Human amniotic membranes were collected from the Gynecology Department, Korea University Guro Hospital, Seoul, South Korea. Prior written consent was obtained from each volunteer donor, and the samples were collected after caesarian sections. The animal experimental protocols were performed according to the guidelines of the ethics committee of Korea University, Guro Hospital, and all the protocols followed in this study were approved by the Institute Review Board (IRB) of Korea University, Guro Hospital, Seoul, South Korea (approval number UGH16046002).

Streptococcus pneumoniae D-39 (NCTC 7466) is Avery’s Virulent, Serotype 2 encapsulated Strain, was purchased from the Health Protection Agency Culture Collections (HPA, Salisbury, United Kingdom). It is fully sequenced (Lanie et al., 2007), forms robust in vitro biofilms (Vidal et al., 2011) and it remained extremely virulent in animal models of infections after many year of its isolation form patient (Benton et al., 1995; Blue et al., 2003; LeMessurier et al., 2006). S. pneumoniae serotype 3 (ATCC strain 6303), 19A, and 19F (ATCC strain 49619) were purchased from the American Type Culture Collection (Manassas, VA, United States); serotype 11 (strain 7101975) was obtained from the Infectious Disease Department of Korea University Medical Center, Guro Hospital, Seoul. S. pneumoniae strains were grown on brain heart infusion (BHI) broth medium or on blood agar plates (BAPs) supplemented with 5% v/v sheep blood. The BAPs were pre-made and purchased from Shin Yang Chemicals Co., Ltd. (Seoul, South Korea). For minimum inhibitory concentrations (MICs) detection bacteria were grown in Muller-Hinton broth and agar (Sigma, St. Louis, MO, United States).

The human AM matrices were collected from the Gynecology Department, Korea University Guro Hospital, Seoul, South Korea. Before the membrane samples were collected, the prospective donors were examined for gestational diabetes, pre-eclampsia, or infectious disease. Those testing positive for any of the above were excluded from the study. AME/CME were prepared using methods described previously by He et al. (2008) with some modifications reported in our previous studies (Go et al., 2016, 2017). Briefly, the tissues were washed with phosphate-buffered saline (PBS) three times to remove blood clots, and then sliced into small pieces, frozen using liquid nitrogen and ground into fine particles using a mortar and pestle. The membrane particles were then mixed with PBS at ratio of 1:1 (wt/vol) and homogenized on ice for 1 h. The membrane lysates were centrifuged twice at 12000 rpm at 4°C for 10 min to obtain the supernatants. The supernatants were filter (0.22-μm pore size) sterilized. The protein concentrations in the extracts were determined using a Detergent Compatible (DC) Protein Assay (Bio-Rad, Hercules, CA, United States) according to the manufacturer’s instructions. Briefly, 5 μL each of AME/CME extract (unknown) and standard solutions (from 2 to 0.125 mg/mL) were inoculated in triplicate in a 96-well plate. Solution A (25 μL; supplied with kit) was added to each well followed by 200 μL of solution B. The plates were incubated at room temperature for 15 min and absorbance was measured at 750 nm using microplate-reader. The standard curve was constructed and concentration of AME/CME extracts was calculated. The extracts were stored at -80°C until use.

Minimum inhibitory concentrations were determined using the broth micro-dilution method as recommended by the Clinical and Laboratory Standards Institute (CLSI, 2005). Overnight cultures of bacteria in MH broth (MHB) were diluted (1:10) in fresh medium and grown to the exponential phase. From the cells growing at the exponential phase, cell suspensions of 1–5 × 10⁵ colony-forming units (cfu)/mL were prepared. Serial twofold dilutions of AME/CME, from 2 to 512 μg, were added to the MHB cell suspensions. Thereafter 200 μL of each suspension were inoculated into each well of a 96-well plate. Negative control wells were inoculated with cell suspension only. The plates were incubated for 24 h at 37°C, and bacterial growth was determined by measuring the optical density at 600 nm (OD₆₀₀) using a microplate reader. To further confirm the bacterial growth in each well after incubation, 10-μL aliquots of the cell suspensions from the 96-well plate were spread on MH agar, and bacterial colonies were counted after 24 h of incubation at 37°C. The MIC was defined as the lowest AME/CME concentration in which no visible growth was observed. The experiment was performed in quadruplicates and was repeated two times.

In this study, biofilm experiments were performed using static biofilm model, with 18 h biofilm growth as per pervious reports (Moscoso et al., 2006; Domenech et al., 2013). A microtiter plate assay was used to grow and quantify in vitro biofilms of the bacteria tested with or without AME/CME extracts (Christensen et al., 1985; Yadav et al., 2017). Briefly, the biofilms were grown to stationary phase in 24- or 96-well polystyrene flat-bottom microtiter plates (BD Falcon, Sparks, MD, United States) for 18 h at 37°C. Fresh bacterial cell suspensions were prepared in BHI medium, diluted 1:200, and inoculated into plates at 200 μL/well in 96-well plates or 1 mL/well in 24-well plates. Twofold diluted AME/CME from 16 to 256 μg, was added to the wells and incubated at 37°C for 18 h. PBS alone was introduced into the vehicle control wells. After inoculation, the media and planktonic bacteria cultures were decanted, washed twice with sterile water, and stained with 50 μl of 0.1% crystal violet (CV) for 15 min. After staining, the plates were washed and dried, and CV was dissolved in 200 μL (96-well plates) or 1 mL (24-well plates) of ethanol for each well. Absorbance at 570 nm was then measured using a microplate reader.

The metabolically active cells in the biofilms were detected using resazurin staining. Resazurin is a blue, non-fluorescent dye that is reduced to the pink and highly fluorescent compound resorufin by metabolically active cells, thus allowing the quantitative measurement of cell viability. Resazurin staining was performed using a previously reported procedure with minor changes (Paytubi et al., 2017). Briefly, 0.02% (w/v) resazurin sodium salt (Sigma) solution was prepared in water and sterilized by filtration. Biofilm were growth in 96-well microtiter plates with different concentration of AME/CME (16–256 μg) for 18 h. The biofilms were washed twice with sterile water and incubated with 25 μL (0.02%) resazurin dye and 100 μL BHI-broth. The plate was incubated at 37°C for 2 h and fluorescence was measured (excitation 570 nm, emission 615 nm) using a Varioskan LUX multimode microplate reader (Thermo Scientific, Waltham, MA, United States).

Eighteen-hour pre-established biofilms of S. pneumoniae strain D-39 were treated with two and fourfold MICs of AME/CME and incubated for an addition 6 h. The biofilm biomass was assessed by using the CV microplate assay, and viable bacteria were enumerated and presented as cfu counts, as described above.

Streptococcus pneumoniae strain D-39 was grown to the log phase, and aliquots were removed, treated with 512 μg (2 × MIC) AME or CME, and further incubated. The bacteria were counted at various time intervals (0, 1, 3, and 6 h). The controls were treated with PBS.

Penicillin-Streptomycin antibiotic solution is commonly used for washing AMs before preservation to eliminate possible microbial contamination. In this study, we evaluated the synergistic effects of AME/CME and P-S antibiotic solution on S. pneumoniae by using the checkerboard test, as previously described (Petersen et al., 2006). S. pneumoniae D-39 was grown with twofold dilutions from 32 μg (0.125 × MIC) to 256 μg (MIC) of AME/CME, as well as 0.025 μg (0.125 × MIC) to 0.2 μg (MIC) of P-S antibiotic solution alone and in combination. In each well of a 96-well plate, the AME/CME or antibiotic solution concentrations were adjusted such that the first antibiotic of the combination was serially diluted along the ordinate, while the second drug was diluted along the abscissa. Growth was monitored by determining the optical density at 600 nm. The fractional inhibitory concentration index (FICI) was calculated as the sum of the fractional inhibitory concentrations (FICs) of the AME/CME and P-S antibiotic solution. FICs are defined as the MIC of each drug when used in combination, divided by the MIC of the drug when used alone. If the FICI is less than or equal to 0.5, the interaction of the two drugs is synergistic. The interaction is additive if the FICI > 0.5 and ≤1. The interaction is indifferent if the FICI is >1.0 and ≤2.0, and antagonistic if the FICI is >2.0.

To evaluate the effect of AME or CME+ P-S antibiotic solution (AME/CME solution) on S. pneumoniae biofilm, in vitro biofilms were grown with different concentration of AME/CME solution (4.003–64.05 μg) and biofilm biomass was quantified using the CV-microtiter plate assay as described above. The metabolically active cells within biofilms were detected using resazurin staining as described above. To evaluate the biofilm eradication potential of AME or CME + P-S antibiotic solution, 18 h pre-established biofilms of S. pneumoniae D-39 were treated with AME or CME + P-S antibiotic solution (2 × MIC and 4 × MIC) and incubated for an additional 6 h. The biofilm biomass was assessed by CV microplate assay, and viable bacteria were detected by resazurin staining as described above.

It is reported that S. pneumoniae in vitro biofilms formed on abiotic surfaces for 10–12 h consist of a three-dimensional structures with significant thickness (25–30 μm) (Moscoso et al., 2006). In this study, we evaluated the effect of AME/CME on 18-h pre-established in vitro biofilm using scanning electron microscope (SEM). In vitro biofilms of S. pneumoniae D-39 were grown as described above for 18 h and treated with indicated concentration of AME/CME solution for 6 h; the control biofilms were treated with PBS. After incubation, the biofilms were washed, pre-fixed by immersion in 2% glutaraldehyde in 0.1 M phosphate buffer, and post-fixed for 2 h in 1% osmic acid dissolved in PBS. The biofilm samples were then treated with increasing concentrations of ethanol (60–95%), followed by t-butyl alcohol. The samples were dried in a freeze dryer (ES-2030, Hitachi, Tokyo, Japan) and coated with platinum using an ion coater (IB-5, Eiko, Kanagawa, Japan). The images were captured under field emission-SEM (FE-SEM; Hitachi, S-4700, Tokyo, Japan).

The biofilm eradication activity of AME/CME was evaluated by confocal microscopy. The biofilms were grown on micro-discs for 18 h and then treated with AME/CME solution (2 × MIC) for 6 h. The control biofilms were treated with PBS only. The biofilms were stained using a LIVE/DEAD Biofilm Viability Kit (Invitrogen) using procedures described by the manufacturer. Control cells were grown without AME/CME solution treatment. After washing, the stained biofilms were viewed under a Nikon A1 confocal microscope (Nikon Instruments Inc., Melville, NY, United States) using fluorescein (green) and Texas red (red) bandpass filter sets.

The inhibitory effect of AME on the in vivo colonization of S. pneumoniae was evaluated using a rat OM model (Yadav et al., 2012, 2017). Twenty-six pathogen-free Sprague-Dawley rats weighing 150–200 g were obtained from Koatech (Pyeongtaek, South Korea). The rats were isolated in a sterile environment for 2 weeks, and middle ear abnormalities were examined before the experiment. The rats were divided into four groups: group 1 rats (n = 9) were inoculated with bacteria only; group 2 rats (n = 9) were inoculated with bacteria and AME solution (0.5 × MIC); group 3 rats (n = 6) received medium only (vehicle control); and group 4 rats (n = 3) had no procedures performed. Animals were euthanized using Zoletil H (tiletamine-zolazepam, Virbac, Carros, France) and Rompun H (xylazine-hydrochloride, Bayer, Leverkusen, Germany) combined at a ratio of 1:1. The S. pneumoniae cell suspension was prepared in BHI medium, and 50 μL (approximately 10⁷ cfu) was injected into the middle ear cavity of each rat in groups 1 and 2 through the tympanic membrane of the right ear using a tuberculin syringe and 27-gauge needle (Yadav et al., 2012). The animals were held for 1 week before they were sacrificed and bullae were acquired aseptically. Tympanic membranes were removed, and the middle ears were dissected and photographed. The representative bullae from rats in each group were preserved in SEM solution (glutaraldehyde and paraformaldehyde solution), aseptically homogenized, serially diluted, and plated on BAP for cfu counts. Colonies were counted after 24 h of incubation at 37°C.

A proteomic analysis of AME/CME was performed using tandem mass tag (TMT)-based quantitative mass spectrometry (MS) using previously described procedures (Park et al., 2016). Briefly, the extracts were denatured with 8 M urea at room temperature for 10 min. The e protein solution was then reduced using dithiothreitol (DTT) at room temperature and alkylated using indole acetic acid (IAA). Proteins were quantified using the Bradford protein assay, using 100-μg aliquots of protein diluted in 50 mM ABC buffer. The diluted protein samples were digested with trypsin at 37°C for 16 h and quenched in 3% FA The digested samples were dissolved in mobile phase A and analyzed using a liquid chromatography (LC)-MS/MS system consisting of a nano liquid chromatograph and a mass spectrometer equipped with a nano electrospray source. The samples were loaded into a C₁₈ trap column with an autosampler. Samples were concentrated and desalted on the trap column for 5–10 min at a flow rate of several μL/min and then separated on a C₁₈ analytical column. For the LC–MS/MS analysis, the mobile phases comprised 99.9% water (A) and 99.9% acetonitrile (ACN) (B), with each containing 0.1% FA. The LC gradient started with 5% of B for 15 min and was ramped to 15% of B for 5 min, 50% of B for 75 min, and 95% of B for 1 min; it was held at 95% of B for 13 min and then at 5% of B for additional 1 min. The column was re-equilibrated with 5% of B for 10 min before the next run. For the serum/plasma sample analysis, the LC gradient time was extended to 180 min. For peptide identification, MASCOT (version 2.4.0, Matrix Science, London, United Kingdom) software was used. Peptides with a p-value < 0.05 (automatically estimated by MASCOT) were considered authentic.

Experiments were performed in quadruplicates and repeated 2–3 times. The initial comparisons were performed with non-parametric Wilcoxon method. The difference in value between control and treated samples were assessed with the Mann–Whitney U tests and represented as median values and quartiles (25/75%). Comparisons with a p-value ≤ 0.05 were considered statistically significant. Statistical analyses were performed using SPSS Statistics package (SPSS Inc., Chicago, IL, United States).

For MIC detection S. pneumoniae D-39, serotype 3, 19A, 19F, and clinical strain 11 were grown with different concentrations of AME/CME (2–512 μg). Bacterial growth was detected by measuring absorbance at 600 nm. The bacteria growth were further confirmed by spreading 10 μL samples from each well on MH agar plate. The results showed no growth of S. pneumoniae D-39 and serotype 3 at AME/CME concentrations of 256 μg and above. Similarly, the growth of S. pneumoniae serotypes 19A and 19F were inhibited at 128 μg of AME/CME. The clinical strain 11 showed no growth at 32 μg of AME/CME. Therefore, the MICs of AME and CME for S. pneumoniae D-39, S. pneumoniae serotype 3, 19A, 19F and clinical strain 11 were 256, 256, 128, 128, and 32 μg, respectively.

To assess the effect of AME/CME on S. pneumonia biofilm growth in vitro, biofilms were grown on microtiter plates with different concentration of AME/CME. The CV microtiter plate assay results revealed a significant (p < 0.05) decrease in biofilm biomass in AME/CME-supplemented samples (Figures 1A,B). The inhibitory effect of AME/CME on pneumococcal biofilms was concentration dependent. At sub-MIC (128 μg) concentrations of the AME/CME supplement, the pneumococci were unable to form robust biofilms, and the biomass decreased by >50%. Similarly, the viable bacteria detected by resazurin staining also decreased with increasing concentrations of AME/CME (Figure 1C). The bacteria appeared unable to grow in the presence of AME/CME, and were not able to establish organized biofilms. These results indicate that AME/CME inhibits in vitro biofilm formation.

The EPS matrices of pneumococcal biofilms contain proteins, e-extracellular DNA, and polysaccharides, and are difficult to eradicate using conventional antibiotics (Moscoso et al., 2006). We evaluated the ability of AME/CME to eradicate pre-established S. pneumoniae biofilms. The results showed strong biofilm eradication activity of AME/CME against pneumococcal biofilms. The biomasses of pre-established pneumococcal biofilms was significantly (p < 0.05) reduced by AME/CME treatment at the MIC (256 μg) and 2 × MIC (512 μg) (Figures 2A,B). Treatment with 2 × MIC (512 μg) AME/CME significantly (p < 0.05) reduced in vitro biofilms > 50%. Similarly, the viable bacteria counts within biofilms also decreased significantly after 6 h of incubation with 256 μg (MIC), 512 μg (2 × MIC) and 1024 μg (4 × MIC) of AME/CME (Figure 2C). AME/CME treatment at 4 × MIC significantly (p < 0.05) decreased viable bacteria by more than 2 log 10 steps.

Strong biofilm eradication activity of AME/CME at concentration of 4 × MIC (1024 μg) against clinical strains of S. pneumoniae was also detected (Figure 3). These results indicated that AME and CME possess strong pneumococci biofilm eradication activity.

Amniotic membrane extract and chorionic membrane extract is a natural cell membrane extract from the human amnion; therefore, these extracts were expected to show significant bacteriostatic effects. We sought to test the bactericidal effects of AME/CME on pneumococci. A time-kill analysis was performed on bacteria in the exponential growth phase. Treatment of log phase bacteria with 512 μg (2 × MIC) of the extracts exhibited a strong bactericidal effect on S. pneumoniae. At 0 h, viable bacteria in the control and treated samples were similar (5.15 × 10⁷); however, 1 h after AME/CME treatment, the viable bacteria decreased significantly (to 6.20 × 10⁶). The viability of bacteria continued to decrease with incubation time, and at 3 and 6 h of AME/CME treatment, viable bacteria were reduced by more than two and three logs, respectively (Figure 4). These results demonstrated the bactericidal activity of AME/CME against pneumococci.

Streptococcus pneumoniae growth was inhibited by 256 μg AME or CME. Similarly, the P-S antibiotic solution at a concentration of 0.2 μg completely inhibited bacterial growth. Therefore, the individual MICs of AME or CME and the P-S antibiotic solution on S. pneumoniae were 256 and 0.2 μg, respectively. In combination (AME+ P-S antibiotic or CME+ P-S antibiotic), the MICs of AME or CME and P-S antibiotic solution decreased to 64 and 0.05 μg, respectively. No bacterial growth inhibition was detected with AME or CME alone at a concentration of 64 μg (0.25 × MIC). Similarly, no growth inhibition was detected with 0.05 μg (0.25 × MIC) of P-S antibiotic solution when applied alone. However, the bacterial growth was inhibited by combining AME or CME at concentration of 64 μg (0.25 × MIC) and P-S antibiotic solution at concentration of 0.05 μg (0.25 × MIC) (Figure 5). According to the definition of FICs, the FIC value of AME or CME was 0.25 [MIC in combination (64 μg)/individual MIC (256 μg). Similarly, the FIC value of P-S antibiotic solution was 0.25 (0.05 μg/0.2 μg). Therefore, the FICI of AME or CME and P-S antibiotic solution in combination was 0.5, indicating a synergistic effect (as defined by an FICI ≤ 0.5). This result implied that in combination AME or CME and P-S antibiotic solution (AME/CME solution) exhibited a synergistic effect against S. pneumoniae.

The synergistic experiment results showed that the AME or CME in combination with P-S antibiotic solutions inhibited bacterial growth at low concentration. Therefore, we prepared AME or CME solution by combining AME or CME and Penicillin-streptomycin solution and evaluated biofilm inhibition and biofilm eradication activity. The results showed that increasing concentrations of the AME+P-S or CME+P-S solution significantly (p < 0.05) decreased biofilm growth in vitro. A significant (p < 0.05) reduced biofilm biomass and viable bacteria were detected at 4.003 μg (0.0625 × MIC) of AME/CME+P-S solutions and the inhibitory effect of AME/CME+P-S solutions was concentration dependent (Figures 6A–C). The AME/CME+P-S solution was also effective to eradicate pre-established biofilms of S. pneumoniae. Treatment of pre-established biofilm with 64.05 μg (MIC) and 128.1 μg (2 × MIC) of AME/CME+P-S solutions significantly (p < 0.05) reduced biofilm biomass (Figures 6D,E). Similarly, viable bacteria were also decreased in AME or CME+P-S solution treated biofilms (Figure 6F). These results indicate that the AME/CME+P-S solution is an effective antimicrobial solution for S. pneumoniae biofilms.

The confocal microscopy results showed that the untreated biofilms were compact and multilayered and the cells were connected to each other to form an organized three-dimensional structure (Figure 7A). However, the biofilms treated with AME/CME solution were scattered and attached to the bottom of the plate (Figures 7B,C). The cell-to-cell connections were not well established, and the biofilms were dispersed.

In this study, the SEM analysis showed compact biofilms in the control samples. The cells were well connected to each other and the bottom of the plate, forming three-dimensional biofilms (Figures 8A–C). The cells in the biofilms produced extracellular polymeric substances that were visible on and between the cells. Conversely, the biofilms treated with AME/CME solution were scattered, irregular, and disrupted (Figures 8D–I). In these biofilms, only a few cells that were scattered on the bottom of the plate were visible. The cell-to-cell connections were not well established, and the biofilms were disorganized. AME/CME solution treatment appeared to disrupt cell-to-cell connections and dismantle biofilm matrices. These results provided further evidence that AME/CME demonstrates a potent biofilm eradication activity.

The inhibitory effects of AME solution on in vivo colonization was studied using a rat OM model. S. pneumoniae D-39 was inoculated into the middle ears of rats, and then bacteria were enumerated after 1 week and the mucosa was visualized under SEM. The results showed that rat bullae inoculated with bacteria only swelled and thickened; the bullae were filled with dense cell debris and biofilm-like material (Figure 9A). However, no visible cell-debris or biofilm-like structures were detected in the rat bullae inoculated with bacteria + AME solution (Figure 9B). Similarly, in vehicle control bullae, no biofilms or cell debris were detected and there was no mucosal swelling (Figure 9C). The mean cfu counts of bacteria in the middle ears inoculated with bacteria only and bacteria with AME were 1.13 × 10⁴ and 1.79 × 10³, respectively. In rat bulla inoculated with AME, significantly (p < 0.05) 84% less bacteria were recovered (Figure 9D).

The colonization of S. pneumoniae in the middle ear mucosa of rats was viewed using SEM (Figure 10). The middle ears treated with the vehicle control were clean, without any cell debris in the mucosa and with intact cilia (Figures 10A,B). The rat bullae inoculated with bacteria only were filled with biofilm-like structures, the mucosa was covered with biofilm debris, and cilia were not visible (Figures 10C,D). The biofilm EPS deposited on the tips of cilia concealed the cilia. In contrast, the rat bullae supplemented with AME solution were relatively clean, and there was no visible biofilm-like structure or cell debris on the middle ear mucosa. No visible cell debris was deposited on the cilia, although the cilia of the epithelium appeared conglomerated (Figures 10E,F).

The proteomics analysis of AME/CME revealed several antimicrobial proteins/peptides (Table 1). The gene ontology (GO) analysis indicated that the identified proteins and peptides are involved in hydrolase, protease, and ribonuclease activities. The pathway analysis of these proteins revealed that, in humans, these proteins/peptides are also involved in chemotaxis and T cells and B cell activity in response to bacterial infection (Figure 11). The major protein/peptides included seven ribonucleases (ribonuclease T2, ribonuclease K6, ribonuclease 7, ribonuclease H2, ribonuclease pancreatic, and ribonuclease 5 [angiogenin]). The lactoferricins, which are antimicrobial peptides present in lactoferrin proteins, were also detected in AME/CME. A lysozyme with recognized muramidase or N-acetylmuramide glycanhydrolase activity was present in the AME/CME extracts. The other antimicrobial proteins and peptides present were dermcidin, a skin peptide, granulysin, the proteins S100-A9 and S100-A8, beta-2 microglobulin, and antileukoproteinase. The other proteins present in the extracts with antimicrobial activity were histone H2B type 1-D, type 1-O, HRA-V, and H1-4,and human beta defensin-3 protein.