Microbial repellence properties of engineered spider silk coatings prevent biofilm formation of opportunistic bacterial strains

Bacterial infections are well recognised to be one of the most important current public health problems. Inhibiting adhesion of microbes on biomaterials is one approach for preventing inflammation. Coatings made of recombinant spider silk proteins based on the consensus sequence of Araneus diadematus dragline silk fibroin 4 have previously shown microbe-repellent properties. Concerning silicone implants, it has been further shown that spider silk coatings are effective in lowering the risk of capsular fibrosis. Here, microbial repellence tests using four opportunistic infection-related strains revealed additional insights into the microbe-repellent properties of spider silk-coated implants, exemplarily shown for silicone surfaces.


Introduction
Multidrug resistant bacteria are recognised as one of the greatest threats to human health worldwide. [1] Further, rising rates of antibacterial resistance impact all aspects of modern medicine and threaten the effectiveness of many medical treatments in cancer care, transplantation and surgical procedures among others. [2] Staphylococcus aureus is probably one of the most prevalent, dangerous and clinically relevant bacteria due to its methicillin resistance. [3,4] It is a leading cause of bacteremia and infective endocarditis and already spread beyond the confines of health-care facilities, emerging as a new threat in the community and becoming a dominant pathogen. [5,6] Apart from S. aureus, other pathogens cause increasingly inflammatory problems. Brevundimonas diminuta found in infections of intravascular catheters are opportunistic pathogens affecting patients suffering from several diseases. [7,8] Ralstonia species are gram-negative bacilli that have increasingly been recognised as emerging nosocomial pathogens, particularly in immune-compromised hosts. Ralstonia pickettii is clinically the most important pathogen from the Ralstonia genus. Nosocomial outbreaks of R. pickettii infections are often based on contaminated medical solutions, including saline, sterile water, as well as disinfectants, and there have been case reports of various invasive infections [9,10] causing meningitis, infective endocarditis, nosocomial pneumonia, central line-associated bloodstream infection, [11] and chronic haemodialysis linked to contamination of the dialysis water. [12] Propionibacterium acnes is a microaerophilic, gram-positive and rod-shaped bacterium that resides in the pilosebaceous follicles of the human skin. [13] It is a low virulence opportunistic pathogen with the potential to cause a wide range of infections, usually following surgery and often associated with the use of medical devices, [14] especially with breast implants. [15] Cases of elbow joint and prosthetic joint infections as well as post-operative discitis are reported. [16][17][18] B. diminuta, R. pickettii, S. aureus and P. acnes could also be found in bacterial inflammation incidences related to breast implant surgery and post-implant complications. [19][20][21] In fact, biofilm formation on the breast implant surface and the subsequent chronic inflammation are considered among the triggers of capsular contracture development [22] representing the most common complication following breast implant insertion and is the leading course for reoperation. [23] To prevent biofilm formation, a material with initial microbial repellence properties would be beneficial to prevent overuse of antibiotics and, therefore, inhibit developing of multidrug-resistant microbes. This material could be used as a coating for established biomaterials such as silicone surfaces. [24,25] Natural spider silk is microbe repellent, hypoallergenic, biocompatible and biodegradable, and possesses outstanding mechanical properties. [26,27] However, the application of natural spider silks in medical devices is limited due to their low accessibility, the lack of possible modifications and a high batch-tobatch variation common to most natural materials. [28] The engineered silk protein eADF4(C16) is based on a major ampullate spidroin of the dragline silk of the European garden spider (Araneus diadematus). Additional data concerning the chemical structure is shown in Online Resource 1. [28,29] eADF4(C16) can be produced recombinantly at large scale as well as processed into a variety of morphologies other than fibres [30] such as films for coatings [27,[31][32][33][34][35] or soft hydrogels for biofabrication. [36][37][38] In this context, it The Author(s) 2021 has been shown to be a promising material for coating of polyurethane, polytetrafluoroethylene or silicone-based materials. [29] Materials made of eADF4(C16) are perfectly suited for biomedical applications since they display the absence of toxicity, lack of immune reactivity and slow biodegradation. [39,40] Recently published results indicate that bacterial infestation of spider silk surfaces is inhibited by microberepellent properties of the material´s surface rather than by antibacterial means. [41][42][43][44] As eADF4(C16) lacks cell binding motifs, like most so far identified spider silk proteins, eADF4(C16)-coated implants and catheters display significantly reduced adhesion and proliferation of any cells as compared to non-treated ones. [31,45] When transplanted into rats in vivo, eADF4(C16)-coated silicone implants exhibited a substantial reduction in capsular contracture. Interestingly, the four bacteria strains tested here are also sometimes associated with capsular contracture. [19,45] In order to analyse the bacterial repellence properties of recombinant spider silk coatings against four selected inflammation-related bacteria, silicone surfaces with different surface topographies were used to further analyse its impact. B. diminuta, R. pickettii, S. aureus and P. acnes were incubated on the respective spider silk-coated silicone samples, and their repellence properties were evaluated accordingly.

Surfaces and coating
Polydimethylsiloxane sheets of commercial materials with different texturing comprised POLYSmoooth © (Smooth), MESMO © (Mesmo) and POLYtxt © (Polytxt) (POLYTECH, Germany) showing an average surface roughness of < 1 µm for Smooth, 25 µm for Mesmo and 42 µm for Polytxt. Such materials are used as implant materials for breast implant surgery. They were washed with 80% ethanol and dried at room temperature overnight. Afterwards they were stored at room temperature for 4 weeks, placed on and covered with aluminium foil for further processing. Dip coating and fixation were performed in a clean bench (Hera guard, Eq ID 0011). The silicone squares were fixed with a tweezer and dipped for 10 s into isopropanol. After draining for 10-20 min, silicone squares were fixed with a tweezer and dipped for 5 min into eADF4(C16) coating solution at a concentration of 10 mg/ mL. After draining the coating solution, the silicone squares were placed on a metal grit to dry for 2-3 h on the clean bench. Samples were dipped into 0.5 M sodium phosphate buffer (pH 7.4) for 30 s and washed with ultrapure (Milli-Q) water before 1 h of drying. An additional cleaning step was performed using isopropanol before the silicone squares were ready to use.

Atomic force microscopy (AFM) measurements
The Smooth silicone samples were immobilised on microscopy slides as solid supports in order to prevent slipping during ongoing AFM measurements. Briefly, the microscopy slides were cleaned in a two-step procedure using alkaline Hellmanex cleaning solution (2% in ultrapure water) and a mixture of isopropanol (VWR) and water at a ratio of 3:1 (v:v). After drying, the glass was treated with air plasma for 10 min (Zepto, Diener Electronics). A thin film of Norland Optical Adhesive 63 (Norland Adhesives) was bladed onto the surface of the glass, and pieces of silicone samples (10 mm × 15 mm) were placed onto it. The glue was cured for 120 s under UV light.
The film thickness and swelling in PBS buffer (Sigma Aldrich, 1:10 diluted) were determined using AFM imaging in PeakForce Tapping Mode. All experiments were performed on a Dimension ICON (Bruker, Santa Barbara, CA) equipped with a Nanoscope V controller using SNL standard liquid imaging cantilevers (Bruker, Santa Barbara, CA) with a nominal spring constant of 0.35 N/m. Prior to the experiment, the spring constants of the cantilevers were calibrated using integrated functions of the used AFM software (NanoScope 9.10) following the thermal noise method. [46] Eventually, the sensitivity was calibrated in either media used for measuring, air or 1:10 diluted PBS buffer, respectively. All images were acquired at a PeakForce Tapping frequency of 1 kHz at lateral scan rates of 0.2 Hz and a peak force of 2 nN. The PeakForce amplitude was set to 150 nm in air and was increased to 300 nm in liquid to improve stability of the data acquisition. The same position on the same scratch was investigated in dried and swollen state. Optical microscopy control allowed to realign the scan area in liquid surrounding. Several consecutive scans at the crack in liquid were performed to verify that the equilibrium degree of swelling had been reached. A step height analysis was conducted using Nanoscope Analysis (version 1.80, Bruker Santa Barbara, CA) to determine the film thickness in dried and swollen state, which allowed to evaluate the swelling ratio.

Contact angle measurement
Silicone surfaces of the type Smooth, Mesmo and Polytxt were coated with eADF4(C16) or tested uncoated with four aqueous liquids, namely, ultrapure water, 0.9% sodium chloride isotonic saline (Berlin-Chemie AG, Germany), betadine betaisodona solution (Mundipharma GmbH, Frankfurt a.M., Germany) twofold diluted in isotonic saline, which resulted in a final composition of 50 mg/mL povidone iodine, 0.45% sodium chloride and minor components like glycerol, nonoxynol 9, disodium hydroxide, anhydrous citric acid, sodium hydroxide and potassium iodate, and Adams solution with 100,000 U/L bacitracin, 2 g/L cefazolin and 160 mg/L gentamicin in isotonic saline. 10 µL of the test fluids were applied on the silicone surfaces and immediately imaged using a Canon EOS 700D camera with a macro objective. The contact angles between the sample surfaces and the fluid drops were assessed using the drop analysis module of the Image J software (NIH, USA).

Bacteria culture
Brevundimonas diminuta (DSM 1635) and Ralstonia pickettii (DSM 6297) were plated from cryostocks on plate count agar (PCA) plates and incubated for 2 days at 29°C. Staphylococcus aureus (DSM 799) was plated from cryostocks on PCA plates and incubated for 1 day at 37°C. Propionibacterium acnes (DSM 1897) was plated from cryostocks on Columbia blood agar plates and incubated for 7 days at 37°C at anaerobic conditions. Using an inoculating loop, bacteria were transferred from agar into 5 mL of 1/500 nutrient broth medium and evenly dispersed. The number of bacteria was determined using spectrophotometry (OD 600 ), and the suspension was diluted 1/500 in nutrient broth to a final concentration of 5.0 × 10 5 cells/mL. The test inoculum was used within 2 h of preparation.
Bacterial solutions (1.25 × 10 4 CFU/cm 2 test area) were applied on the sample surfaces. To avoid drying out, samples were kept in a water-saturated environment for the 24 h cultivation period at 37°C. In order to remove the unbound cells, the surfaces were washed with sterile phosphate-buffered saline for 10 min with continuous shaking. Adhered cells were removed using ultrasound treatment and vortexing. Different dilutions of the washing solutions were incubated on culture plates, and colony numbers were determined.

Scanning electron microscopy (SEM)
To analyse the morphological structure and bacterial infestation using SEM, the surfaces were fixed in paraformaldehyde solution (3.7% in 1× PBS buffer, Carl Roth GmbH, Germany) for 30 min, washed afterwards using 1× PBS buffer and dried for 2 days at 37°C. The samples were fixed on SEM stubs using conductive adhesive pads (12 mm, PLANO GmbH, Germany), sputter coated with 1.3 nm platinum (Sputter Coater EM ACE600, Leica, Germany) and imaged at 5 kV accelerating voltage using the SE2-Detector in Standard Mode using a FEG-SEM after oxygen plasma treatment for 1 min to minimise contaminations (Apreo VS, FEI/ThermoFisher Scientific, Germany).

Contact angle measurements
Surface hydrophobicity is important for a material's interaction with its environment; therefore, contact angles were analysed at the silicone-(coated/uncoated) air interface. Previous studies showed that bacterial growth on surfaces is inter alia traced back to hydrophilicity as bacteria need an optimised environment for growth and reproduction. [47] Table 1 shows contact angles of four different solutions typically used for handling and processing of breast implants. Ultrapure water, isotonic saline, betaisodona solution and Adams solution used as covering solutions ahead of implantation by surgeons. It needs to be stated that contact angles are influenced by the roughness of a surface according to Cassie and Baxter resulting in higher values than on smooth surfaces. [48] As the surface roughness of Mesmo and Polytxt silicone samples is 25 µm and 42 µm and the surface coating thickness is at the submicron level, the overall surface structure was not affected by the coating. Online Resource 2 shows SEM images of spider silk-coated and uncoated silicone surfaces at ×500 magnification indicating no topographical differences. All coated surfaces displayed a contact angle between 35° and 105°, and in some cases, significant differences were obtained concerning the hydrophilic properties in comparison to uncoated template surfaces (Smooth, Mesmo and Polytxt) with contact angles between 74° and 112°.
As described above, rougher surfaces had higher contact angles due to incomplete wetting of the material interface. This influenced significantly the results on textured surfaces, [48][49][50] but results of Smooth samples were significant. Apart from the water contact angle on Mesmo surfaces, all coated samples showed enhanced wetting with a contact angle reduction of up to 57% for betaisodona solution on Smooth samples, confirming water contact angle data in the literature. [34,51] Betaisodona solution and Adams solution showed the highest wettability of the tested solutions on spider silk coatings with 35°-75° (betaisodona) and 58°-88° (Adams), respectively. The highest hydrophilicity was detected for betaisodona solution on eADF4(C16) films with a contact angle of 35°.

Coating thickness and swelling behaviour
Film thickness in dried state shown in Fig. 1(a) was determined from five PeakForce AFM images acquired on three different Smooth sample pieces. The average height was 801 nm ± 185 nm. For every image, the step height at the edge of the crack was determined across the whole image area in dry state [ Fig. 1(c)] as well as in swollen state [ Fig. 1(d)]. The roughness of the eADF4(C16) film and the silicone was evaluated to be 20.4 nm and 16.1 nm in an area of 5 µm × 5 µm, respectively.
In swollen state, a distinct increase in film thickness occurred to a height of approximately 1.6 µm [ Fig. 1(b)]. The swelling factor accounted to 2.03 ± 0.03. In addition to swelling in height, lateral swelling of the film was observed. No changes in film thickness were observed upon consecutive scans within the measurement time of 3 h. Thus, it can be concluded that the equilibrium swelling was reached before the beginning of the first scan at time scales < 30 min.

Bacterial repellence of spider silk-coated silicone surfaces
Biofilm formation of B. diminuta, R. pickettii, S. aureus and P. acnes was investigated on spider silk-coated as well as uncoated Smooth, Mesmo and Polytxt silicone surfaces. Microbial viability was quantified by plating the initial media and the washing buffers on media plates and counting cell colonies. Tables with raw data are shown in Online Resource 3. The reduction of cell numbers found on coated surfaces was normalised to uncoated surfaces for B. diminuta and R. pickettii and is shown in Fig. 2(a).
Bacterial repellence properties of the eADF4(C16)-coated silicone surfaces were predominant for B. diminuta and R. pickettii, which showed a reduction of adhesion between 79.4 and 99.7%, respectively. The negligible adhesion of S. aureus and P. acnes on the surfaces resulted in very low colony numbers, which prevented analysis of adhesion differences. In case of P. acnes incubated on coated and uncoated Polytxt surfaces, no difference in colony numbers could be observed after washing, as repellence properties were high regardless of the surface coating. The reduction of cell numbers found on coated surfaces normalised to cell numbers in the initial media for B. diminuta and R. pickettii, S. aureus and P. acnes is shown in Fig. 2(b). It could be concluded that S. aureus and P. acnes showed in general a low adhesion to the silicone surfaces; hence, a low ratio of adhered colonies after incubation to colony numbers could be detected in solution prior to incubation. Nevertheless, the spider silk coating promoted and improved repellence down to 0.01-1.41% for P. acnes and 0.01-2.17% for S. aureus [ Fig. 2(b)]. It has been previously shown that surfaces of recombinant spider silk materials have bacterial repellence properties for other pathogens like E. coli, S. mutans, P. pastoris and C. albicans. In this previous study, S. aureus was also tested and showed a low adhesion force to the spider silk surfaces, which explains the resulting low colony numbers. [44] Next, uncoated and coated silicone surfaces incubated with B. diminuta, R. pickettii, S. aureus and P. acnes (Fig. 3) were analysed using SEM, which confirmed that regardless of the silicone surface, the spider silk coating substantially restricted the attachment, growth and microbial colonisation. Uncoated textured surfaces like Polytxt and Mesmo showed slightly higher biofilm formation than Smooth ones due to a higher surface area. [52] In general, low adherence for S. aureus and P. acnes could be observed in both the absence and presence of the spider silk coatings [ Fig. 3(c) and (d)]. Figure 2. Results of adhesion tests using bacterial strains B. diminuta, R. pickettii, S. aureus and P. acnes. The ratio of colony numbers of spider silk-coated silicone sheets after washing are shown normalised to (a) colony numbers of uncoated silicone surfaces after washing and (b) to initial colony numbers on uncoated surfaces before incubation and washing. Ratio is given in %.

Conclusion
Coatings of silicone surfaces made of recombinant spider silk eADF4(C16) improved wetting of such surfaces using different pre-surgical aseptic solutions, which in combination with spider silk´s repellence properties enhances their applicability against opportunistic bacterial strains associated with inflammation. The repellence properties of spider silk coatings reduced microbial adherence up to 99.7% in comparison to uncoated silicone surfaces, and SEM images showed a significant decrease in biofilm formation after 24 h of incubation. Together with their biocompatibility and biodegradability, spider silk coatings could be part of a strategy to prevent inflammation and to reduce post-operative complications of implants.

Acknowledgments
The author gratefully acknowledges POLYTECH Health & Aesthetics GmbH (Germany) for providing silicone implant materials. This work was supported by Dr. Brünke MTC e.K.-Microbiological Testing Competence (Germany) concerning bacterial repellence tests. We thank Prof. Dr. Papastavrou and Dr. Nicolas Helfricht and the Department of Physical Chemistry II for support concerning AFM measurements.

Funding
Open Access funding enabled and organized by Projekt DEAL. No explicit funding was obtained for this study.

Conflict of interest
The authors declare the following competing financial interest: TS is co-founder and shareholder of AMSilk GmbH.

Open Access
This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.