Limits...
Bacterial surface appendages strongly impact nanomechanical and electrokinetic properties of Escherichia coli cells subjected to osmotic stress.

Francius G, Polyakov P, Merlin J, Abe Y, Ghigo JM, Merlin C, Beloin C, Duval JF - PLoS ONE (2011)

Bottom Line: Additionally, for a given surface appendage, the magnitude of the nanomechanical parameters decreases significantly when increasing bulk salt concentration.This effect is ascribed to a bacterial exoosmotic water loss resulting in a combined contraction of bacterial cytoplasm together with an electrostatically-driven shrinkage of the surface appendages.Altogether, AFM and electrokinetic results clearly demonstrate the intimate relationship between structure/flexibility and charge of bacterial envelope and propensity of bacterium and surface appendages to contract under hypertonic conditions.

View Article: PubMed Central - PubMed

Affiliation: Laboratoire de Chimie Physique et Microbiologie pour l'Environnement, Nancy Université, CNRS UMR7564, Villers-lès-Nancy, France. gregory.francius@lcpme.cnrs-nancy.fr

ABSTRACT
The physicochemical properties and dynamics of bacterial envelope, play a major role in bacterial activity. In this study, the morphological, nanomechanical and electrohydrodynamic properties of Escherichia coli K-12 mutant cells were thoroughly investigated as a function of bulk medium ionic strength using atomic force microscopy (AFM) and electrokinetics (electrophoresis). Bacteria were differing according to genetic alterations controlling the production of different surface appendages (short and rigid Ag43 adhesins, longer and more flexible type 1 fimbriae and F pilus). From the analysis of the spatially resolved force curves, it is shown that cells elasticity and turgor pressure are not only depending on bulk salt concentration but also on the presence/absence and nature of surface appendage. In 1 mM KNO(3), cells without appendages or cells surrounded by Ag43 exhibit large Young moduli and turgor pressures (∼700-900 kPa and ∼100-300 kPa respectively). Under similar ionic strength condition, a dramatic ∼50% to ∼70% decrease of these nanomechanical parameters was evidenced for cells with appendages. Qualitatively, such dependence of nanomechanical behavior on surface organization remains when increasing medium salt content to 100 mM, even though, quantitatively, differences are marked to a much smaller extent. Additionally, for a given surface appendage, the magnitude of the nanomechanical parameters decreases significantly when increasing bulk salt concentration. This effect is ascribed to a bacterial exoosmotic water loss resulting in a combined contraction of bacterial cytoplasm together with an electrostatically-driven shrinkage of the surface appendages. The former process is demonstrated upon AFM analysis, while the latter, inaccessible upon AFM imaging, is inferred from electrophoretic data interpreted according to advanced soft particle electrokinetic theory. Altogether, AFM and electrokinetic results clearly demonstrate the intimate relationship between structure/flexibility and charge of bacterial envelope and propensity of bacterium and surface appendages to contract under hypertonic conditions.

Show MeSH

Related in: MedlinePlus

Electrokinetic analysis of the bacterial electrophoretic mobilities.Quantitative analysis of the electrokinetic patterns of (A) E2152, (B) E2146, (C) E2302, (D) E2498. The charge density , the characteristic flow penetration length , the heterogeneity length scale  and the thickness d of the soft bacterial component are indicated below. (A) E2152: nm. mM, nm, . (B) E2146: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (C) E2302: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (D) E2498: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. N.B.: Values for  and  are indicated with a precision of ±10% in relation with error bars of experimental data.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3105017&req=5

pone-0020066-g007: Electrokinetic analysis of the bacterial electrophoretic mobilities.Quantitative analysis of the electrokinetic patterns of (A) E2152, (B) E2146, (C) E2302, (D) E2498. The charge density , the characteristic flow penetration length , the heterogeneity length scale and the thickness d of the soft bacterial component are indicated below. (A) E2152: nm. mM, nm, . (B) E2146: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (C) E2302: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (D) E2498: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. N.B.: Values for and are indicated with a precision of ±10% in relation with error bars of experimental data.

Mentions: In details, three types of behavior may be distinguished: that of the reference cell E2152, for which /µ/ and its increase with decreasing ionic strength is most significant; that of E2146 strain for which /µ/ is lowest at fixed concentration and its increase when lowering electrolyte concentration is least pronounced; and finally an intermediate behavior which is that of the strains E2302 and E2498 for which /µ/ significantly increases when decreasing ionic strength from 100 mM to about 10 mM, and subsequently levels off for concentrations lower than 10 mM. Analysis of the data was carried out according to a methodology detailed elsewhere [65] on the basis of the theory outlined in previous section. For that purpose, the length scale associated to the thickness of surface appendage and/or cell wall was estimated from nanomechanical AFM analysis of the bacteria in 1 mM KNO3 solution, i.e. from the spatial range where non-linearity between loading force and nano-indentation applies, thus providing d∼25 nm for E2152, d∼40 nm for E2498 and d∼100 nm for E2146 and E2302. It is indeed under such low ionic strength conditions that deformation of the very bacterial soft component upon compression/indentation by the AFM tip is least impeded by a possible concomitant deformation of bulk cytoplasm [65]. This important element, further commented in the Discussion section, is supported by the large Turgor pressure of the cells obtained in 1 mM solution as compared to that determined in 100 mM salt concentration, in line with results and observations by Gaboriaud et al. [65]. Analysis of the data are given in Fig 7 for each strain, recalling that the electrokinetic parameters ρ0 (expressed below as a volume concentration of charges) and λ0 are jointly determining the slope and magnitude of the mobility as a function of ionic strength in the high salt concentration regime, while interphasial heterogeneity of the bacterium (parameter α) may further come into play at sufficiently low ionic strengths as a result of possible swelling of the cell wall/surface appendage. The reader is referred to [68] for further details. The legend of Fig 7 collects the magnitude of the electrokinetic parameters obtained for the various bacteria, having in mind that values are given with a precision of±10%, in relation with the error bars of experimental data. Remarkably, data for the reference bacteria E2152 could be reproduced by theory over the whole range of ionic strength examined in this study. The charge density for this strain is high (ρ0 = −170 mM) as compared to that obtained for other gram-negative bacteria like Shewanella [65] and is in good agreement with that determined by Sanohara et al. for E. coli [101]. The analysis suggests that it is not necessary to introduce any interphasial heterogeneity of the bacteria for adequately reproducing electrokinetic behavior at low ionic strengths. The rate of increase of /µ/ upon decrease of ionic strength may be indeed solely attributed to a significant polarization of the electric double layer around/within the bacteria, as expected for such large values of charge density [68]. Also, an hydrodynamic penetration length 1/λ0 of about 0.7 nm is determined, which denotes a poor intrusion of the electroosmotic flow within the thin cell wall that serves as only soft component for the reference strain devoid of type 1 fimbriae, F-pili or Ag43 protein. For bacteria with type 1 fimbriae (E2146), we obtain significantly lower charge density (ρ0 = −30 mM) and larger hydrodynamic penetration length (1/λ0 = 1.7 nm) than for the E2152 strain. This indicates that the charges responsible for the motion of the particles upon action of the electric field are either present in lower number or distributed over larger volume than for E2152 cells and that the supporting soft structure is significantly permeable, i.e. inhibits flow penetration to a lower extent than within the cell wall surrounding the reference bacteria E2152. Additionally, mobility values for ionic strengths lower than ∼20 mM are compatible with an increase of α, i.e. a heterogeneous extension of the bacterial appendage by electrostatically-driven swelling processes connected to repulsive interactions between neighboring charges carried by the fimbriae. For the E2498 strain covered by Ag43 protein layer, electrokinetic data are excellently reproduced in the ionic strength range 5 mM to 100 mM adopting the values ρ0 = −170 mM and 1/λ0 = 0.6 nm. For lower ionic strengths, the slight decrease in /µ/ with decreasing ionic strengths is interpreted by a swelling of the outer edge of the protein layer, as judged by the increase in length scale that allows the recovering of data points collected at KNO3 concentrations lower than 5 mM. Similarly to E2152, the set of electrokinetic parameters (ρ0, λ0) obtained for E2498 typically pertains to a soft structure that is rather compact, highly charged, and poorly permeable. Finally, despite the proven presence of F-pili at the surface of E2302 strain, the dependence and magnitude of /µ/ with ionic strength depicts features that are surprisingly similar to those obtained for the strain E2498. It was systematically verified that an analysis taking into account a thickness d∼100 nm for these F-pili rendered impossible any appropriate fit of the data at large ionic strengths and further systematically required a charge density and hydrodynamic penetration length of magnitudes ρ0∼−160 mM, 1/λ0∼0.6 nm that mark the presence of compact soft structures (see values derived for E2152 and E2498) rather than that of the expected long, flexible filaments (see values obtained for E2146). An explanation for this counterintuitive result is inferred upon closer inspection of the AFM images detailed in Fig 3. Indeed, in the case of E2146, type 1 fimbriae clearly constitute a continuous polymeric layer surrounding the cell membrane whereas for E2302, the number of observed F-pili is extremely low (to a maximum of 5) and by no means may be assimilated to a polymeric layer of which electrohydrodynamic properties can be tackled on the basis of a mean field model. In other words, the quantities ρ0 and 1/λ0 derived for E2302 suggest that the electrokinetic properties for these bacteria are mainly governed by the cell membrane supporting the F-pili rather than by the F-pili themselves which are either too scarce for generating any predominant contribution to the overall mobility of the bacteria, or completely retracted along the cell wall, thereby covering heterogeneously a insignificant spatial fraction of the overall bacterial surface. This is supported by the excellent quantitative recovering of the electrokinetic data for E2302 at ionic strengths larger than 10 mM, taking for the cell surface appendage thickness the value d∼25 nm, i.e. that associated to the cell wall thickness for the reference strain E2152. For salt concentration below 10 mM, data measured for E2302 are further in line with a heterogeneous extension of the bacterial soft interphase, which is probably associated to a stretching of the few F-pili surrounding the cell and/or that of the supporting cell membrane.


Bacterial surface appendages strongly impact nanomechanical and electrokinetic properties of Escherichia coli cells subjected to osmotic stress.

Francius G, Polyakov P, Merlin J, Abe Y, Ghigo JM, Merlin C, Beloin C, Duval JF - PLoS ONE (2011)

Electrokinetic analysis of the bacterial electrophoretic mobilities.Quantitative analysis of the electrokinetic patterns of (A) E2152, (B) E2146, (C) E2302, (D) E2498. The charge density , the characteristic flow penetration length , the heterogeneity length scale  and the thickness d of the soft bacterial component are indicated below. (A) E2152: nm. mM, nm, . (B) E2146: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (C) E2302: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (D) E2498: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. N.B.: Values for  and  are indicated with a precision of ±10% in relation with error bars of experimental data.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3105017&req=5

pone-0020066-g007: Electrokinetic analysis of the bacterial electrophoretic mobilities.Quantitative analysis of the electrokinetic patterns of (A) E2152, (B) E2146, (C) E2302, (D) E2498. The charge density , the characteristic flow penetration length , the heterogeneity length scale and the thickness d of the soft bacterial component are indicated below. (A) E2152: nm. mM, nm, . (B) E2146: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (C) E2302: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. (D) E2498: nm. mM, nm, (curve a) , (curve b) nm, (curve c) nm, (curve d) nm. N.B.: Values for and are indicated with a precision of ±10% in relation with error bars of experimental data.
Mentions: In details, three types of behavior may be distinguished: that of the reference cell E2152, for which /µ/ and its increase with decreasing ionic strength is most significant; that of E2146 strain for which /µ/ is lowest at fixed concentration and its increase when lowering electrolyte concentration is least pronounced; and finally an intermediate behavior which is that of the strains E2302 and E2498 for which /µ/ significantly increases when decreasing ionic strength from 100 mM to about 10 mM, and subsequently levels off for concentrations lower than 10 mM. Analysis of the data was carried out according to a methodology detailed elsewhere [65] on the basis of the theory outlined in previous section. For that purpose, the length scale associated to the thickness of surface appendage and/or cell wall was estimated from nanomechanical AFM analysis of the bacteria in 1 mM KNO3 solution, i.e. from the spatial range where non-linearity between loading force and nano-indentation applies, thus providing d∼25 nm for E2152, d∼40 nm for E2498 and d∼100 nm for E2146 and E2302. It is indeed under such low ionic strength conditions that deformation of the very bacterial soft component upon compression/indentation by the AFM tip is least impeded by a possible concomitant deformation of bulk cytoplasm [65]. This important element, further commented in the Discussion section, is supported by the large Turgor pressure of the cells obtained in 1 mM solution as compared to that determined in 100 mM salt concentration, in line with results and observations by Gaboriaud et al. [65]. Analysis of the data are given in Fig 7 for each strain, recalling that the electrokinetic parameters ρ0 (expressed below as a volume concentration of charges) and λ0 are jointly determining the slope and magnitude of the mobility as a function of ionic strength in the high salt concentration regime, while interphasial heterogeneity of the bacterium (parameter α) may further come into play at sufficiently low ionic strengths as a result of possible swelling of the cell wall/surface appendage. The reader is referred to [68] for further details. The legend of Fig 7 collects the magnitude of the electrokinetic parameters obtained for the various bacteria, having in mind that values are given with a precision of±10%, in relation with the error bars of experimental data. Remarkably, data for the reference bacteria E2152 could be reproduced by theory over the whole range of ionic strength examined in this study. The charge density for this strain is high (ρ0 = −170 mM) as compared to that obtained for other gram-negative bacteria like Shewanella [65] and is in good agreement with that determined by Sanohara et al. for E. coli [101]. The analysis suggests that it is not necessary to introduce any interphasial heterogeneity of the bacteria for adequately reproducing electrokinetic behavior at low ionic strengths. The rate of increase of /µ/ upon decrease of ionic strength may be indeed solely attributed to a significant polarization of the electric double layer around/within the bacteria, as expected for such large values of charge density [68]. Also, an hydrodynamic penetration length 1/λ0 of about 0.7 nm is determined, which denotes a poor intrusion of the electroosmotic flow within the thin cell wall that serves as only soft component for the reference strain devoid of type 1 fimbriae, F-pili or Ag43 protein. For bacteria with type 1 fimbriae (E2146), we obtain significantly lower charge density (ρ0 = −30 mM) and larger hydrodynamic penetration length (1/λ0 = 1.7 nm) than for the E2152 strain. This indicates that the charges responsible for the motion of the particles upon action of the electric field are either present in lower number or distributed over larger volume than for E2152 cells and that the supporting soft structure is significantly permeable, i.e. inhibits flow penetration to a lower extent than within the cell wall surrounding the reference bacteria E2152. Additionally, mobility values for ionic strengths lower than ∼20 mM are compatible with an increase of α, i.e. a heterogeneous extension of the bacterial appendage by electrostatically-driven swelling processes connected to repulsive interactions between neighboring charges carried by the fimbriae. For the E2498 strain covered by Ag43 protein layer, electrokinetic data are excellently reproduced in the ionic strength range 5 mM to 100 mM adopting the values ρ0 = −170 mM and 1/λ0 = 0.6 nm. For lower ionic strengths, the slight decrease in /µ/ with decreasing ionic strengths is interpreted by a swelling of the outer edge of the protein layer, as judged by the increase in length scale that allows the recovering of data points collected at KNO3 concentrations lower than 5 mM. Similarly to E2152, the set of electrokinetic parameters (ρ0, λ0) obtained for E2498 typically pertains to a soft structure that is rather compact, highly charged, and poorly permeable. Finally, despite the proven presence of F-pili at the surface of E2302 strain, the dependence and magnitude of /µ/ with ionic strength depicts features that are surprisingly similar to those obtained for the strain E2498. It was systematically verified that an analysis taking into account a thickness d∼100 nm for these F-pili rendered impossible any appropriate fit of the data at large ionic strengths and further systematically required a charge density and hydrodynamic penetration length of magnitudes ρ0∼−160 mM, 1/λ0∼0.6 nm that mark the presence of compact soft structures (see values derived for E2152 and E2498) rather than that of the expected long, flexible filaments (see values obtained for E2146). An explanation for this counterintuitive result is inferred upon closer inspection of the AFM images detailed in Fig 3. Indeed, in the case of E2146, type 1 fimbriae clearly constitute a continuous polymeric layer surrounding the cell membrane whereas for E2302, the number of observed F-pili is extremely low (to a maximum of 5) and by no means may be assimilated to a polymeric layer of which electrohydrodynamic properties can be tackled on the basis of a mean field model. In other words, the quantities ρ0 and 1/λ0 derived for E2302 suggest that the electrokinetic properties for these bacteria are mainly governed by the cell membrane supporting the F-pili rather than by the F-pili themselves which are either too scarce for generating any predominant contribution to the overall mobility of the bacteria, or completely retracted along the cell wall, thereby covering heterogeneously a insignificant spatial fraction of the overall bacterial surface. This is supported by the excellent quantitative recovering of the electrokinetic data for E2302 at ionic strengths larger than 10 mM, taking for the cell surface appendage thickness the value d∼25 nm, i.e. that associated to the cell wall thickness for the reference strain E2152. For salt concentration below 10 mM, data measured for E2302 are further in line with a heterogeneous extension of the bacterial soft interphase, which is probably associated to a stretching of the few F-pili surrounding the cell and/or that of the supporting cell membrane.

Bottom Line: Additionally, for a given surface appendage, the magnitude of the nanomechanical parameters decreases significantly when increasing bulk salt concentration.This effect is ascribed to a bacterial exoosmotic water loss resulting in a combined contraction of bacterial cytoplasm together with an electrostatically-driven shrinkage of the surface appendages.Altogether, AFM and electrokinetic results clearly demonstrate the intimate relationship between structure/flexibility and charge of bacterial envelope and propensity of bacterium and surface appendages to contract under hypertonic conditions.

View Article: PubMed Central - PubMed

Affiliation: Laboratoire de Chimie Physique et Microbiologie pour l'Environnement, Nancy Université, CNRS UMR7564, Villers-lès-Nancy, France. gregory.francius@lcpme.cnrs-nancy.fr

ABSTRACT
The physicochemical properties and dynamics of bacterial envelope, play a major role in bacterial activity. In this study, the morphological, nanomechanical and electrohydrodynamic properties of Escherichia coli K-12 mutant cells were thoroughly investigated as a function of bulk medium ionic strength using atomic force microscopy (AFM) and electrokinetics (electrophoresis). Bacteria were differing according to genetic alterations controlling the production of different surface appendages (short and rigid Ag43 adhesins, longer and more flexible type 1 fimbriae and F pilus). From the analysis of the spatially resolved force curves, it is shown that cells elasticity and turgor pressure are not only depending on bulk salt concentration but also on the presence/absence and nature of surface appendage. In 1 mM KNO(3), cells without appendages or cells surrounded by Ag43 exhibit large Young moduli and turgor pressures (∼700-900 kPa and ∼100-300 kPa respectively). Under similar ionic strength condition, a dramatic ∼50% to ∼70% decrease of these nanomechanical parameters was evidenced for cells with appendages. Qualitatively, such dependence of nanomechanical behavior on surface organization remains when increasing medium salt content to 100 mM, even though, quantitatively, differences are marked to a much smaller extent. Additionally, for a given surface appendage, the magnitude of the nanomechanical parameters decreases significantly when increasing bulk salt concentration. This effect is ascribed to a bacterial exoosmotic water loss resulting in a combined contraction of bacterial cytoplasm together with an electrostatically-driven shrinkage of the surface appendages. The former process is demonstrated upon AFM analysis, while the latter, inaccessible upon AFM imaging, is inferred from electrophoretic data interpreted according to advanced soft particle electrokinetic theory. Altogether, AFM and electrokinetic results clearly demonstrate the intimate relationship between structure/flexibility and charge of bacterial envelope and propensity of bacterium and surface appendages to contract under hypertonic conditions.

Show MeSH
Related in: MedlinePlus