We investigated the molecular basis of interaction between E. coli
cells and metal substrates presenting different surface nanotopographies (namely, flat and nanorough gold). For substrates fabrication, we exploited a method discussed elsewhere
]. Briefly, we used a spontaneous galvanic replacement reaction (SGDR), which allows metal deposition in the absence of an external reducing agent
]. This electroless plating approach is fairly cheap, highly reproducible, and enables the fabrication of metal films with highly controlled surface topographies that are uniform over wide areas. Representative SEM images of flat and nanorough gold substrates are reported in Figure . In particular, as shown in Figure , a homogeneous, flat gold film was used as reference substrate. On the other side, the rough Au film (Figure ) obtained by SGDR shows a randomly organized and uniform nanoroughness, presenting hollow/porous nanostructures that are regularly extended over a wide area. We also carried out atomic force microscopy (AFM) characterizations of the substrates. In particular, the AFM line profiles (Figure , bottom) illustrate that, while flat gold surfaces display a clear smooth profile (with a Ra
< 1 nm), the nanostructured Au surfaces have an average roughness profile of c.a.
100 nm. These substrates were exploited to investigate the early stage of E. coli
adhesion capability, focusing on the possible activation of specific bio-molecular pathways.
Substrate characterizations. SEM investigation of flat (A) and nanorough (B) gold substrates, with their respective AFM line profiles at the bottom part of the pictures.
We first performed a preliminary counting experiment of bacteria growing onto the two distinct substrates in order to verify whether nanotopography may affect the number of adhering bacteria. Then, we carried out morphological investigations, by AFM, to detect any phenotypical changes of microorganisms upon interaction with the different nanotopographies. Experimental data show that, although surface nanoroughness does not directly influence the adhesion capability of E. coli
cells in terms of total number of adherent cells (Figure ,B), it significantly impacts their adaptation, morphology, and physiology (Figure ,D). In particular, the bacterial colonization on the abiotic surfaces was not found to rely on nanoscale changes in surface nanoroughness as the average density of adherent bacteria was practically the same in the two samples (Figure ,B). It should be mentioned that, in this case, the two substrates display similar wettability properties, so such surface parameter does not play a significant role in the bacterium/substrate interaction. In fact, albeit nanorough gold surfaces are fairly hydrophilic (with a static water contact angle(WCA) of approximately 25°) as compared to their flat counterpart (static WCA of approximately 85°), after incubation with the bacterial culture medium, both nanotopographies acquire a rather hydrophilic character (WCA of approximately 10° and 30° for nanorough and flat samples, respectively). This is due to the adsorption of medium proteins onto the gold surfaces, which leads to a variation in the wettability properties of the substrates toward hydrophilicity, regardless of the original surface properties
]. Consistent with the literature, our results confirm the contrasting data regarding the influence of surface nanotopography and wettability on bacterial adhesion (e.g., the reported increase or decrease in the number of adherent bacteria as a function of surface nanoroughness). This suggests that a general explanation or theory about the adhesion mechanism is not feasible since bacterial interaction and persistence on abiotic surfaces are strongly dependent on the specific physicochemical properties of the substrates employed as well as on the bacterial strains used (e.g., Gram-positive or Gram-negative) and their growth conditions (i.e., incubation time, growth medium, ionic strength of the medium, temperature, shaking/flowing or static incubations).
Figure 2 Impact of nanoroughness on the morphology of E. coli. Confocal microscopy images of DAPI-stained bacteria growing onto flat (A) and nanorough (B) gold substrates revealing that the total number of bacteria is almost the same between the two samples. SEM (more ...)
Notably, Figure ,D shows that the population of E. coli
adhering onto nanostructured surfaces underwent an important phenotypical change with respect to those adhering onto flat films. Specifically, the SEM investigations illustrate that E. coli
growing onto flat gold film strongly adhered onto the surface, as demonstrated by the presence of the type-1 fimbriae. Such structures are, in fact, adhesive organelles that bacteria employ to contact and robustly interact with both host cells and abiotic surfaces
]. They also promote biofilm formation and development
]. On the contrary, bacteria attached onto nanostructured surfaces did not phenotypically display type-1 fimbriae, thus suggesting a weak interaction with the surfaces. This latter finding highlights that, although the total number of adherent bacteria is roughly the same, E. coli
cells growing onto nanostructured substrates exhibit the typical features of cells that are not able to make a correct and strong interaction with the surface. In a previous study, we found out that nanotopography may induce important changes in fimbrial expression, mainly related to the over-expression of one fimbrial operon repressor, namely LrhA; the detailed molecular activity of LrhA, however, has not been completely clarified yet
]. In this work, we aimed at uncovering the molecular mechanisms underlying fimbrial expression as a function of surface-related physical stimuli as well as to understand the molecular bases of bacterium/abiotic substrate interaction at the interface in the early stage of adhesion event. In particular, we incubated E. coli
with the two different nanotopographies and investigated the expression level of several genes that are involved in fimbrial synthesis, inter- and intra-species communication, biofilm formation, response to stress stimuli, and adhesion to both host cells and abiotic surfaces. The results of RT-qPCR of bacteria growing onto nanorough surfaces, compared to the reference flat substrate, are illustrated in Figure .
Figure 3 Results of RT-qPCR. RNA expression level (%), by RT-qPCR, of E. coli growing onto flat (left histograms) and nanorough (right histograms) substrates. All data relative to PCR experiments were analyzed by a statistical software to evaluate the significant (more ...)
Notably, we found a significant over-expression of cpxP
genes, which are involved in the bacterial envelope stress response, named as Cpx two-component system
]. This pathway is activated by the presence of large amounts of misfolded fimbrial protein aggregates, which are associated with the inner membrane. In particular, the periplasmic fimbrial misfolded subunits titrate cpxP
and further activate cpxA
; the latter then shifts its own phosphatase activity to a kinase and autokinase activity, leading to an accumulation of a phosphorylated transcription factor CpxR in the cytoplasm. This protein activates the expression of envelope folding and degrading factors, including dsbA
. However, as indicated by our data, bacteria growing onto nanorough Au surface do not up-regulate the periplasmic protein disulfide isomerase dsbA
, which is involved in protein quality control and refolding processes. On the other side, the over-expression of degP
suggests that E. coli
cells prefer to shift their molecular activity on removing misfolded proteins in the periplasmic space by degrading them, instead of trying to refold them, most probably because of the high presence of extremely unfolded/damaged proteins. Moreover, we found that bacteria growing onto nanostructured gold substrates over-express the fimE
encodes for a recombinase protein involved in the on-to-off fimbrial switching (i.e., FimE), leading bacteria to repress the type-1 fimbrial synthesis under particular conditions
]. These data are in good agreement with the SEM investigation of Figure and better explain also our previous findings
Taking into account all these data, it is likely that E. coli
adhering onto nanostructured gold substrates undergo a general stress condition, which results in two distinct biological responses: (1) The two-component system Cpx pathway ‘senses’ the external stimulus (i.e., the nanoscale variation of surface roughness) by detecting periplasmic and/or external misfolded proteins (thanks to the cpxP
recruitment), including the fimbrial subunits; as a consequence, bacteria activate the degP
-related degradation of fimbrial proteins for the recycle of amino acids. (2) FimE recombinase is over-expressed, which switches off the fimbrial operon, thus inhibiting the transcription of all the fimbrial subunits. As a result, bacteria adhering onto nanorough gold substrates repress the fimbrial transcription and, at the same time, degrade the fimbrial protein subunits, which are present in the periplasmic space. The scheme in Figure summarizes the possible molecular mechanisms involved in the bacterium/nanotopography interaction. This is also consistent with our previous proteomic data, in which some proteins involved in general stress response were found to be up-regulated in E. coli
attached onto rough substrates
Figure 4 Scheme of the molecular mechanism of bacteria/nanorough substrate interaction. (A) E. coli growing on flat gold substrates present the typical type-1 fimbriae; the two-component system Cpx pathway is inactive, thus CpxA posses a phosphatase activity which (more ...)
We also found an up-regulation of luxS
gene in the nanorough samples. Such gene is involved in the biosynthesis of a quorum sensing (QS) autoinducer molecule (AI-2), which has been demonstrated as a universal signal that could be used by a variety of bacteria for communication, also among different species
]. QS molecules are used by microorganisms to coordinate the gene expression also of the surrounding community, thus enabling bacteria to behave like a quasi
complex multicellular organism. This phenomenon occurs when bacteria have to overcome some environmental difficulties; in our case, such stress condition is represented by the nanotextured substrates.
On the other hand, the ompC
gene, which codifies for the outer membrane porin C, lpxC
, which is required for lipid A expression, and murA
, which is important for external wall synthesis, are not regulated upon interaction with the nanostructured substrates. Also, the fliC
gene that codifies for a flagella subunit, as well as cpxR
, which is an effector of the two-component system CpxR-A pathway, is not regulated in the treated samples. In this respect, we can envisage that, although nanostructured Au substrates strongly impact the bacterial adhesion capability, the genes codifying for the biofilm expression
] seem to be unregulated in the early stage of the adhesion event. Further and more systematic studies are required in order to evaluate any possible influence of nanotopographies on biofilm formation after longer incubation periods. On the other hand, our data suggest that the mechanosensing machinery of E. coli
feels the change in surface nanotopography as a physical stress signal. Hence, the bacteria focus their molecular activities on regulating and triggering specific pathways, which are important for recovery from stress conditions.