In this study, we investigate the electrohydrodynamic and nanomechanical characteristics of two Saccharomyces cerevisiae yeast strains, a wild-type (WT) strain and a strain overexpressing (OE) Hsp12p, in the presence and absence of hydrophobic Congo red compound. By combining these two advanced biophysical methods, we demonstrate that Hsp12p proteins are mostly located within a thin layer (c. 10 nm thick) positioned at the external side of the cell wall. However, this Hsp12p-enriched layer does not prevent Congo red from entering the cell wall and from interacting with the chitin therein. The entrance of Congo red within the cell wall is reflected in an increase of the turgor pressure for the OE strain and a decrease of that for the WT strain. It is shown that these opposite trends are consistent with significant modulations of the water content within the cell wall from/to the cytoplasm. These are the result of changes in the hydrophobicity/hydrophilicity balance, as governed by the intertwined local concentration variations of Congo red and Hsp12p across the cell wall. In particular, the decrease of the turgor pressure in the case of WT strain upon addition of Congo red is shown to be consistent with an upregulation of Hsp12p in the close vicinity of the plasma membrane.
Congo red, the sodium salt of benzidinediazo-bis-1-napthylamine-4-sulphonic acid, is a diazo dye known to bind to the cell wall of the yeast Saccharomyces cerevisiae. It has been proposed that Congo red forms a complex with chitin, a poly-N-acetyl-d-glucosamine polysaccharide (Vannini, 1983; Pancaldi, 1985; Kopecka & Gabriel, 1992), found in small amounts dispersed over the entire cell periphery in normally growing cells (Molano, 1980) and especially in the bud scars (Cabib, 2001). Binding of Congo red to chitin results in altered cell wall assembly and cell separation, and induces numerous genes mainly concerned with cell wall organization, morphogenesis, signal transduction and stress response (Garcia, 2004). This latter class includes the fourfold upregulation of HSP12, which codes for a 12-kDa hydrophilic protein largely localized in the plasma membrane (Sales, 2000). This protein protects against desiccation- and ethanol-induced stress (Motshwene, 2004), and acts as a plasticizer in the cell wall (Karreman & Lindsey, 2005; Karreman, 2007).
It was previously shown that the presence of Hsp12p reduces the sensitivity of yeast cells to Congo red (Motshwene, 2004). As an example of this action, HSP12 deletion mutants failed to grow on plates containing a low concentration (0.43 mM) of Congo red dye (Motshwene, 2004). In addition, when these HSP12 deletion mutants were grown in the presence of lower concentrations of Congo red (<0.14 mM), the cells displayed elevated levels of cellular clumping, enhanced sedimentation and aggravated septation defects when compared with wild-type (WT) cells (Karreman & Lindsey, 2007) with a certain concentration level of Hsp12p within the cell wall. The Δhsp12 cells had up to five daughter cells attached to the maternal cell compared with a maximum of two for WT cells. Because chitin has been reported to be the most abundant component of the primary septum that is formed between mother and daughter cells during the budding process and because it is required for cell separation (Cabib, 2001), we postulated that one function of Hsp12p is to protect chitin from interacting with phenolics such as Congo red (Karreman & Lindsey, 2007). Interestingly, we were able to demonstrate that Hsp12p inhibited the binding of Congo red to insoluble chitin using an in vitro assay.
If this hypothesis is correct, then overexpressing (OE) Hsp12p, using a constitutive promoter, should result in a yeast strain that is less sensitive to Congo red than WT cells. In this manuscript, we quantify the effect of Congo red on yeast cells (OE or not Hsp12p) by measuring their electrohydrodynamic and nanomechanical properties in the presence and absence of Congo red compound. In particular, the interpretation of the electrohydrodynamic data is based on recent theoretical developments (Duval & Ohshima, 2006) that have been applied to numerous environmental and biological particles, in particular humic acids (Duval, 2005a), bacteria (Duval, 2005b; Dague, 2006; Gaboriaud, 2008a), viruses (Langlet, 2008) and more recently yeasts (Karreman, 2007). Additionally, we have combined advanced biophysical tools such as force spectroscopy (Gaboriaud & Dufrêne, 2007) to address the impact of Hsp12p on the physicochemical properties and molecular-level structure of the yeast envelope exposed or not to Congo red dye.
Materials and methods
Yeast strains and culture conditions
The S. cerevisiae WT (MATa, ade2-1, trp1-1, leu2-3, his3-11, ura3 and canr1-100) originates from the W303 background. To overexpress HSP12, the coding sequence of the HSP12 gene was cloned into the pDLG1 plasmid under the control of the alcohol dehydrogenase 2 promoter (La Grange, 2001). The resulting plasmid was transformed into the WT strain using electroporation. Electrocompetent cells were prepared by diluting overnight W303 WT cultures to an OD of c. 25 OD units (ODU) and then sequentially collected by centrifugation at 3000 g for 10 min at 4 °C, washed twice with ice-cold sterile water, washed twice with ice-cold 1 M sorbitol and then resuspended in a minimal volume of ice-cold 1 M sorbitol. After 1 μg DNA was added to 50 μL electrocompetent yeast cells, the latter were pulsed for 5.1 ms on the SC2 setting using a Biorad Micropulser (Biorad, CA) (Delorme, 1989). A 500-μL aliquot of ice-cold 1 M sorbitol was immediately added to the cells and subsequently plated on SC-URA medium containing 1 M sorbitol to select for positive transformants.
Yeast cells were routinely grown in 1% (w/v) yeast extract, 2% (w/v) peptone and 2% (w/v) glucose (YEPD) media at 30 °C on a rotary shaker at c. 200 r.p.m. The strains were grown either to the midexponential (10 h cultures, 2.6 ODU for WT and 2.9 ODU for Hsp12p overexpressed) or to the stationary phase (38 h cultures, 7.6 ODU for WT and 8.1 ODU for Hsp12p overexpressed), 1 ODU being equivalent to 3 × 107 cells mL−1. Yeast growth was monitored by measurement of the OD of the culture at 600 nm (OD600 nm) and expressed in ODU.
Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE)
Cell wall proteins extracted using 0.2 M NaOH were separated on the basis of mass and visualized by SDS-PAGE as previously described (Laemmli, 1970). A total extract of chicken erythrocyte histones served as a standard, their approximate molecular weights (kDa) being, H1, 22.5; H5, 20.6; H3, 15.3; H2B, 13.7; H2A, 14.0 and H4, 11.2.
Force spectroscopy measurements
Force curves were obtained at room temperature, using an MFP-3D instrument (Asylum Research, Santa Barbara). All force experiments were performed in 1 mM KNO3 solution at neutral pH using microlever probes (MLCT-EXMT-BF, Santa Barbara) with the spring constant systematically determined following the thermal calibration method (mean value equal to 55±5 pN nm−1). Glass slides (38 × 26 mm; Menzer GmBH, Braunschweig, Germany) were washed overnight in 70% nitric acid before being rinsed with distilled water. The slides were then immersed in 0.2% (w/v) polyethyleneimine solution (Sigma Chemical Co.), left for a 4-h delay, after which they were washed with distilled water. To immobilize the cells on the dried polyethyleneimine-prepared slides, a droplet of yeast cells (OD600 nm of 0.06) prepared in 1 mM KNO3 solution was deposited onto the slide for a period of 1 h; the slide was subsequently rinsed and directly transferred into the liquid cell mounted in the atomic force microscopy (AFM) instrument. The cells were precisely located using an optical image obtained from the inverted microscope that the AFM apparatus is equipped with. The cantilever was then moved to the apex of the cell and force curves were recorded.
The measured force profiles were converted into force-indentation curves and various key mechanical parameters were evaluated according to a procedure described in detail elsewhere (Gaboriaud, 2008b) and illustrated in Fig. 1. The linear behaviour at high loading forces, as measured upon approach of the tip to the cell, was quantitatively interpreted in terms of a cell spring constant, denoted as kcell. This parameter basically corresponds to the slope of the constant compliance region, with δ(Linear) the indentation value marking the separation between the linear portion of the force profiles and the nonlinear regime (Fig. 1). While kcell is related to the turgor pressure that counteracts the compression of the cell's cytoplasm by the AFM tip, the δ(Linear) values correspond to the thickness of the outer gel-like layer, which is assimilated into the cell wall in our study (Gaboriaud, 2005, 2008a). The nonlinear regime at low loading forces was interpreted using the Hertz model taking into account the pyramidal geometry of the AFM tip, according to the following:Touhami, 2003), α is the cone opening angle and E is the sample's Young's modulus that stands for the mechanical softness/rigidity of the external cell wall layer of the yeast.
Soft particle electrokinetic analysis
Electrophoretic mobility measurements were carried out in a Zetaphorimeter IV (SEPHY-CAD Instrumentations) equipped with a laser illumination and video interface, via a CCD camera and image software analysis. Under the action of a direct-current electric field applied between two electrodes mounted in a quartz suprasil cell thermostated at 25 °C, yeast cells migrated and their trajectories were recorded and processed for evaluating the corresponding electrophoretic mobility defined as the ratio between electrophoretic velocity and electric field strength. Measurements were performed for yeast cells cultured as detailed above and subsequently washed and dispersed in a KNO3 aqueous electrolyte, of which the concentration was varied in the range of 1–100 mM at a constant pH c. 6.5. For a given electrolyte concentration, at least 50 measurements of cell mobility were recorded.
Because yeast cells are paradigms of soft permeable bioparticles (Duval, 2007; Karreman, 2007), the electrokinetic results were quantitatively interpreted on the basis of the numerical theory developed by Duval & Ohshima (2006). This theory, which abandons the necessarily inappropriate concept of electrokinetic potential for permeable particles, allows the evaluation of the interphasial electrostatic and hydrodynamic properties of bioparticles without any assumption on the magnitude of the charge they carry, their size and the salt level in the medium. More comprehensively, the cells are viewed as core–shell particles with core radius a of about 1.5 μm and shell thickness δ 100 nm, dimensions that basically correspond to yeast cell size and cell wall thickness, respectively. Experimental data were consistently reconstructed with theory by adjusting the two relevant electrokinetic parameters: one denoted as ρo, which pertains to the volume density of ionogenic charges distributed throughout the permeable shell, and another denoted as 1/λ0, which corresponds to the typical hydrodynamic flow penetration length within the hydrodynamically soft structure of the yeast cell. For more information, technical details may be found in Duval & Ohshima (2006) and Gaboriaud (2008a). It is stressed that the advanced electrokinetic analysis used for interpreting electrokinetic data under the ionic strength conditions of the current study is independent of the size a of the core particle as well as the thickness δ of the soft component because the criteria κδ≫1 and κa≫1 are satisfied with κ, the reciprocal screening Debye layer thickness (Duval & Ohshima, 2006). Under the conditions where electrokinetic experiments were performed, such criteria are always met for sizes between 3 and 10 μm and for shell thickness between 50 and 150 nm. In brief, regardless of the setting of these dimensions (core and shell), the same electrohydrodynamic characteristics are obtained (which was verified theoretically) because we are essentially investigating systems that comply with the conditions of thin electric double layers within the ionic strength regime of the experiments.
SDS-PAGE analysis of the Hsp12p content
We initially compared the Hsp12p content in the cell wall of the WT strain with that of the cell wall of the strain (OE) that supposedly overexpresses Hsp12p. Cells of both strains were grown to both midlog and stationary phases and the cell wall proteins were extracted with 0.2 M NaOH (Motshwene, 2004). SDS-PAGE of the extracts (Fig. 2) showed that while the WT strain had significant quantities of Hsp12p only in the stationary phase, the OE strain exhibits considerable amounts of Hsp12p in both midlog and stationary phases. The relative amounts of Hsp12p were determined by scanning the stained gel. It was found that the WT strain had doubled its Hsp12p content from the midlog to the stationary phase. Furthermore, the OE strain had approximately four times more Hsp12p in both midlog and stationary phases when compared with the WT strain in the midlog phase.
Macroscopic observations of yeast cells
We next investigated whether an increase in the Hsp12p content of the yeast cell wall would lead to reduced binding of Congo red. Both WT and OE strains were grown to the midlog phase in YEPD medium; Congo red was then added to 0.02 mg mL−1 and the yeast was grown for a further 16 h. After extensive washing of the cells with 50 mM phosphate and 150 mM NaCl at pH 7.4, it was apparent that WT cells were markedly pinker than OE cells (Fig. 3). This observation suggests that Congo red binds significantly more to WT than to OE cells, in line with expectation and former reports in the literature. We were, however, not able to quantitatively measure the difference in binding, as scattering caused by the cells occurs in a range of wavelengths similar to that of Congo red (498 nm).
Electrohydrodynamic characteristics of OE and WT strains as determined by soft particle electrokinetic theory
Figure 4 displays the electrokinetic properties of WT and OE cells in the presence and absence of Congo red. In the absence of Congo red (Fig. 4, top panels), the respective values obtained for the volume charge densities (ρo) of WT and OE strains are similar within experimental error. Like most microorganisms, yeast cells are negatively charged at a neutral pH. The resulting ionic strength–mobility patterns obtained for both S. cerevisiae strains used in this study are in line with our previous investigation (Karreman & Lindsey, 2007) and with the results obtained by Tazhibaeva (2003). In the latter study, the authors, however, incorrectly converted electrophoretic mobility into ζ-potential and surface charge density for reasons mentioned and previously discussed by Duval & Ohshima (2006) and Duval (2007). The charge distributed throughout the cell wall probably originates from the presence of carboxyl and phosphoric acid functional groups typically present in yeast cell walls (Takashima & Morisaki, 1997). Without any detailed information on the cell wall composition and on the impact of the presence of Hsp12p, it is highly speculative to unambiguously address the origin of the slight variation in ρo associated with the soft cell wall structure of the WT and OE strains. The values obtained for the volume charge density of microbial cells usually lie in the range of −1 to −50 mM (see e.g. Takashima & Morisaki, 1997; Vadillo-Rodriguez, 2002; Gaboriaud, 2008a). Regardless of the experimental conditions (i.e. overpresence or not of Hsp12p), the values obtained in this study (−1.5 to −2.1 mM) are not significantly different, and they indicate a similar low number of charges within the part of the cell wall that is probed by the electro-osmotic flow.
In contrast, the hydrodynamic flow penetration length significantly differs according to the type of strain examined, with around 3.0 nm for WT and 4.4 nm for OE. As explained elsewhere (Duval & Ohshima, 2006), characterizes the permeable nature of a given soft interface. In more detail, it refers to the typical flow penetration length within the soft structure analysed. Differences in as measured for various soft bioparticles may be interpreted in terms of differences in soft surface layer compactness (i.e. degree of steric inhibition to fluid flow penetration) or differences in the hydrophilic/hydrophobic balance of the interface. In the literature on the electrokinetic properties of soft bioparticles [bacteria (Duval, 2005b; Dague, 2006; Gaboriaud, 2008a), viruses (Langlet, 2008) and yeast (Karreman, 2007)], the magnitude of typically ranges from 0. to 4.0 nm, and a variation in as small as 0.2 nm is significant from an electrokinetic point of view (Duval & Ohshima, 2006; Langlet, 2008).
In the presence of Congo red (Fig. 4, bottom panel), the electrohydrodynamic characteristics of the OE strain are basically identical to those evaluated in the absence of Congo red. On the contrary, major differences were found for the WT in the presence and absence of Congo red. In particular, a 100% increase is obtained for of WT strain in the presence of Congo red as compared with the situation where Congo red is absent from the solution. This increase in the propensity of the WT cells to hydrodynamic flow penetration is directly related to their aggregate state (maternal cell with attached daughter cells as previously demonstrated by Karreman & Lindsey, 2007; see Fig. 3). The impact of soft particle aggregation on the hydrodynamic softness parameter is well known and has already been observed and demonstrated for various biological and environmental systems (e.g. Duval, 2005a; Langlet, 2008). Given these elements, the electrokinetic data obtained for WT cells in the presence of Congo red cannot be easily interpreted in terms of changes in the interfacial properties of individual cells because they are mostly present in the form of aggregates. It is important to point out that we chose not to prevent aggregation by adding some molecules because such an operation would inevitably alter the electrohydrodynamic characteristics of the cell interface (via e.g. adsorption on cells) and thus mask the targeted effects of Congo red. Despite this, the electrokinetic results unambiguously underline the intimate relationship that exists between the amount of Hsp12p present in the cell wall and the binding action of Congo red with consequences for cell aggregation.
Nanomechanical properties measured by force spectroscopy
Figure 5a–c depicts the different mechanical characteristics for the WT and OE strains grown in either YEPD medium or YEPD supplemented with 0.02% (w/v) Congo red solution. Figure 5d schematically illustrates the relationship between yeast cell wall structure and the mechanical characteristics measured from the force-curve analysis. In simple terms, the cell spring constant (kcell) reflects the inner turgor pressure of the cell, Young's modulus (E) characterizes the elasticity of the yeast cell wall and the onset of the linear compliance [δ(linear)] is a measure of the cell wall thickness.
In the absence of Congo red, the mechanical properties of both strains studied were very similar. The Young moduli (Fig. 5b) were found to be c. 1–2 MPa for both the WT and the OE strains, which agrees reasonably well with the values previously reported for a different type of S. cerevisiae strain (E=0.6 MPa, Touhami, 2003). The thickness of the cell wall estimated from the nonlinear nanoindentation regime (i.e. the onset of the linear compliance, Fig. 5c) was about 50 nm, with a broader distribution for the WT strain. Cell spring constant values (Fig. 5a) ranged between 100 and 200 mN m−1, which are 10 times higher than the values reported for the WT in our previous study (kcell=17 mN m−1, Karreman, 2007). To explain this difference, it is important to note that in our previous study, we used a different cantilever (cantilever spring constant, tip geometry) and we pinpointed the apical surface of the yeast from the AFM images. In the latter case, we recently demonstrated that contact-mode imaging does not accurately locate the apical surface because a component of the applied load laterally deforms the cell during the raster scan (Gaboriaud, 2008b). Therefore, the low value obtained in our previous study (Karreman, 2007) may be attributed to the limitation of contact-mode imaging in accurately locating the apical surface of a yeast cell.
In the presence of Congo red, the magnitude of Young's modulus (Fig. 5b) and the cell wall thickness (Fig. 5c) remained roughly constant (P>0.1) for both strains. In contrast, the presence of Congo red induces remarkable variations in the cell spring constants and thus in the turgor pressure of the cells (Fig. 5a). In the case of Hsp12p-enriched cells (OE), the cell spring constant significantly increases whereas for the WT strain it shows a reversed trend.
It was clearly observed that the addition of Congo red leads to distinct responses in the WT and OE strains when subjected to electrical or mechanical constraints as done within the framework of electrophoresis and AFM experiments, respectively. Before discussing the results, it is important to emphasize that both methods probe the cells according to very different spatial scales. With electrophoresis, one investigates the electrohydrodynamic properties of the most external layer of the cell wall, typically over a spatial range that spans a few times the fluid flow penetration length . In this study, this corresponds to a thickness of about 10 nm at most. In contrast, AFM force spectroscopy basically remains insensitive in modulations of the structure of the aforementioned external part of the cell wall, but rather provides quantitative information on the physical properties of the whole cell wall, including the whole cytoplasm compartment. Bearing these elements in mind, electrokinetics and AFM provide complementary information for analysing the dynamic response of cells in terms of changes in the structure and chemical composition when exposed to Congo red dye.
Impact of the Hsp12p protein on the physical properties of S. cerevisiae
As a first step, we may compare the values of for WT and OE strains in the absence of Congo red as obtained in this study with those previously reported (Karreman, 2007) for WT and yeast cells lacking the gene encoding Hsp12p (Δhsp12). Such a comparison allows us to establish the following sequence in flow permeability: . This sequence suggests that the presence of the hydrophilic stress response protein Hsp12p within the yeast cell wall increases the fluid flow permeability of the latter. On the basis of simple thermodynamic considerations, it may be expected that the affinity of water for cell wall compounds is increased upon reduction of overall cell wall hydrophobicity following the presence of Hsp12p. In addition, force spectroscopy did not reveal any major differences between the nanomechanical properties of the cell wall (Young's modulus, Fig. 5b) and the cell cytoplasm (spring constant, Fig. 5a) of WT and OE cells. These results, together with those obtained by electrokinetics, strongly indicate that Hsp12p is confined to the most external side of the cell wall (that is, the cell wall part probed by electrokinetics) and that Hsp12p is present there at a higher concentration for the OE strain than for the WT strain as illustrated in Fig. 6 (left part). This result is consistent with the previous study of immunolocalization of Hsp12p (Motshwene, 2004), which clearly showed the presence of the protein in the cell wall.
Effect of Congo red on the OE Hsp12p S. cerevisiae
No significant changes were observed for the electrokinetic parameters pertaining to the OE cells when Congo red was added to the growth media (Fig. 4). In contrast, the analysis of cell nanomechanics shows that the addition of Congo red leads to a significant decrease of Young's modulus (Fig. 5b). This indicates that Hsp12p is probably exclusively confined within the external cell wall layer, in line with results obtained in the previous section, and that Congo red mostly interacts with compounds located in the inner part of the cell wall (probed by AFM). In more detail, the most external layer of cell wall that is probed by electrokinetics (around c. 10 nm thick) is Hsp12p rich, and prevents Congo red from interacting with specific cell wall constituents, in particular chitin, located at the cell wall periphery. Because previous reports demonstrated that chitin is homogeneously distributed throughout the cell wall (Molano, 1980), we can envisage that the Hsp12p content in the inner layer (c. 40 nm thick) of the OE strain cell wall is rather low as compared with that in the outer cell wall periphery. As a result, the transport of Congo red within this thick inner layer is not strongly inhibited by Hsp12p and Congo red may interact freely with chitin components, leading to variations in the nanomechanical properties as reported in Fig. 5b. Because Congo red is a significantly hydrophobic diazo dye (Vannini, 1983; Pancaldi, 1985; Kopecka & Gabriel, 1992), its presence within the heart of the cell wall is expected to strongly modulate the water content therein. The concomitant decrease in Young's modulus and increase in the cell spring constant (Fig. 5a) for OE cells in the presence of Congo red are in line with the process of water release from the cell wall towards the external media and the cytoplasm, as illustrated in Fig. 6. To summarize the preceding arguments, the thin external layer rich in Hsp12p prevents Congo red from interacting with chitin located within this peripheral cell wall layer, in agreement with electrokinetic results. Such an Hsp12p-rich top layer does not, however, inhibit the transport of Congo red molecules further within inner cell wall compartments where Hsp12p is present at a lower concentration. There, Congo red may interact with chitin, this interaction being favoured by the absence of a significant amount of Hsp12p. The presence of Congo red in these inner cell wall spatial regions results in modulation of the water content within the cell wall and within the outer part of the cytoplasm, as illustrated in Fig. 6, where a schematic view is presented for the different processes that come into play. The comparison between the electrokinetic and the nanomechanical properties of cells in the presence and absence of Congo red unambiguously shows that the action of this dye on cells takes place in a nonuniform manner throughout the cell wall (anisotropic action).
Effect of Congo red on the WT S. cerevisiae
In the case of WT cells, the electrokinetic and AFM data are consistent with a dye that operates according to a somewhat different mode as compared with that for OE strains. The Hsp12p content of the cell wall is now four times lower than that for OE cells (Fig. 2). We assume from the analysis detailed in the previous section that the HSP12p proteins are also located within a thin top layer on the external part of the cell wall. In the presence of Congo red, Young's modulus decreases in the same order of magnitude as that observed for the OE cells (Fig. 5b). The process is similar to that explained for the OE cells: the transport of hydrophobic Congo red within the WT cell wall is accompanied by water expulsion from the cell wall to the external medium. However, this decrease of the Young modulus is now accompanied by a decrease in the cell spring constant (Fig. 5a). In line with the literature (Garcia, 2004; Karreman & Lindsey, 2007), we argue that HSP12 expression is upregulated by the very presence of Congo red, thus leading to the synthesis of Hsp12p in the case of WT cells. This synthesis is likely to be concomitant with the entrance of Congo red in the interior of the cell wall. As a result, a three-layered structure within the cell wall may be visualized: the top (external part of the cell wall) and bottom (close to the plasma membrane) Hsp12p-rich layers sandwich the intermediate Congo red–chitin layer, as illustrated in Fig. 6. We suggest that the innermost (hydrophilic) Hsp12p-rich layer (which results from upregulation of Hsp12 due to the presence of Congo red) alters the water balance of the cell wall by allowing water molecules from the cytoplasm to enter the cell wall, which ultimately leads to the observed decrease in the spring constant.
In summary, we have shown that the hydrophilic Hsp12p protects the cells from the surrounding environment by forming a thin layer on the most external part of the cell wall. The electrohydrodynamic properties of such a thin layer positioned at the periphery of the cell wall may be investigated by means of electrokinetics that reveal significant differences in flow permeability depending on the amount of Hsp12p present, or equivalently, depending on the type of yeast strain analysed (OE and WT). Based on AFM nanoindentation analysis, we state that this enriched Hsp12p layer does not, however, prevent the exchange of Congo red molecules between the outer electrolytic environment and the inner part of the cell wall up to the inner cell wall regions close to the cytoplasm. We found that the addition of Congo red molecules does indeed induce a significant modulation of the water content within the cell wall and cytoplasm with, as a result, significantly different nanomechanical characteristics for the cell as a whole. By combining advanced biophysical methods that probe cells at various spatial scales, we were able to propose and identify the respective structures and chemical compositions of yeast cells containing different Hsp12p contents (Fig. 6), in line with all electrokinetic and nanomechanical observations. It is believed that the use of these types of approaches will provide new routes to better understand the dynamic changes of microbial interfaces in response to the addition of molecules with defined properties such as antibiotics, antiadhesives or dyes. Combining electrokinetics and atomic force microscopy allows the recording of a physical signature of the impact of specific molecules at the interphase formed by biological cells with their surrounding aqueous medium and within the bulk region of the cell, respectively.
V.J.S. and G.G.L. acknowledge the support of the National Research Foundation of South Africa and the University of Cape Town Research Fund. F.G. and G.G.L. thank the cooperation program between NRF and CNRS (No. 19796, 2005–2007).
We dedicate this article to the memory of Prof. George G. Lindsey.