Revealing Silver Effects on Bacterial Cell Structure by Atomic Force and Scanning Electron Microscopies

The use of silver (Ag + ) as an antimicrobial under different forms and at different 23 scales, appears in numerous applications such as in health care, food industry, clothing, 24 fabrics and disinfectants. Yet, there is still important gaps regarding the complete 25 comprehension of the mechanisms of its actions on bacteria. In a previous work 1 , we 26 demonstrated that, silver and copper severly damage membrane proteins involved in 27 photosynthesis and respiration in bacteria exposed to metal excess. Here, we are presenting 28 complementary data using AFM and SEM microscopies, that reveals (i) the drastic effects of 29 Ag + ions on the morphology and structure of cell membrane and (ii) the formation of Ag + 30 aggregates that adhere to the bacterial cell surface in Rubrivivax (R.) gelatinosus . Impacts of 31 Ag + ions on R. gelatinosus are compared to those on the most commonly studied bacteria 32 (Escherichia (E.) coli and Bacillus (B.) subtilis ), while considering the effect of culture grown 33 media on the modification of silver ions. Altogether, these results reveal other levels and 34 subtle aspects of Ag + toxicity to be taken into account in understanding the general 35 mechanisms of metal toxicity in bacteria.


39
In the context of antibiotic crisis, where lack of efficiency is linked to both multi-40 resistance bacteria and increasingly reduced offer of antibiotic solutions 2-5 , silver has been 41 considered as a very promising alternative proposal 4,6 . It is therefore, widely used in industry, 42 agriculture and health care as an antimicrobial agent 5,7,8 . It can be found as pure and/or alloy 43 with other metals, at different scale and forms. It inhibits, causes or enhances mechanisms of 44 bacterial death, disrupts growth or reduces bacterial biofilm proliferation 9,10 . However, its use 45 is not without danger, since excess of silver in both ions (Ag + ) and nanoparticles (Ag-NPs) 46 forms shows toxic impact toward all types of living cells. Yet, the underlying mechanisms 47 driving its toxicity and its impact on bacteria, are still under intense investigations 11-13 . 48 Unlike copper, iron or zinc for example 14 , silver is not an essential metal for cell 49 growth. In bacteria, iron or zinc enter the cell trough specific import systems. Silver ions 50 could diffuse and enter the cells through the outer membrane via nonspecific importers and 51 poison the membrane or the cytoplasm of the bacterium 15 . Ag + could also affect the 52 membrane integrity and damage the cell structure if it accumulates outside the cell, although 53 direct evidences are still missing. In general, bacterial cells are programmed to maintain and 54 to resist metal effects and toxicity. They have the ability to reach metal homeostasis by using 55 different defense systems. They can, for example, repress the import system, induce the 56 required detoxification enzymes including metal sequestration and/or active efflux 57 systems 16,17 . A dysfunction in the homeostasis system (import/efflux) of these metals can 58 cause physiological disorders in both prokaryotes and eukaryotes cells 18 . Many studies have 59 reported these damages 1,4,12,19 . For example, human cells (skin and lung cells) are easily 60 exposed to silver nitrate or silver oxide by touching and breathing 20 ; this exposure can cause 61 breathing problems, lung and throat irritation, stomach pain, and argyria 21,22 . In the green 62 algae Chlamydomonas reinhardtii, silver ions disrupt cellular metabolism and inhibit important functions such as photosynthesis 23,24 . In Escherichia (E.) coli, Bacillus (B.) subtilis 64 or Salmonella (S.) thyphimirium model bacteria, Ag + inhibits the growth by targeting various 65 metabolisms 25-28 .In photosynthetic bacterium only few studies have addressed metal toxicity 66 mechanisms and homeostasis of silver. Therefore, we used to tackle the effect of Ag + on 67 photosynthesis and respiration. 68 In a previous study 1  required for photosynthesis and respiration respectively. Moreover, our data showed that Ag + 74 has effects on the succinate dehydrogenase (SDH) complex in E. coli, but not in B. subtilis 1 . 75 In this study, we carried out an investigation, using Scanning Electron Microscope (SEM) and 76 Atomic Force Microscope (AFM), to provide data on the morphological and structural 77 changes that occur upon exposure of bacteria to silver ions due to metals influence. The aim 78 of these investigations, in this context, was to elucidate the influence of silver Ag + ions 79 outside the bacterial cells and essentially to reveal Ag + interaction with the cell membrane 80 surface. The results showed that high resolution images could discern and characterize the 81 detailed changes that occurred on the bacterial cell membrane after treatment with AgNO3. To 82 our knowledge, this presents the direct evidence of Ag + silver ions morphological damages to 83 R. gelatinosus cell membrane. Our data analyses showed drastic changes that increased with 84 the increasing of incubation time, while untreated samples remain unaffected. Ag + exposure 85 leads to i-the formation of irregular deep grooves vesicles on the cell surface, ii-to cell 86 membrane shrinking and iii-to structure breaking in the extreme case of extended exposure. observed using SEM, raising the questions of the media effect on silver precipitation, the 89 influence of cell membrane surface and eventually suggest a hypothetical scenario of cell 90 defense.

Consequences of Ag + exposure on cell surface and morphology revealed by SEM and AFM
93 microscopy 94 To assess the toxicity effect of Ag + on cell's morphology and structure, we imaged R. 95 gelatinosus cells that were exposed to 1 mM of AgNO3 and deposited using dip coating on 96 substrate for both SEM (on freshly cleaved graphite HOPG) and AFM (on freshly cleaved 97 Mica). We have prepared same samples, under same conditions, for both untreated and 98 treated cultures after 1 and 24 hours (h) incubation with AgNO3. Results showed that 99 untreated cells of R. gelatinosus have rod-shape and a relatively smooth surface with no 100 ruptures or swellings (Fig. S1)). The images, at the same scales, showed also that the there was no effect of different deposition procedure, nor drying technics, nor ambient air or 108 high vacuum. Membranes cells were continuous, and homogeneous. We did not observe any observed bacteria that appeared very bright (statistically one fourth, at that concentration of 116 cells and ions) (Fig. 1A). Higher magnification showed an impressive accumulation of 117 metallic aggregates on the cell surface ( Fig. 1B and Fig. S2). The size distribution of these 118 aggregates on the membrane was compared with the distribution size of the ones that can be 119 observed also around the cells (Fig. 1C). It seems, that during the first hour, the first stage was 120 silver precipitation to form these aggregates. The superposition of several normal distributions 121 of the aggregates size distribution suggested several, may be competitive, stochastic 122 nucleation processes 29,30 . However, we observed a difference between distribution on and 123 around the cell's membrane. Some of the aggregates around the cells (2 nd and 3 rd pics of 124 density profile) can be attributed directly to detached particles from the membranes. These 125 data raised questions related to the interaction of bacteria with silver ions in the medium. 126 First, since not all bacteria displayed these particles on their surface, this may suggest a 127 difference in the surface component (exopolysaccharides/ capsule) of our cells, and/or a 128 difference in the growth stage (dividing versus static or dying cells). Second, we might 129 consider that these particles arose from interaction of silver with components in the growth 130 media. Nevertheless, we can also speculate at that point, that precipitation of silver ions was 131 mediated by cells eventually as a first stage of a defence mechanism, since insoluble particles 132 could be more difficult to penetrate into the cell with increasing size of the particles. This was 133 suggested in Pseudomonas aeruginosa in which extracellular synthesis of nanoparticles was 134 also observed after exposure to cadmium or selenium 31-33 . This mechanism was suggested to 135 contribute to metal resistance in Pseudomonas. 136 Furthermore, the AFM images of untreated or exposed bacteria to AgNO3 for one 137 hour, confirmed the presence of silver aggregates on the cell's membrane (Fig. 2). The size 138 distribution corresponds to the one observed with the SEM. Aggregates on the cells were not 139 displaced by the scanning and showed important adhesion to the cell surface. Yet, we cannot exclude that some aggregates grown on cell surface can be detached during washing and 141 deposition on substrate processes. No membrane deformation, breaking or change in 142 hardness-softness were observed at that point. We imaged also, using AFM, bacteria cells 143 after extended incubation time (24 h) to AgNO3. Figure 3 shows the comparison between 144 untreated cells, and AgNO3 exposed bacteria after 24 h incubation. The figure shows, again 145 topography and peak force error images, but also longitudinal and transverse cross section 173 When , is equal to 3, this indicates a gaussian amplitude distribution, smaller than 3 174 means that the surface is flattening and higher than 3, that the surface has a lot of peaks.
175 Table 1, summarize these morphological parameters for the control and the exposed 176 bacteria to silver ions. Errors on the values are estimated around 10% for average height and 177 rugosities, and around 20% for 3 rd and 4 th moments.

178
The most important change was an important increasing of surface roughness, along 179 bacteria length axis, till the integrity of cell's membranes was not ensured anymore. It was 180 difficult to localise silver aggregates under these extended exposure condition. Cells with longer incubation time (> 24 h) showed greater damages (Fig. 4)  Samples were collected after incubation time and they were prepared to be imaged by SEM 194 and AFM, using same protocol for deposition and same substrate as above for R. gelatinosus.  This phenomenon was reported with E. coli cells exposed to carbon dots stress for instance 38 .

214
The imaging of B. subtilis exposed to silver ions showed that Ag + had less effect comparing to 215 R. gelatinosus and E. coli during the first hour of incubation (Fig. 6). Cell morphology differs 216 slightly from control one. However, we could observe a drastic effect after 24 h exposition to 217 silver. Apparent smaller size and irregular shape were the main morphological 218 transformations, as reported for E. coli cells exposed to silica nanoparticles 39.

219
Collectively, these results showed that silver has a strong impact on bacteria. It

Effect of culture media (malate medium versus LB medium) on the interaction of R.
251 gelatinosus with silver ions. 252 Given that accumulation of silver aggregates was observed for R. gelatinosus in

253
Malate medium, but not in E. coli or B. subtilis in LB, we asked whether these aggregates 254 could also form when R. gelatinosus was grown in LB medium. To answer this question, bacteria were grown as previously in Malate or LB medium. SEM imaging was used to 256 discriminate any effects between untreated or AgNO3 exposed cells in LB or Malate medium.

257
The images revealed several differences. First for unknown reason, R. gelatinosus cells grown 258 in LB were longer than those grown in Malate medium (Fig. S5). Second, the comparison 259 between exposed bacteria to silver showed that unlike malate case, bacteria grown in LB did 260 not exhibit any silver accumulation on their surface. Bright halos were observed around the 261 cells; likely, due to trace of the medium, but we did not observe any metallic aggregates on 262 the bacterial surface (Fig. 8)  there are present in both media, and before further and more detailed study, we decided to

295
To demonstrate rapidly this assumption, we carried the following experiment. 6 mM 296 of AgNO3 was directly added to the media, malate and LB, under same conditions. The 297 concentration was increased to enhance the suspected effects. We analysed then the optical 298 properties, using spectrophotometry, of the compounds to detect possible formation of silver 299 nanoparticles throw their well-known plasmonic response 50-52 . (Fig. 9) shows the optical 300 absorption for AgNO3 added to culture media (Malate vs LB) and for comparison with citrate 301 that is commonly used to reduce AgNO3 to produce Ag nanoparticles. Usually in citrate 302 assisted production of silver nanoparticles, borates are used instead of phosphates 53 . The            Images shows topological (left side) and peak force error (right side) of respectively untreated 685 bacteria (1 st line) and bacteria exposed to 1mM AgNO3 for 1 h (2 nd line) and 24 h (3 rd line).    Figure 1 SEM image at different scale (A) and their zoom (B) of R. gelatinosus bacteria, exposed to AgNO3 for 1 h, deposited on HOPG graphite as described previously. One can observe silver (bright dots) concentration on the outer membrane. In (C) the graph shows the size distribution (histogram and density) of metallic aggregates on cells surface and around the cells.

Figure 2
AFM image of R. gelatinosus deposited as described previously. Images shows topological i.e. height (A and C) and peak force error (B and D) of untreated bacteria (1st line) and bacteria exposed to AgNO3 for 1 h (2nd line). The pro le curves correspond to relative (difference with mean height) average height across (transversal (E)) and along (longitudinal (F)) bacteria cell axis for control bacteria as pro le references.

Figure 3
Images show topological, peak force error and pro le cross section analysis. AFM image of R. gelatinosus bacteria untreated (A) and exposed (C) to AgNO3 for 24 h. Relative average heights are, calculated compared to height mean value, along and across the bacteria cell's length axis (B and D).  Morphological changes on E. coli deposited as described previously. Images shows topological height (left side) and peak force error (right side) of respectively untreated bacteria (1st line) and bacteria exposed to AgNO3 for 1 h (2nd line) and 24 h (3rd line).

Figure 6
Morphological changes on B. subtilis bacteria deposited as described previously. Images shows topological (left side) and peak force error (right side) of respectively untreated bacteria (1st line) and bacteria exposed to 1mM AgNO3 for 1 h (2nd line) and 24 h (3rd line).  SEM images (and zoom) of R. gelatinosus bacteria grown on malate (A) and LB (B) media, exposed during 1 h to AgNO3.

Figure 9
Optical spectroscopy curves for Malate, LB and citrate-phosphate solution toward nanoparticles synthesis. The dashed lines are just an eye-guide around the curve maximum.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Tambosi2021SupplementaryFigures.pdf