., J. Colloid Interface Sci. 2004, 279, 410–417. [18] Eboigbodin, K. E., Newton, J. R. A., Routh, A. F., Biggs, C. A., Appl. Microbiol. Biotechnol. 2006, 73, 669–675.
This study was supported by grant no. N N204 369040 from the National Science Centre in Warsaw, Poland, and a grant from the European Social Fund, Polish National Budget, and budget of the Kujawsko–Pomorskie Region as part of the “Krok w przyszło´sc´” (Step into the Future) Sectoral Operational Program—Human Resources. The authors have declared no conflict of interest.
[19] Doyle, R. J., Matthews, T. H., Streips, U. N., J. Bacteriol. 1980, 143, 471–480. [20] van der Wal, A., Norde, W., Zehnder, A. J. B., Lyklema, J., Colloids Surf. B Biointerfaces 1997, 9, 81–100. [21] Hammond, S. M., Lambert, P. A., Rycroft, A. N., The Bacterial Cell Surface, Croom Helmm, London 1984. ˙ [22] Kunicki-Goldfinger, W. J. H., Zycie bakterii, PWN, Warsow 1998. [23] Beveridge, T. J., Davies, J. A., J. Bacteriol. 1983, 846–858.
5 References [1] Armstrong, D. W., Schulte, G., Schneiderheinze, J. M., Westenberg, D. J., Anal. Chem. 1999, 71, 5465–5469. [2] Lim, O., Suntornsuk, W., Suntornsuk, L., J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2009, 877, 710–718. [3] Gao, P., Xu, G., Shi, X., Yuan, K., Tian, J., Electrophoresis 2006, 27, 1784–1789. [4] Armstrong, D. W., Schneiderheinze, J. M., Kullman, J. P., He, L., FEMS Microbiol. Lett. 2001, 194, 33–37. [5] Armstrong, D. W., Schneiderheinze, J. M., Anal. Chem. 2000, 72, 4474–4476.
[24] Hermansson, M., Colloids Surf. B Biointerfaces 1999, 14, 105–119. [25] Dziubakiewicz, E., Hrynkiewicz, K., Walczyk, M., Buszewski, B., Colloids Surf. B Biointerfaces 2013, 104, 122–127. [26] Rijnaarts, H. H. M., Norde, W., Lyklema, J., Zehnder, A. J. B., Colloids Surf. B Biointerfaces 1995, 4, 191–197. [27] Mignot, T., Denis, B., Couture-Tosi, E., Kolstø, A.-B., Mock, M., Fouet, A., Environ. Microbiol. 2001, 3, 493–501. [28] Karunakaran, E., Biggs, C., Appl. Microbiol. Biotechnol. 2011, 89, 1161–1175. [29] Orsod, M., Joseph, M., Huyop, F., Malays. J. Microbiol. 2012, 8, 170–174.
[6] Zheng, J., Yeung, E. S., Anal. Chem. 2003, 75, 818–824.
[30] Bos, R., van der Mei, H. C., de Vries, J., Busscher, H. J., Colloids Surf. B Biointerfaces 1996, 7, 101–112.
´ [7] Szumski, M., Kłodzinska, E., Buszewski, B., Microchim. Acta 2009, 164, 287–291.
[31] van der Mei, H. C., de Vries, J., Busscher, H. J., Appl. Environ. Microbiol. 1993, 59, 4305–4312.
´ [8] Szumski, M., Kłodzinska, E., Buszewski, B., J. Chromatogr. A 2005, 1084, 186–193.
[32] Handley, P. S., Harty, D. W. S., Wyatt, J. E., Brown, C. R., Doran, J. P., Gibbs, A. C. C., J. Gen. Microbiol. 1987, 133, 3207–3217.
[9] Petr, J., Ryparova, O., Znaleziona, J., Maier, V., Sevcik, J., Electrophoresis 2009, 30, 3863–3869.
[33] Sheng, G.-P., Yu, H.-Q., Li, X.-Y., Biotechnol. Adv. 2010, 28, 882–894.
[10] Kłodzinska, E., Szumski, M., Dziubakiewicz, E., Hrynkiewicz, K., Skwarek, E., Janusz, W., Buszewski, B., Electrophoresis 2010, 31, 1590–1596.
[34] Oukacine, F., Quirino, J. P., Garrelly, L., Romestand, B., Zou, T., Cottet, H., Anal. Chem. 2011, 83, 4949–4954.
´ [11] Buszewski, B., Szumski, M., Kłodzinska, E., Dahm, H., J. Sep. Sci. 2003, 26, 1045–1049.
[35] Oukacine, F., Garrelly, L., Romestand, B., Goodall, D. M., Zou, T., Cottet, H., Anal. Chem. 2011, 83, 1571–1578.
[12] Schneiderheinze, J. M., Armstrong, D. W., Schulte, G., Westenberg, D. J., FEMS Microbiol. Lett. 2000, 189, 39–44.
[36] Phung, S. C., Nai, Y. H., Powell, S. M., Macka, M., Breadmore, M. C., Electrophoresis 2013, 34, 1657–1662.
[13] Armstrong, D. W., Girod, M., He, L., Rodriguez, M. A.,
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[37] Tong, M.-Y., Jiang, C., Armstrong, D. W., J. Pharm. Biomed. Anal. 2010, 53, 75–80.
www.electrophoresis-journal.com
of Environmental Chemistry and Bioanalytics, Faculty of Chemistry, Nicolaus ´ Copernicus University, Torun, Poland 2 Centre for Modern Interdisciplinary Technologies, Nicolaus Copernicus University, ´ Poland Torun,
Received August 22, 2013 Revised December 4, 2013 Accepted December 9, 2013
Electrophoresis 2014, 35, 1160–1164
Research Article
Capillary electrophoresis of microbial aggregates Uncontrolled aggregation of bacterial cells is a significant disadvantage of electrophoretic separations. Various aspects of the electrophoretic behavior of different strains of Grampositive Bacillus cereus, Bacillus subtilis, Sarcina lutea, Staphylococcus aureus(1), and Micrococcus luteus bacteria and Gram-negative Escherichia coli bacteria were investigated in this study. Our findings indicate that bacteria can be rapidly analyzed by CZE with surface charge modification by calcium ions (Ca2+ ). Bound Ca2+ ions increase zeta potential to more than 2.0 mV and significantly reduce repulsive forces. Under the above conditions, bacterial cells create compact aggregates, and fewer high-intensity signals are observed in electropherograms. The above can be attributed to the bridging effect of Ca2+ between bacterial cells. CE was performed to analyze bacterial aggregates in an isotachophoretic mode. A single peak was observed in the electropherogram. Keywords: Bacterial aggregation / Capillary electrophoresis / Modified cell surface charge / Zeta potential DOI 10.1002/elps.201300588
Additional supporting information may be found in the online version of this article at the publisher’s web-site
1 Introduction CZE supports fast detection and identification of microorganisms, and it has a variety of practical applications in bioengineering, medical diagnosis, environmental protection, and food analysis [1–5]. Despite its popularity, the method has certain limitations, such as the aggregation and/or adhesion of bacteria to the surface of solids [6–14]. Every bacterial species has a complex and characteristic composition of the cell wall. Macromolecules that are present in the cell wall and bacterial membranes, including proteins, phospholipids, teichoic acid, teichuronic acid, and lipopolysaccharides, produce unique biochemical fingerprints. Those macromolecules also contribute to the surface charge of bacterial cells through the ionization of protonactive functional groups, such as carboxyl, phosphate, amino, and hydroxyl groups, and the adsorption of ions from the solution [15, 16]. Research studies have demonstrated that the discussed macromolecules play a significant role in bacterial aggregation and adhesion to solid and contribute to microbial differentiation [17, 18].
Correspondence: Professor Bogusław Buszewski, Department of Environmental Chemistry and Bioanalytics, Faculty of Chemistry, ´ Poland Gagarina 7, 87-100 Torun, E-mail: [email protected] Fax: +48-56-6114837
Abbreviation: PEO, poly(ethylene oxide) C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Bacteria may be divided into two major groups based on the Gram-staining reaction. Peptidoglycan is a significant component of the cell wall and is found in both bacterial groups. The cell wall of Gram-positive bacteria is composed of a rigid and relatively thick peptidoglycan layer with adjacent teichoic and teichuronic acids [19, 20]. In more complex envelopes of Gram-negative cells, peptidoglycan is found only in a thin layer between two membranes—the inner cytoplasmic membrane and outer membrane. The outer membrane consists of proteins, phospholipids, and lipopolysaccharides [21–23]. Bacterial adhesion was first explained by the DLVO (Derjaguin, Landau, Verwey, Overbeek) theory. Bacterial adhesion is mediated by an interplay between Lifshitz-van der Waals forces and electrostatic interactions originating from the overlap of electrical double layers and the solid surface. This theory has been extended to acid–base interactions [15, 24]. The leading theories explaining aggregation and adhesion effects during electrophoretic analysis were proposed by Armstrong and co-workers [12, 13], Zheng et al. [6], and Buszewski and co-workers [7, 10, 11] who characterized the electrophoretic behavior of bacteria based on analyses of electropherograms, zeta potential measurements, and microscopic observations. Despite those efforts, the described phenomena still lack a detailed theoretical explanation. The objective of this study was to investigate various aspects of the electrophoretic behavior of different strains of Gram-positive Bacillus cereus, Bacillus subtilis, Sarcina lutea, Staphylococcus aureus(1), and Micrococcus luteus bacteria, and Gram-negative Escherichia coli bacteria and to demonstrate www.electrophoresis-journal.com
CE and CEC
Electrophoresis 2014, 35, 1160–1164
that bacteria can be rapidly analyzed by CE with surface of bacteria charge modification by calcium ions (Ca2+ ) in an isotachophoretic mode.
1161
Between runs, the capillaries were washed with 1.0 M NaOH, deionized water for 2 min each, and running buffer for 4 min. A total of 0.5 mLstock bacterial suspension was used for electrophoretic measurements.
2 Materials and methods 3 Results and discussion 2.1 Preparation of bacterial suspensions Escherichia coli , B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacterial strains were obtained from the collection of the Department of Microbiology (Nicolaus Copernicus University, Poland). Bacterial cells were cultivated according to Dziubakiewicz et al. [25]. The final pellet of washed bacterial cells was suspended in 5 mL of 0.005 M NaNO3 and used for zeta potential measurements. The same procedure was applied to cultivate bacterial suspension for CE analysis, but the pellet with bacterial cells was washed and the final pellet of washed bacteria was suspended in a running buffer. The final pellet of washed bacterial cells was suspended in 5 mL of 0.005 M Ca(NO3 )2 to modify surface charges, and after 6 h, the pellet with bacterial cells was washed to remove free Ca2+ ions for future analysis. The OD (OD measured spectrophotometrically at 590 nm) of final bacterial suspensions varied between 0.097 and 0.320 (109 –1012 cells per mL).
Figure 1 shows the results of CZE analysis of E. coli, B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria in fused silica capillaries in TBE buffer. The obtained electropherograms indicate that the use of the TBE buffer in electrophoretic analyses of the examined bacteria is not effective. Similar results were reported for other buffers, including Tris and MES. Bacteria migrated close to electroosmotic flow (tEOF 1.02 min). The above conditions did not support a reproducible electrophoretic analysis of the examined bacteria. Bacterial adhesion to the inner surface of the capillary and the formation of bacterial aggregates with different size and surface charge, which produced several signals in the electropherograms (Fig. 1), were observed. The inner wall of the capillary was modified both dynamically and chemically to minimize surface adhesion and
2.2 Zeta potential The zeta potential of E. coli, B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacterial strains was measured in 0.005 M NaNO3 at 25°C using the Malvern Zetasizer Nano ZS system (Malvern Instrument, UK) equipped with a titrator (Malvern Instrument) and the Malvern polystyrene U-shaped cell. The suspension was acidified to pH ࣈ 2.0 with 0.1 M HNO3 (POCh, Gliwice, Poland) and was titrated to pH ࣈ 10.5 with 0.1 M NaOH (POCh, Gliwice). All titrations were performed using a titrator assembly consisting of a glass vessel with a lid. The electrode was calibrated over the entire pH range of 2.0–11.0. Zeta potential ( ) was calculated from electrophoretic mobility () with the use of Smoluchowski’s equation: =
, ε0 εr
(1)
where is the viscosity of the electrolyte solution, and ε r and ε o represent the relative permittivity of the electrolyte solution and vacuum, respectively. 2.3 Capillary electrophoresis CE experiments were performed in a HP3D CE system (Agilent Technologies, Waldbronn, Germany) equipped with a DAD with the use of fused silica capillaries (id = 75 m; Ltot = 33.5 cm; Leff = 25 cm; Composite Metal Services, Shipley, UK). New capillaries were rinsed with 1.0 M NaOH, deionized water, and BGE (outlet buffer) for 10 min each before use. C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. Electropherograms of E. coli, B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria. Conditions: TBE buffer, Leff = 25 cm, Ltot = 33.5 cm, detection at = 214 nm, V = 20 kV, injection – 100 mbar·s
www.electrophoresis-journal.com
1162
E. Dziubakiewicz and B. Buszewski
Electrophoresis 2014, 35, 1160–1164
Figure 2. The effect of calcium ions on the zeta potential of E. coli (A), B. cereus (B), B. subtilis (C), S. lutea (D), S. aureus(1) (E), and M. luteus (F) bacteria
slow down EOF. Dynamic modification involved the addition of poly(ethylene oxide) (PEO) to the buffer that increased buffer viscosity and decreased EOF. According to some authors [5,6,12], PEO contributes to the aggregation of bacterial cells. EOF was decreased, but reproducible analyses were not achieved, probably due to system instability. Casual interactions between PEO and bacterial cells, and the inner surface of the capillary (different film thickness) further deepened system instability. The use of capillaries modified with acrylamide eliminated EOF and enabled bacterial cells to migrate according to their own electrophoretic mobility. Despite the above, several signals were observed in the electropherograms (data not shown). The resulting curves of zeta potential of the examined bacteria as a function of pH (Fig. 2) indicate that the observed phenomenon can be attributed to the high content of peptides and polysaccharides in bacterial walls and membranes, which are responsible for uncontrolled aggregation and adhesion to the hydrophilic capillary surface. Gram-positive bacteria, excluding B. cereus, have the pI of ࣘ 2.2. Rijnaarts et al. [26] observed the presence of teichoic acids (R−HPO4 − /R−PO4 2− ) and anionic polysaccharides on the surface of bacterial cells with pKa ࣘ 2.8 (R−HPO4 − /R−PO4 2− ; R−COOH/ R−COO− ). The pI ࣙ 3.6 for B. cereus and E. coli bacteria cannot be unambiguously interpreted. All components of the bacterial wall and membrane should be taken into account. The pI can be determined by pI values of surface peptides and peptidoglycans (R−NH3 + /R−NH2 ; R−COOH/ R−COO− ), polymers, including amino groups (R−NH3 + /R−NH2 ) and polysaccharides (R−HPO4 − /R−PO4 2− ; R−COOH/ R−COO− ). The pI can also vary within a wide spectrum of values, subject to
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
differences in peptidoglycan composition in various bacterial species. The ratio of amino groups to carboxyl groups in peptidoglycan is minimum 1:3, which produces pI ⬎ 3.8. The interactions between amino and carboxyl groups can decrease pI values, whereas the amidation of carboxyl group can lead to an increase in pI. The pI of ࣙ 3.6 for B. cereus and E. coli bacteria suggests the presence of an additional protein layer. According to previous research, B. cereus, Bacillus antracis, and Bacillus brevis are surrounded by an S-layer composed of protein [27] or a mucous coating, which consists of extracellular polymeric substances (26–40% protein) [28, 29]. In E. coli, this phenomenon can be explained by the presence of lipoproteins, porins, and large numbers of protein appendages (flagella, fimbriae, pili) on the surface of the outer membrane. The modification of the surface charge of bacterial cells with Ca2+ facilitates cell aggregation and improves the selectivity of electrophoretic analysis. As demonstrated by Fig. 2, E. coli, B.cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria are charged negatively at pH ⬎ 4. Attractive electrostatic forces (van der Waals and acid–base forces) have to dominate repulsive forces for bacterial cells to aggregate. Deprotonated functional groups on the surface of bacterial cells support the formation of cationic bridges. Calcium ions bound to the surface of cells improve their hydrophobicity, and they evoke the participation of attractive acid–base forces. Calcium ions facilitate overcoming of the electrostatic energy barrier between cells, which in turn facilitates cell aggregation. Extracellular structures, including fimbriae, which bind Ca2+ ions at terminal sites also overcome repulsive forces. As demonstrated by Fig. 2, the modified surface charge of bacteria with calcium ions leads to changes in their
www.electrophoresis-journal.com
CE and CEC
Electrophoresis 2014, 35, 1160–1164
1163
Figure 3. Electropherograms of E. coli bacteria surface modification with Ca2+ (A) and without Ca2+ and application of CE in an isotachoretic mode (B). Conditions: Leff = 25 cm, Ltot = 33.5 cm, detection at = 214 nm, V = 15 kV, injection – 100 mbar·s
Table 1. Buffer compositions
Buffer composition
TRIS Boric acid HCl
Leading buffer
Terminating buffer
c [mM]
pH
c [mM]
pH
5.00 50 4.32
7.11
5.00 50 0
8.23
electrophoretic mobility. Bound Ca2+ ions increase zeta potential to more than 2.0 mV that, according to Bos et al. [30], significantly reduces repulsive forces. Under the above conditions, bacterial cells create compact aggregates [31–33], and fewer high-intensity signals are observed in electropherograms (Fig. 3A). Application of buffers with different ions mobilities (an isotachophoretic mode) and without bacterial surface modification with Ca2+ , according to Oukacine et al. [34,35] and Phung et al. [36] allowed focusing the zones. However, in our case we did not obtain single peak but multiplet (Fig. 3B). This phenomenon is probably due to high concentrations of bacterial cells (109 –1012 cells/mL). An optimal composition of buffers (leading buffer and terminating buffer) is presented in Table 1. Finally, CE in an isotachophoretic mode was performed to focus the migrating bacterial aggregates (after surface modifications with Ca2+ ) in a single zone, and as the result, a single signal was obtained in the electropherograms (Fig. 4). Sharpening of the peaks and an improvement in the shape of the basic line were observed. The high reproducibility of electrophoretic evaluations supported quantitative analyses of bacteria. The applied electrophoretic conditions resulted in good linearity. The correlation coefficient for the examined bacteria was determined at 0.98 (an example of quantification of E. coli bacteria is inserted in the Supporting Information Fig. 1). Application of developed technique did not allow for separation of bacteria and we observed their coelution. From this reason, this technique could be valuable and useful as a sterility test as previously described and published by Tong et al. [37]. C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 4. Electropherograms of E. coli (A), B. cereus (B), B. subtilis (C), S. lutea (D), S. aureus(1) (E), and M. luteus (F) bacteria. Conditions: buffer (Table 1), Leff = 25 cm, Ltot = 33.5 cm, detection at = 214 nm, V = 15 kV, injection – 100 mbar·s
4 Concluding remarks Various aspects of the electrophoretic behavior of different strains of Gram-positive B. cereus, B. subtilis, S. lutea, S. aureus(1), and M. luteus bacteria, and Gram-negative E. coli bacteria were investigated in this study. The following conclusions can be formulated based on our findings: (i) The modification of surface bacterial groups with Ca2+ ions contributed to the aggregation of bacterial cells and www.electrophoresis-journal.com
1164
E. Dziubakiewicz and B. Buszewski
decreased the number of signals in the electropherograms. (ii) CE in an isotachoretic mode was performed to focus bacterial aggregates in a single zone, and it supported quantitative analyses of bacterial cells (R2 = 0.98). (iii) The results of this study contribute to our knowledge of electrophoretic separation bacteria and supply new information about the surface charge of bacterial cells. Our findings proved to be useful for describing the fundamental mechanisms of bacterial aggregation.
Electrophoresis 2014, 35, 1160–1164
Wei, W., Zheng, J., Yeung, E. S., Anal. Chem. 2002, 74, 5523–5530. [14] Desai, M. J., Armstrong, D. W., Microbiol. Mol. Biol. Rev. 2003, 67, 38–51. [15] Poortinga, A. T., Bos, R., Norde, W., Busscher, H. J., Surf. Sci. Rep. 2002, 47, 1–32. [16] Beveridge, T. J., Murray, R. G., J. Bacteriol. 1980, 141, 876–887. [17] Tsuneda, S., Aikawa, H., Hayashi, H., Hirata, A., J. Colloid Interface Sci. 2004, 279, 410–417. [18] Eboigbodin, K. E., Newton, J. R. A., Routh, A. F., Biggs, C. A., Appl. Microbiol. Biotechnol. 2006, 73, 669–675.
This study was supported by grant no. N N204 369040 from the National Science Centre in Warsaw, Poland, and a grant from the European Social Fund, Polish National Budget, and budget of the Kujawsko–Pomorskie Region as part of the “Krok w przyszło´sc´” (Step into the Future) Sectoral Operational Program—Human Resources. The authors have declared no conflict of interest.
[19] Doyle, R. J., Matthews, T. H., Streips, U. N., J. Bacteriol. 1980, 143, 471–480. [20] van der Wal, A., Norde, W., Zehnder, A. J. B., Lyklema, J., Colloids Surf. B Biointerfaces 1997, 9, 81–100. [21] Hammond, S. M., Lambert, P. A., Rycroft, A. N., The Bacterial Cell Surface, Croom Helmm, London 1984. ˙ [22] Kunicki-Goldfinger, W. J. H., Zycie bakterii, PWN, Warsow 1998. [23] Beveridge, T. J., Davies, J. A., J. Bacteriol. 1983, 846–858.
5 References [1] Armstrong, D. W., Schulte, G., Schneiderheinze, J. M., Westenberg, D. J., Anal. Chem. 1999, 71, 5465–5469. [2] Lim, O., Suntornsuk, W., Suntornsuk, L., J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2009, 877, 710–718. [3] Gao, P., Xu, G., Shi, X., Yuan, K., Tian, J., Electrophoresis 2006, 27, 1784–1789. [4] Armstrong, D. W., Schneiderheinze, J. M., Kullman, J. P., He, L., FEMS Microbiol. Lett. 2001, 194, 33–37. [5] Armstrong, D. W., Schneiderheinze, J. M., Anal. Chem. 2000, 72, 4474–4476.
[24] Hermansson, M., Colloids Surf. B Biointerfaces 1999, 14, 105–119. [25] Dziubakiewicz, E., Hrynkiewicz, K., Walczyk, M., Buszewski, B., Colloids Surf. B Biointerfaces 2013, 104, 122–127. [26] Rijnaarts, H. H. M., Norde, W., Lyklema, J., Zehnder, A. J. B., Colloids Surf. B Biointerfaces 1995, 4, 191–197. [27] Mignot, T., Denis, B., Couture-Tosi, E., Kolstø, A.-B., Mock, M., Fouet, A., Environ. Microbiol. 2001, 3, 493–501. [28] Karunakaran, E., Biggs, C., Appl. Microbiol. Biotechnol. 2011, 89, 1161–1175. [29] Orsod, M., Joseph, M., Huyop, F., Malays. J. Microbiol. 2012, 8, 170–174.
[6] Zheng, J., Yeung, E. S., Anal. Chem. 2003, 75, 818–824.
[30] Bos, R., van der Mei, H. C., de Vries, J., Busscher, H. J., Colloids Surf. B Biointerfaces 1996, 7, 101–112.
´ [7] Szumski, M., Kłodzinska, E., Buszewski, B., Microchim. Acta 2009, 164, 287–291.
[31] van der Mei, H. C., de Vries, J., Busscher, H. J., Appl. Environ. Microbiol. 1993, 59, 4305–4312.
´ [8] Szumski, M., Kłodzinska, E., Buszewski, B., J. Chromatogr. A 2005, 1084, 186–193.
[32] Handley, P. S., Harty, D. W. S., Wyatt, J. E., Brown, C. R., Doran, J. P., Gibbs, A. C. C., J. Gen. Microbiol. 1987, 133, 3207–3217.
[9] Petr, J., Ryparova, O., Znaleziona, J., Maier, V., Sevcik, J., Electrophoresis 2009, 30, 3863–3869.
[33] Sheng, G.-P., Yu, H.-Q., Li, X.-Y., Biotechnol. Adv. 2010, 28, 882–894.
[10] Kłodzinska, E., Szumski, M., Dziubakiewicz, E., Hrynkiewicz, K., Skwarek, E., Janusz, W., Buszewski, B., Electrophoresis 2010, 31, 1590–1596.
[34] Oukacine, F., Quirino, J. P., Garrelly, L., Romestand, B., Zou, T., Cottet, H., Anal. Chem. 2011, 83, 4949–4954.
´ [11] Buszewski, B., Szumski, M., Kłodzinska, E., Dahm, H., J. Sep. Sci. 2003, 26, 1045–1049.
[35] Oukacine, F., Garrelly, L., Romestand, B., Goodall, D. M., Zou, T., Cottet, H., Anal. Chem. 2011, 83, 1571–1578.
[12] Schneiderheinze, J. M., Armstrong, D. W., Schulte, G., Westenberg, D. J., FEMS Microbiol. Lett. 2000, 189, 39–44.
[36] Phung, S. C., Nai, Y. H., Powell, S. M., Macka, M., Breadmore, M. C., Electrophoresis 2013, 34, 1657–1662.
[13] Armstrong, D. W., Girod, M., He, L., Rodriguez, M. A.,
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[37] Tong, M.-Y., Jiang, C., Armstrong, D. W., J. Pharm. Biomed. Anal. 2010, 53, 75–80.
www.electrophoresis-journal.com
