3.1. PTU/ZnO Composites Have Anti-Adhesive Surface Properties
Given the importance it may have for cell adhesion, the first property of the PTU/ZnO composites that we chose to explore in this study was surface roughness. For this, we used two established techniques to characterize the surface materials, confocal laser-scanning microscopy (CLSM) and atomic force microscopy (AFM) [
51], and we measured the average roughness R
a of the PTU/ZnO composites and the PDMS specimen. The data obtained are presented in
Table 1.
The roughness analysis using CLSM was performed on images of 1 × 1.5 mm
2, while for AFM it was performed on images of 20 × 20 µm
2. Thus, the area scanned using CLSM was 3750 times larger than the area scanned using AFM, explaining the differences observed in the values between the two techniques. The first interesting point to note, is that while CLSM measurements gave a similar value for the roughness of PTU and PDMS, measurements at the micro-scale with AFM show an important difference between PDMS and PTU, with a roughness value of 13.6 ± 4.6 nm for PDMS and of 0.401 ± 0.075 nm for PTU. This difference may, indeed, be due to the size of the areas scanned in each case, but also to the different detection methods involved in the two techniques (CLSM, reflection of visible light; AFM, physical deflection of a cantilever). However, both techniques show the same tendency: while the addition of 1 wt.% of spherical ZnO particles tends to smoothen the surface, although the difference with PTU in both cases is not significant, increasing the amount of particles from 1 to 5 wt.% increases the roughness of the material. In the situation where tetrapodal particles are added to PTU, the roughness is slightly increased when 1 wt.% of particles were added compared to PTU; this increase is more pronounced with the addition of 5 wt.% of particles. It is important to note that in all cases, no Zn
2+ was found at the surface of the materials, neither by EDX analysis nor by Zn
2+ release measurements (
Supplementary Tables S1 and S2), which implies that the roughness differences observed are only influenced by the filling material, and not by the physical presence of ZnO particles at the surface. The literature on the influence of roughness on cell adhesion is conflicting. While some studies suggest that an
Ra smaller than 200 nm does not affect adhesion [
52], different authors showed differences in the adhesion of cells on nanorough surfaces (30 nm to 120 nm) [
20,
23]. However, as far as we know, the roughness in the sub-nanometer range has not been extensively investigated in the context of microbial adhesion. Thus, it is difficult to state at this stage whether the small roughness differences measured between the PTU variations will influence cell adhesion.
Another component considered in this study is the material stiffness. In material science, material stiffness is a basic parameter, most relevant in evaluating fundamental phenomena like adhesion, friction or crack propagation [
51,
53]. However, within the interdisciplinary field of bio-adhesion between abiotic surfaces and microorganisms, this issue has not yet been widely addressed, although it has been shown that the stiffness of a material can influence the adhesion of bacteria onto it [
39]. In the case of the PTU/ZnO composites produced in this study, the stiffness of the material variations was measured using tensile tests; the results are presented in
Table 2. They show that while the stiffness of PTU is five times higher than that of PDMS, there are no significant differences observed between the PTU variations, which is in accordance with previous measurements performed on other batches of PTU/ZnO materials produced in our team [
28]. Only PTU with 5 wt.% of t-ZnO shows a significant difference to the other PTU variations; this might be due to the geometric arrangement of the tetrapods in the material. If we put these results in light of the study by Song and Ren, then the adhesion to the stiffer material, namely PTU and PTU/ZnO variations, should be less favored by the microorganisms tested in this study.
The hydrophobicity of a material is also a crucial factor to consider when evaluating the adhesion of microorganisms to artificial surfaces [
54,
55,
56]. For instance, it has been already shown that microorganisms such as
S. aureus or
C. glabrata, used in this study, adhere more to hydrophobic surfaces, as do many other microbes [
17,
57,
58,
59,
60,
61]. Thus, we characterized the wetting properties of the PTU/ZnO variation compared to PDMS; the results are presented in
Table 3.
Figure S3 (Supplementary Materials) shows examples of the corresponding micrographs from which the measurements were performed. While PDMS is hydrophobic with a Water Contact Angle (WCA) of 114.9 ± 1.4°, PTU and PTU/ZnO variations are hydrophilic (WCA comprised between 78.6–73.3°). No significant differences can be observed among the PTU/ZnO variations. As well as for the increased stiffness, the hydrophilic properties of PTU/ZnO composites should also lead to the reduced adhesion of microbial cells.
Finally, the last parameter we measured to characterize the PTU/ZnO composites was their surface charge. Indeed, during the first phase of cell adhesion to an abiotic surface, electrostatic interactions are at play; alongside to van der Waals and hydrophobic forces, they mediate non-specific cell adhesion [
61]. For this reason, notably, the surface charge is one of the most studied cell wall properties of bacteria cells [
15]. The net surface charge of microbial cells in physiological medium is negative [
62]. To evaluate the strength of electrostatic interactions between
S. aureus respectively
C. glabrata and the PTU/ZnO compositesthe
ζ-potential of PDMS, PTU and PTU with 5 wt.% of spherical and tetrapodal ZnO particles was measured: the results are presented in
Table 4. The results show that all the surfaces are negatively charged; no significant differences can be observed between PTU and PTU functionalized by 5 wt.% ZnO, whether of spherical or tetrapodal filler shape. An important point to note, is that the PDMS surface is six times less charged than PTU: thus, on this surface there will be less repulsion with cells than for PTU and PTU variations, which may also have an important influence on cell adhesion. Indeed, as the cell wall of microbial cells is negatively charged, the repulsion forces will increase with increasing the negative charge of the material surface.
All these experiments have thus allowed us to characterize the properties of the PTU/ZnO composites investigated in this study, and to compare them with the ones of PDMS. Our analyses show that PTU/ZnO composites have a decreased roughness compared to PDMS, an increased stiffness, they have a lower WCA and are hydrophilic, and finally, they present a more negatively charged surface. Within the PTU variations, the only significant difference was found concerning the stiffness, where PTU with 5 wt.% of tetrapodal ZnO particles had an increased stiffness compared to the other PTU variations. Based on these findings, the potential of these materials to reduce microbial adhesion, atomic force microscopy was used to probe the interactions between microbial cells and the different material surfaces at the molecular scale.
3.2. Adhesion of S. aureus to PTU/ZnO Composites Is Significantly Decreased Compared to PDMS
To evaluate whether the PTU/ZnO polymer variations are efficient anti-fouling materials, single-cell force spectroscopy experiments were conducted first using S. aureus cells. In this type of experiment, a single living bacterial cell is attached to a colloidal cantilever and directly used, in force spectroscopy mode, to probe the interactions with the surface. This gives access to matrices of force–distance curves recorded over a micrometer-sized (1 × 1 µm²) area of the surface, from which different information, such as the maximum adhesion peak and the rupture distance, can be extracted. The maximum adhesion peak quantifies the maximum force of the interaction, and the rupture distance gives information on the length of the unfolded molecules at the surface of the cell, and thus, on the nature of the interaction probed. Moreover, the percentage of adhesion is determined by the number of force curves presenting adhesive events in each case.
The histograms of the results obtained are presented in
Figure 1. For each type of polymer composite and PDMS, the interactions with seven different cells, coming from at least three independent cultures are presented. The results obtained show that PDMS presents the highest average adhesion force of 726 ± 397 pN, as well as the highest average rupture distance of 212 ± 116 nm among the material variations, containing PTU/ZnO composites and PDMS (
Figure 1a). Compared to PDMS, PTU shows a significantly lower number of adhesive events (89.5% instead of 100% for PDMS), a lower average adhesion force with 237 ± 289 pN and a corresponding average rupture distance of 191 ± 146 nm (
Figure 1b). With the addition of ZnO particles into the PTU, the number of force curves presenting adhesive events and adhesion forces decrease, as well as the corresponding rupture distances. The minimum average adhesion force recorded, of 86 ± 120 pN, is reached for PTU with 5 wt.% t-ZnO (only 40.4% of adhesive curves,
Figure 1f). The average rupture distance for PTU with 5 wt.% t-ZnO is 103 ± 61 nm, which is comparable to PTU containing 5 wt.% s-ZnO. These results are also summarized in
Table 5. Statistical analysis (two-sample t-test) showed that the differences recorded in the adhesion forces are significantly different (
p-value < 0.001), with only one exception made by PTU with a filler content of 1 wt.% of t-ZnO, for which the adhesion force is not significantly different from PTU and the other PTU/ZnO composite variations. From these results it thus seems that PTU variations display anti-adhesive properties as the number of adhesive events and the adhesion force values are significantly decreased compared to PDMS. While this was known for classic polyurethane materials coatings [
63], it is the first time this has been shown for PTU. Moreover, the increasing amount of both types of ZnO particles present in the PTU further decreases both the frequency of adhesion of
S. aureus; while in the case of s-ZnO particles, the adhesion force does not significantly decrease with the increasing amount of particles, in the case of t-ZnO, the adhesion force is reduced by a factor of 2.2. Therefore, it seems the tetrapodal shape of ZnO particles is more efficient at decreasing the adhesion of cells. The significant decrease in the adhesion forces indicate the reduced bond-strengthening on PTU/ZnO composites; thus, irreversible adhesion is weaker. This proves the suitability of PTU/ZnO composites as anti-adhesive surfaces in the case of
S. aureus.
The rupture distances obtained for PTU/ZnO composites are rather long and indicate that the interactions between the cells and the surfaces do not rely only on physico-chemical forces, but also on specific interactions involving specific cell-surface polymers. Indeed, when looking at the retract force curves obtained, multiple peaks can be observed after the contact point, thus proving that molecules at the surface of cells interact with the material and are unfolded upon retraction. Such molecules could be surface proteins or strands of peptidoglycan present on the surface of Gram-positive bacteria such as
S. aureus, interacting with the polymer surface and thus strengthening the initial physico-chemical bonds between cells and surfaces, to transition to an irreversible adhesion [
5].
The reasons explaining the decrease in adhesion force can be directly correlated to the characteristics of the surfaces described earlier. The first reason, is that PTU variations, compared to PDMS, are hydrophilic (for details see
Table 3). Indeed, several studies have shown that an increased hydrophobicity of the material was responsible for a high level of bacterial initial binding to the surface [
56,
64]. A study conducted using
S. epidermidis even showed that an increase in the degree of hydrophobicity of materials was linearly correlated with the number of adherent bacterial cells on the surface [
57]. The propensity of
S. aureus cells to adhere more to a hydrophobic surface such as PDMS compared to the hydrophilic PTU/ZnO variations, may be due to the presence of hydrophobic components at its cell surface, such as proteins also known as adhesins, which are important factors promoting strong adhesion to hydrophobic surfaces [
58,
65]. Furthermore, Maikranz et al. recently showed that
S. aureus binds with many weakly binding macromolecules to hydrophobic surfaces, while only a few selected but strong macromolecules bind to hydrophilic surfaces. It is also suggested that the hydrophobic molecules form rather fast bonds while the hydrophilic binding molecules have to overcome a potential barrier [
66], which may explain the decreased adhesion of cells on hydrophilic surfaces such as the PTU/ZnO composites produced in this study. Another reason that could explain the differences between PDMS and the PTU/ZnO composites is the surface charge of the material. The cell surface of
S. aureus is moderately negatively charged (−4 to 6 mV) [
67,
68], thus, slightly negatively charged surfaces could be colonized by
S. aureus cells, as long as van der Waals forces or other forces overcome the repulsion [
69]. Furthermore, a study by Gross et al. suggests that increasing the repulsive forces between
S. aureus and the polymer surface might complicate biofilm formation [
70]. Thus, the fact that PTU/ZnO composites have a more negatively charged surface is also an important factor that can notably explain the smaller percentages of adhesion forces on these surfaces compared to PDMS. Another factor that seems to influence cell adhesion in the case of
S. aureus is the stiffness of the materials. Only a few studies have reported on the influence of material stiffness on cell adhesion, but no consistent trend has been found on the subject [
71]. For example, Lichter et al. found that
S. epidermidis adhered better to stiff polyelectrolyte multilayered thin films [
72], while Wang et al. found that the adhesion of
S. aureus was reduced on polyacrylamide hydrogels with increasing stiffness [
73]. In our case, we show that adhesion to stiffer materials is reduced, as the adhesion force recorded on PDMS is much higher than the one recorded on PTU variations. The stiffnesses of the PTU variations are not significantly different from each other except in the case of PTU containing 5 wt.% of t-ZnO, which is stiffer than the other PTU/ZnO composites and where the adhesion force recorded is significantly lower (86 ± 120 pN). While it is difficult to evaluate the independent effects of each parameter (charge, wetting property and stiffness) on the cell adhesion, in this case, we could suggest the hypothesis that at similar charge and WCA, an increased stiffness of the material might reduce the adhesion of cells onto it.
The same type of experiment was conducted with
C. glabrata. Given the large size of these cells compared to
S. aureus (five times larger), they were directly attached to tipless cantilevers and used to probe the interactions with the different material variations. The adhesion force and rupture histograms obtained are presented in
Figure 2. As for
S. aureus, in this case PDMS also shows the highest adhesion force of 11,164 ± 5387 pN and the highest rupture distance of 1187 ± 424 nm among the material variations (
Figure 2a), including PDMS and PTU/ZnO composites. In comparison, pure PTU shows a significantly lower average adhesion force with 4881 ± 1744 pN and a corresponding rupture distance of 825 ± 404 nm (
Figure 2b). With addition of 1 wt.% of t-ZnO and s-ZnO, the average adhesion force and rupture distance values only change slightly, whereas the addition of 5 wt.% of s-ZnO raises the adhesion force to 4909 ± 1733 pN. These results are summarized in
Table 6. Statistical tests showed that the adhesion forces are significantly different from each other (
p-value < 0.01), with an exception made by PTU with a filler content of 5 wt.% of s-ZnO, which is not significantly different from pure PTU. Thus, the results that are obtained in the case of
C. glabrata show the same tendency as the ones obtained with
S. aureus: the adhesion of the cells is strongly decreased on all PTU variations compared to PDMS. However, for
C. glabrata, increasing the amount of spherical ZnO particles present in PTU does not seem to further decrease cell adhesion, but it does in the case of PTU filled with 5 wt.% of t-ZnO, where the lowest adhesion force is recorded. This was already the case for
S. aureus. Moreover, it is interesting to note that compared to
S. aureus, the amount of force curves presenting adhesive events for all investigated polymer surfaces was 100%, and that both adhesion forces and rupture distance values were higher for
C. glabrata. This is probably due to the increased size of cells (approximately 5 µm in diameter), which, thus, increases the surface contact between the cells and the material, thus explaining the higher forces recorded. In this case, the long rupture distances measured and the presence of multiple peaks located after the contact point on the retract force curves also show that specific bonds are at play in the interaction. Here, strengthening the initial physico-chemical bonds also resulted in a transition from reversible to irreversible attachment [
5].
The reasons for these lower adhesion values on PTU and PTU/ZnO variations compared to PDMS are probably the same as discussed previously in the case of
S. aureus, i.e., lower water contact angle, higher negatively charged surface and higher stiffness of the PTU/ZnO composites. In particular, in the case of yeast cells, a study by El-Kirat-Chatel et al. demonstrated that the hydrophobic surface chemistry of the substrates plays an important role in the adhesion of the
C. glabrata [
60]. It shows that the adhesion of
C. glabrata is mostly promoted by adhesins binding to hydrophobic surfaces. Which means that
C. glabrata adheres more strongly to hydrophobic surfaces. This is in line with our findings. Regarding the surface charge of the materials,
C. glabrata was found to bear a surface
ζ-potential of −24 mV [
74]. Thus, in this case, the repulsion of
C. glabrata may also be more pronounced on PTU/ZnO composite surfaces than on PDMS, which has a smaller negative surface charge. Thus, perhaps explaining in part the reduced adhesion forces, although in this case, the percentage of force curves presenting adhesive events is still 100% for PTU/ZnO variations. Finally, regarding the stiffness of the materials, as far as we are aware, no studies have reported its influence on yeast cell adhesion. For PTU/ZnO composites, the adhesion force recorded is the lowest in the case of PTU with a filler content of 5 wt.% of t-ZnO particles. While the surface charge and the water contact angle are similar for all PTU/ZnO composites, the stiffness is significantly different for PTU/ZnO with 5 wt.% of t-ZnO. Thus, also in this case, it seems that stiffness is an important parameter to control the adhesion of microbes.