Understanding the impact of crosslinked PCL/PEG/GelMA electrospun nanofibers on bactericidal activity

Herein, we report the design of electrospun ultrathin fibers based on the combination of three different polymers polycaprolactone (PCL), polyethylene glycol (PEG), and gelatin methacryloyl (GelMA), and their potential bactericidal activity against three different bacteria Staphylococcus aureus (S. aureus), Pseudomonas aeruginosa (P. aeruginosa), and Methicillin-resistant Staphylococcus aureus (MRSA). We evaluated the morphology, chemical structure and wettability before and after UV photocrosslinking of the produced scaffolds. Results showed that the developed scaffolds presented hydrophilic properties after PEG and GelMA incorporation. Moreover, they were able to significantly reduce gram-positive, negative, and MRSA bacteria mainly after UV photocrosslinking (PCL:PEG:GelMa-UV). Furthermore, we performed a series of study for gaining a better mechanistic understanding of the scaffolds bactericidal activity through protein adsorption study and analysis of the reactive oxygen species (ROS) levels. Furthermore, the in vivo subcutaneous implantation performed in rats confirmed the biocompatibility of our designed scaffolds.


Introduction
Microbial infections are immense health problems, in particular with respect to wound healing. 1 Here, a microbial infection at the wound site can severely prolong the healing process, leading to necrosis, sepsis, and even death. 2 The most common pathogenic bacteria found in the hospital environment are Escherichia coli (E. coli), Pseudomonas aeruginosa (P. aeruginosa), Staphylococcus aureus (S. aureus), Staphylococcus epidermis (S. epidermis), and various filamentous fungi as well as yeasts (i.e., Candida spp.). [3][4][5] The use of antibiotics can results in an escalating drug resistance in pathogenic microorganisms, which is associated with increased morbidity and mortality and therefore should be avoided when possible. 6 In fact, according to the World Health Organization (WHO), antibiotic resistance is one of the biggest threats to global health. Therefore, the design of materials with the ability to significantly reduce the administration of antibiotics and avoid infections from highly antibiotic-resistant bacteria in the hospital environment are highly desirable. In this context, degradable polymers with tunable mechanical, biological and chemical properties and ease of fabrication can be attractive for aforementioned applications. 7 Through the combination of several biomaterials with inimitable properties, a tailor-made hybrid material can be designed without the use of any complicated fabrication approaches or chemistry. However, one requirement is that such materials are compatible with each other avoiding interference or inhibition due to mismatched chemistry or structure, but instead promote properties synergistically.

Preparation of the solutions prior electrospinning
The samples were prepared separately by mixing PCL (0.24 g), PEG (0.15 g) and GelMA (0.05 g) in 5 mL of HFIP for each sample in vials. Subsequently, the vials were closed and dissolved under stirring at room temperature for 24 h. Then, the solutions were placed in a syringe (BD Yale, 3 mL) followed by connecting a needle to the syringe (Inbras, 23G). The electrospinning was carried out at 17 kV (Nanospinner Machine, Inovenso) as a positive voltage.
An aluminum foil collector plate (20 cm x 20 cm square plate) was used as the anode, a needlecollector distance of 10 cm was used, and 1 mL/h was used as the solution flow rate (Harvard, PHD 2000). The temperature and humidity were controlled at 21-23 °C and 44-54%, respectively.

Photocrosslinking of the produced scaffolds
The photocrosslinking solution was prepared by adding 1.0 g of photoinitiator to 10 mL of ethanol in the absence of the light and stirred until completely dissolved. The uncrosslinked electrospun mats was first immersed in a glutaraldehyde (25% in deionized (DI) water ) solution (10 mL/500 mL ethanol) overnight and then extensively washed with a solution of glycine (3.75 g in 250 mL of DI water). Subsequently, the mats were immersed in an ethanol solution containing photoinitiator (10 wt%) for 2 h and then exposed to a 365-nm UV light (OminiCure ® -2000 Series) at a 10-cm working distance for 10 min. Then, the crosslinked mats were washed 3 times with ethanol. Afterwards, the mats were washed in DI water and dried under vacuum. Fig.   1c illustrates a possible mechanism of PCL:PEG:GelMA interaction after photocrosslinking.

Characterization of the produced scaffolds
A scanning electronic microscope (FEI Quanta 200, 3 kV) was used to analyze the scaffold morphology. A thin gold layer was evaporated onto all scaffold surfaces before analysis to improve image acquisition.
Fourier transform infrared spectrometry (FTIR, Spotlight-400, Perkin Elmer FTIR Imaging System) was used to analyze the chemical groups in the biomaterial before and after photocrosslinking under a controlled atmosphere and transmittance mode. The spectra were collected at regions between 4000−400 cm -1 and were analyzed in three different regions. The regions of interest were analyzed, and their vibrational modes indexed.
For wettability analysis, the contact angle (CA) between the surface of the scaffold and a DI water or diiodomethane drop was monitored in the dynamic mode, using a contact angle device (Krüss, Model DSA 100). Before analysis, a careful calibration was performed in accordance to the equipment manual and all the scaffolds were fixed on a Teflon base. Then, DI water and diiodomethane (2 µL) were dropped onto the scaffold surfaces and measured after 1 s.
The atmosphere and humidity (~ 60%) were controlled during all the measurements. The surface tension and adhesion force were calculated and described in the supplementary data.

Bacteria tests
S. aureus (ATCC 25923), P. aeruginosa (ATCC 25668), and Methicillin-resistant S. aureus (MRSA) (ATCC 43300) were inoculated in 3% Tryptic soy broth (TSB) and after 14 h, the bacteria were plated at a concentration of 10 4 cells/mL onto each sample (1 cm x 1 cm). After 24 h, the samples were lightly rinsed with PBS (x2) to remove any non-adherent bacteria. This bacterial solution was then diluted serially (10X; 100X; 1000X) and plated on tryptic soy agar plates and incubated overnight (~14 h). Afterwards, the colonies were counted to calculate the colony-forming units (CFUs). The data were normalized by raw PCL data to highlight the bactericidal activity of the produced scaffolds without the use of antibiotics.

Protein adsorption
For protein adhesion test, the fibers were placed in a 24-well plate, 0.5 mL of TSB or casein solution were added and the plate was kept inside an incubator at 37 ºC for 24 h. After this time, the samples were transferred to a new plate, washed twice with PBS and 0.5 mL of immunoprecipitation assay (RIPA) buffer was added to each well. After 10 min in contact with RIPA buffer, the solution inside the wells was mixed with pipette to ensure homogeneity. Using a 96-well plate, 25 µl of the samples were mixed with 200 µl of working reagent from Pierce TM bicinchoninic acid (BCA) Protein Assay Kit, Thermo Scientific. The plate was placed on a plate shaker for 30 s and then incubated at 37 ºC for 30 min. After incubation, the plate was coolED to room temperature and the absorbance was measured at 562 nm using a spectrophotometer (SpectraMax® M3, Molecular Devices).

Analyses of the oxygen species
The reactive oxygen species (ROS) levels before and after UV crosslinking was analyzed using Fluorimetric Hydrogen Peroxide (H 2 O 2 ) Assay Kit (Sigma-Aldrich) and the Cellular ROS/Superoxide (O 2 · -) Detection Assay (Abcam). 1 mL of S. aureus, P. aeruginosa and Methicillin-resistant S. aureus (MRSA) solutions with concentration of 10 4 CFU/mL were placed into 24-well plates containing the fibers. The plates were kept inside the incubator at 37 ºC for 24 h. After this time, the bacteria solution was removed, subsequently, the scaffolds were washed 2x with 1 mL PBS each, and then placed inside an eppendorf tube containing 1 mL of PBS. After vortex for 15 min, the solutions were used to perform the 2 tests.
For the hydrogen peroxide assay, 50 µl of each solution was placed in a 96-well plate and mixed with 50 µl of working solution from the assay kit. The 96-well plate was incubated at room temperature for 30 min followed by measuring the absorbance at 540 nm using a spectrophotometer (SpectraMax® Paradigm®, Molecular Devices).
In the more generalized ROS and superoxide detection process, a microplate assay procedure was implemented in which a mixture containing reactive cell permeable probes was added to the test samples. Specifically, 100 μ L of each test and control was transferred from the eppendorf tubes after vortexing to an opaque 96-well plate. Positive controls were established in which a set of cells were treated with Pyocyanin in order to induce ROS generation. 100 μ L of the ROS/superoxide detection mixture was added to the cell suspensions inside the 96-well plate, and the plate was allowed to incubate at 37 °C for 1 h in the dark. After the staining period, the 96-well plate was loaded onto the spectrophotometer with standard fluorescein (Excitation/Emission = 490/525 nm) and rhodamine (Excitation/Emission = 550/620 nm) filter sets for detecting ROS and superoxides, respectively.

Statistical analysis
All experiments were conducted in triplicate and were repeated at least three different times with analysis of variance (ANOVA) followed by Student's t-tests to determine statistical differences between mean values. The populations from the analyzed scaffolds were obtained with normal distribution and independent to each experiment.

Results
The surface and structural characterization of the scaffolds are summarized in Fig. 2. One clearly observation is that ultrathin and microfibrous scaffolds in the absence of any beads were obtained independently of the analyzed groups. When the PCL solution was electrospun, a nonhomogeneous behavior, and some defected fibers were produced (Fig. 2a1). More details about the surface defects can be seen in Fig. 2a2. and Fig. 2a3, which highlights the partial roughness structure and defects on the scaffold. The distribution of diameters is depicted in Fig. 2a4, where PCL presented microfibrous structures (diameter of 1.53 µm ± 0.74 µm). After the incorporation of PEG, a discrete regularity in the material was obtained (Fig. 2b1) and the PCL:PEG scaffolds presented a better smoothness (Fig. 2b2). The smooth surface of the fibers could be more easily identified in the magnified Fig. 2b3 image. The thickness also decreased by 30% compared to PCL (Fig. 2b4, 0.92 µm ± 0.43 µm). Furthermore, after the addition of GelMA the electrospun fibers transformed to an even more regular scaffold (Fig. 2c1). Fig. 2c2 illustrates details from the non-defected structures with a homogeneous behavior. By comparing the detailed SEM image Fig. 2c3, with PCL ( Fig. 2a1-a3) and PCL:PEG ( Fig. 2b1-b3) scaffolds, a smoother surface was obtained. Moreover, the distribution of the diameters from the PCL:PEG:GelMA scaffolds showed ultrathin fibers (0.24 µm ± 0.10 µm) smaller than PCL. The crosslinked PCL:PEG:GelMA scaffolds also presented a more homogeneous surface without any further defects ( Fig. 2d1-d3), and with increases in diameter ( Fig. 2d4) after crosslinking compared to non-crosslinked PCL:PEG:GelMA ( Fig. 2c1-c3).
The structural analysis and chemical interactions were recorded on the FTIR analysis ( Fig. 2e1-e3). The main peaks for each separate polymer are shown in Fig. 2e1. In the FTIR from the PCL, the main peaks could be identified at: 731 cm -1 (-(CH 2 ) n ), 1157 cm -1 (C-O and C-C stretching in the amorphous phase), 1170 cm -1 (symmetric C-O-C stretching), 1190 cm -1 ((ν(O-C-O)), 1240 cm -1 (asymmetric C-O-C stretching), and 1342 cm -1 (CH 2 ) 38, 39 . Also, the fingerprint region center was observed at: 1637 cm −1 (amide I), 1529 cm −1 (amide II), and 1448 cm −1 (amide III), corresponding to the stretching of the C=O bond, bending of the N-H bond, and plane vibration of C-N and N-H, respectively 40 . Meanwhile, the main peaks related to PEG are: the region 1100−1200 cm -1 of the mixed polymers, the intermolecular forces such as hydrogen bonds, ions, and charge transfers could be observed 43 . Fig. 2e2 and 2e3 show the difference between non-and crosslinked scaffolds. It is known that FTIR helps to analyze the fraction of unreacted C=C double bonds of methacrylate or methacrylamide groups 44 . The band centered at 1634 cm −1 indicating the presence of the double bond was absent after UV irradiation (Fig. 2e2), whilst this was present on the scaffold before UV irradiation (identified at Fig. 2e2). The increased chemical crosslinking resulted in an enhanced physical network 44, 45 . After UV irradiation, a peak centered at 1600 cm −1 appeared (acrylic double bonds, C=C) and at the same time, the peak at 1100 and 1170 cm −1 disappeared ( Fig. 2e3, related to the -C-O-C-vibration commonly found when scaffolds contains GelMA).
This can be associated to the reaction of the methacryloyl moiety during crosslinking 43, 46 .
We compared these results to raw PCL. After UV crosslinking, the scaffolds inhibited the growth of S. aureus (Fig. 3a, gram-positive) compared to raw PCL. Interestingly, the PCL:PEG:GelMA-UV scaffolds reduced P. aeruginosa by 10-fold comparable to the material without UV crosslinking (Fig. 3b). Moreover, our developed scaffolds showed high antibacterial activity against multiple drug-resistant S. aureus bacteria (Fig. 3c, MRSA) and the same efficiency was observed for the materials before and after UV crosslinking. We measured the contact angle (CA) from our developed scaffolds to explain this bactericidal activity presented for the PCL:PEG:GelMA before and after UV crosslinking and calculated their surface energy and adhesion force of bacteria attached using thermodynamic approaches (see supplementary information). Interesting, the PCL:PEG:GelMA-UV presented a CA (water, 69 o ) higher than the same material without UV crosslinking (water, 25 o ), probably due to a more dense material obtained after crosslinking introducing covalent bonds to the material. However, the CA using diiodomethane gave the same result for both of the materials with/without crosslinking (~5 o , supplementary information, Table S2). Then, we calculated the surface energy of our developed scaffolds and the material after UV crosslinking promoting the bactericidal activity showed a lower surface energy (53.3 mJ/m 2 ) compared to the material without UV crosslinking (71.4 mJ/m 2 ). Moreover, the adhesion force was calculated using a thermodynamic approach where the average value of work obtained for the adhesion of bacteria from PCL:PEG:GelMA were ΔF Adh = -34.6 mJ/m 2 (without crosslinking) and ΔF Adh = -22.6 mJ/m 2 (after crosslinking). These results indicated that the scaffold surfaces before UV crosslinking was more favorable for bacteria adhesion than after 47 .
In order to analyze the effect of the UV crosslinked fiberson the bactericidal properties, the non-crosslinked and crosslinked fibers were submitted to protein adsorption assay (Fig. 4a), hydrogen peroxide assay (Fig. 4b), reactive oxygen species (ROS) assay (Fig. 4c) and superoxide assay (Fig. 4d). The protein adsorption assay showed for both TSB and casein an increase of the protein concentration adsorbed onto the crosslinked fibers. Impressively, bacterial adhesion to the implant surfaces is inhibited, despite the generalized understanding that TSB and casein aid in bacterial growth [34] . The reason for this is that casein adsorption to an implant surface prevents biomolecular attachment, and the increased protein adsorption over a restricted time period correlates with antibacterial properties due to short-scale bacterial adhesion dynamics [35] . From this study, it is possible to conclude that the UV crosslinking of scaffold composites improves protein adsorption, and, correspondingly, reduces bacterial attachment and growth.
The UV crosslinking also influences the production of reactive oxygen species, such as hydrogen peroxide (Fig. 4b). The data obtained using the Fluorimetric Hydrogen Peroxide Assay Kit indicates that crosslinking of the fibers contributed to a decrease in the production of H 2 O 2 for all three of the bacterial types evaluated. Nevertheless, the overall oxidative stress imposed on the P. aeruginosa and S. aureus strains due to the combined effect of H 2 O 2 , hydroxyl radicals (HO·), peroxynitrite (ONOO-), and peroxy radicals (ROO·), as measured using the ROS/Superoxide Detection Assay, is augmented for the UV crosslinked samples (Fig. 4c). This indicates that scaffold crosslinking plays a role in promoting the production of ROS that facilitate the self-destruction of P. aeruginosa and S. aureus cells. This conclusion, however, cannot be drawn for the drug-resistant MRSA strain. Indeed, a modest reduction in the ROS levels for MRSA suspensions that were grown along UV-crosslinked scaffolds was documented.
Moreover, an increase in the overall oxidative stress imposed on the P. aeruginosa and S. aureus strains due to scaffold crosslinking did not correlate with an increase in superoxide production ( Fig. 4d), and the amount of O 2 · generated by these bacteria upon exposure to both the UVcrosslinked and non-crosslinked samples was approximately the same. A significant decrease in the production of O 2 · by MRSA cells incubated on the crosslinked scaffolds, relative to the noncrosslinked scaffolds, however, was observed.

Discussion
Herein, we combined three different polymers to obtain ultrathin electrospun meshes. To date, we are not aware of any prior reports that have successfully demonstrated the electrospinning of the combination of PCL, PEG, and GelMA and evaluation on their bactericidal activity. Their efficiency was more evident after UV photocrosslinking, which could be associated to the closer interaction between the polar groups (such as -NH 2 , -OH, and the carbonyl group (-CO)) in GelMA and -OH found on the PEG chain, and also due to the more compact scaffold obtained. The surface energy change of the scaffolds after UV crosslinking (Table S2) can lead to changes in protein adsorption and bioactivity altering cell functions and, at the same time, reducing bacteria adhesion and proliferation. 48 Furthermore, it is widely accepted that reactive chemical groups on micro and nano-fibers can disrupt bacteria membranes via oxygen free radical formation. Therefore, it can be inferred, especially in the case of MRSA, that factors besides ROS and superoxide production contribute to the bactericidal efficacy of PCL:PEG:GelMA scaffolds that are crosslinked. This is beneficial considering that cellular mechanisms have evolved to evade killing by self-generated ROS (Fig. 4c). For instance, adhesion prevention appears to factor significantly into disabling MRSA attachment and growth along scaffolds that incorporate GelMA.
Scaffolds based on different amounts of PCL, GelMA, and gelatin have been reported by Correa et al. 49 However, the authors first produced PCL mats and then incorporated GelMA and gelatin afterwards using in situ coating. The process presented here is different and more interesting to tissue engineering applications since we electrospun PCL, PEG and GelMA together and the subsequent photocrosslinked composite promoted bactericidal activity. A more hydrophilic scaffold (CA ~25 o ) was obtained after GelMA and PEG inclusion. The potential of water sorption of GelMA and GelMA:PLGA mats were also investigated by Zhao et al. 50 . The authors obtained a highly porous scaffold which increased the diameter to ~70% after 24 h.
Other authors have explored the influence of GelMA and PEG to induce vascularization and extracellular matrix calcification 51-57 , however, the bactericidal properties have not yet been investigated.
Anabi et al. reported that hydrogels based on GelMA conjugated with an amphiphilic peptide reduced bacteria growth 58 . However, the bactericidal effect against S. aureus was only observed after peptide incorporation. Commonly, micro-and nanoparticles have been associated to short sequences of cationic amino acids to possess a broad-spectrum of bactericidal activity against Gram (+/−) bacteria 59-62 . Here, impressively, S. aureus and P. aeruginosa were reduced to a half amount compared to raw PCL without requiring additional steps of peptide functionalization or antibiotic incorporation. Additionally, no statistical difference was observed compared to PCL:PEG:GelMA before and after UV against MRSA.
Rujitanaroj et al. combined gelatin and GelMA with silver nanoparticles and evaluated their bactericidal effect 63 ; the use of silver has recently been questioned due to high mammalian cell toxicity. Moreover, the authors only observed the bactericidal effect when silver nanoparticles were added. In this manner, it is plausible to at least preliminarily compare the antimicrobial mechanism of our scaffold to those obtained when amphiphilic peptides (AMPs) were used. It is noteworthy that our developed scaffold had no positive charge in its structure, indicating that the binding interactions between the hydrogel and gram-positive and negative bacteria were unlikely to be caused by electrostatic interactions, as seen in AMPs, which involved bacterial cell wall and membrane disruption. [64][65][66] We believe that the design of a more hydrophilic scaffold promoted the promising results obtained from the bacteria study by altering initial protein adsorption that bacteria depend upon to attach 67 . All these characteristics highlight the application of PCL:PEG:GelMA as novel scaffolds for tissue engineering purposes. Further studies will be needed to determine the exact mechanisms (including initial ion/protein adsorption) responsible for the reduced bacteria growth on the present scaffolds.

Conclusions
In this report we disclose, for the first time, the preparation and bactericidal proficiency investigation of electrospun scaffolds by merging PCL:PEG:GelMA polymers. The designed ultrathin fibers scaffolds were hydrophilic comparable to raw PCL. Moreover, the developed scaffolds were also able to efficiently prevent the growth of S. aureus, P. aeruginosa, and MRSA after UV crosslinking. In conclusion, all these findings suggest that the designed biomaterial is a promising candidate for various biomedical applications.   C  o  m  p  a  r  i  s  o  n  o  f  a  b  i  o  t  i  c  a  n  d  b  i  o  t  i  c  d  e  g  r  a  d  a  t  i  o  n  o  f  P  D  L  L  A  ,  P  C  L  a  n  d  p  a  r  t  i  a  l  l  y  m  i  s  c  i  b  l  e  P  D  L  L  A  /  P  C  L  b  l  e  n  d   .   E  u  r  o  p  e  a  n  P  o  l  y  m  e  r  J  o  u  r  n  a  l   .  2  0  1  3  ;  4  9  :  7  0  6  -7  1  7   P  .  D  a  l  l  a  s  ,  D  .  N  i  a  r  c  h  o  s  ,  D  .  V  r  b  a  n  i  c  ,  N  .  B  o  u  k  o  s  ,  S  .  P  e  j  o  v  n  i  k  ,  C  .  T  r  a  p  a  l  i  s  ,  e  t  a d  i  f  f  e  r  e  n  t  i  a  t  i  o  n  o  f  h  u  m  a  n  m  e  s  e  n  c  h  y  m  a  l  s  t  e  m  c  e  l  l  s  a  n  d  e  n  d  o  t  h  e  l  i  a  l  c  o  l  o  n  y  -f  o  r  m  i  n  g  c  e  l  l  s  c  o  e  n  c  a  p  s  u  l  a  t  e  d  i  n  a  p  a  t  t  e  r  n  e  d  h  y  d  r  o  g  e  l   .   J  o  u  r  n  a  l  o  f  c  o  n  t  r  o  l  l  e  d  r  e  l  e  a  s  e  :  o  f  f  i  c  i  a  l  j  o  u  r  n  a  l  o  f  t  h  e  C  o  n  t  r  o  l  l  e  d  R  e  l  e  a  s  e  S  o  c  i  e  t  y   .  2  0  1  6  ;  2  2  3  :  1  2  6  -1  3  6   M  .  K  .  A  p  a  r  n  a  t  h  i  a  n  d  J  .  S  .  P  a  t  e  l  ,   B  i  o  d  e  g  r  a  d  a  b  l  e  G  e  l  a  t  i  n  M  e  t  h  a  c  r  y  l  a  t  e  G  e  l  a  s  a  P  o  t  e  n  t  i  a  l  S  c  a  f  f  o  l