The purpose of this in vitro study was to evaluate the antibacterial effect of a novel non-resorbable, bioactive polymeric nanostructured membrane (NMs), when doped with zinc, calcium and doxycycline.
A validated in vitro subgingival biofilm model with six bacterial species ( Streptococcus oralis, Actinomyces naeslundii, Veillonela parvula, Fusobacterium nucleatum, Porphyromonas gingivalis and Aggregatibacter actinomycetemcomitans ) was used. The experimental NMs, with and without being doped with doxycycline, calcium and zinc, were placed on hydroxyapatite (HA) discs. As positive control membranes, commercially available dense polytetrafluoroethylene (d-PTFE) membranes were used and, as negative controls, the HA discs without any membrane. The experimental, positive and negative control discs were exposed to a mixed bacterial suspension, at 37 °C under anaerobic conditions, during 12, 24, 48 and 72 h. The resulting biofilms were analyzed through scanning electron microscopy (SEM), to study their structure, and by quantitative polymerase chain reaction (qPCR), to assess the bacterial load, expressed as colony forming units (CFU) per mL. Differences between experimental and control groups were evaluated with the general linear model and the Bonferroni adjustment.
As shown by SEM, all membrane groups, except the NMs with doxycycline, resulted in structured biofilms from 12–72 hours. Similarly, only the membranes loaded with doxycycline demonstrated a significant reduction in bacterial load during biofilm development, when compared with the control groups (p < 0.001).
Doxycycline-doped nanostructured membranes have an impact on biofilm growth dynamics by significant reducing the bacterial load.
Periodontitis is a chronic inflammatory disease associated with bacterial dysbiosis and characterized by progressive destruction of the tooth-supporting structures [ ]. Severe periodontitis, the form with the highest morbidity, is the sixth most frequent disease worldwide, affecting 11.2% of the population [ ]. This disease not only can lead to tooth loss, but has a relevant impact on quality of life [ ] and may influence the incidence and progression of different systemic diseases [ ], and even overall death [ ]. While the aetiology of this chronic inflammatory disease is multifactorial, subgingival bacteria organised in biofilms are accepted as the primary etiological factor [ ].
The treatment of periodontitis is mainly based on infection control measures, by mechanically disrupting and reducing the sub-gingival bacterial biofilm [ ]. In specific periodontal lesions (intrabony defects), however, different regenerative treatments have shown the ability to regenerate the periodontal attachment apparatus, although this outcome is not always predictable [ ]. Among these regenerative technologies, guided tissue regeneration (GTR), based on the concept of tissue compartmentalization and homing, has demonstrated the achievement of periodontal regenerative outcomes [ , ], although these were often compromised by bacterial colonization and infection, mainly when these barrier membranes were exposed [ ]. Hence, a new generation of barrier membranes with enhanced properties is being developed. Among these, nanostructured membranes have been proposed due to their ability to enhance cell adhesion, migration, proliferation and cell differentiation, thus promoting regenerative outcomes [ , , ]. Furthermore, the presence of carboxylic groups (COOH) on their surface [ ] allows for the ability to chelate metal ions and other substances, such as antibiotics, conferring these biomaterials with antimicrobial properties. Metal cations, such as zinc and calcium, have shown certain antibacterial effect [ , ] and the use of doxycycline has shown multiple beneficial effects beyond its antimicrobial activity [ , ].
It was, therefore, the aim of this in vitro investigation, to evaluate the antibacterial activity of a biocompatible, non-resorbable polymeric and nanostructured membrane [ ], with and without doping agents (zinc, calcium or doxycycline), and to compare it with positive and negative control membranes, using a validated multi-species subgingival biofilm model.
Materials and methods
Bacterial strains and culture conditions
The following reference strains were used: Streptococcus oralis CECT 907 T, Veillonella parvula NCTC 11810, Actinomyces naeslundii ATCC 19039, Fusobacterium nucleatum DMSZ 20482, Aggregatibacter actinomycetemcomitans DSMZ 8324 and Porphyromonas gingivalis ATCC 33277. These bacteria were grown on blood agar plates (Blood Agar Oxoid Nº 2; Oxoid, Basingstoke, UK), supplemented with 5% (v/v) sterile horse blood (Oxoid), 5.0 mg L −1 hemin (Sigma, St. Louis, MO, USA) and 1.0 mg L −1 menadione (Merck, Darmstadt, Germany) under anaerobic conditions (10% H 2 , 10% CO 2 , and balance N 2 ) at 37 °C for 24−72 hours.
Nano-structured membrane technology
Nano-structured membranes (NMs) were fabricated by electrospinning, using a novel polymeric blend (PolymBlend®, NanoMyP, Granada, Spain) optimized for the production of non-woven nanofiber mats. Novel membranes were fabricated by electrospinning with a mixture of (MMA)1-co-(HEMA)1 and (MA)3-co-(HEA)2. Electrospun fibers at NMs are randomly distributed and have a diameter of approximately 300 nm. Micro- and nano-pores are found between fibers, resembling the collagen structure of human bone [ ]. The membrane surface was modified by chemical functionalization and doped with calcium (Ca-NM), zinc (Zn-NM) or doxycycline (Dox-NM). This was done by pre-activating the NMs with a sodium carbonate buffer solution (333 mM; pH = 12.5) for 2 h and then washed with water, what resulted in the partial hydrolysis of the ester bonds and carboxyl groups on the surface. Surfaces were later doped by incubating the NMs with different aqueous solutions of CaCl2, ZnCl2 (containing zinc or calcium at 40 ppm, for 3 days) (pH 6.5) or doxycycline (40 mg ml-1 of doxycycline hyclate, during 4 h) at room temperature with continuous shaking [ ]. Morphology, topography and NMs structure were not altered after doping [ ]. The commercially available control membrane (Cytoplast®, Osteogenic Biomedical, Lubbock, TX, USA) is made of dense polytetrafluoroethylene (d-PTFE). These membranes are not nanostructured, but they are arranged in symmetrical rows of regular hexagons with sides 200 microns long, separated one to the other about 0.5 mm.
Calcium hydroxyapatite (HA) discs of 7 mm in diameter and 1.8 mm in thickness (standard deviation, SD = 0.2) (Clarkson Chromatography Products, Williamsport, PA, USA) were covered with the tested NMs and also with the positive control membranes made of dense polytetrafluoroethylene (d-PTFE) (Cytoplast®, Osteogenic Biomedical, Lubbock, TX, USA). As negative control, the HA discs were coated with phosphate buffer saline (PBS) without any barrier membrane.
Pure cultures of each bacterium which were grown anaerobically in brain-heart infusion (BHI) (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) culture medium supplemented with 2.5 g L −1 mucin (Oxoid), 1.0 g L −1 yeast extract (Oxoid), 0.1 g L −1 cysteine (Sigma), 2.0 g L −1 sodium bicarbonate (Merck), 5.0 mg L −1 hemin (Sigma, St. Louis, MO, USA) and 1.0 mg L −1 menadione (Merck, Darmstadt, Germany) and 0.25% (v/v) glutamic acid (Sigma). At mid-exponential phase of bacterial growth (measured by spectrophotometry), a mixed bacterial suspension containing 10 3 colony forming units (CFU) mL −1 of S. oralis , 10 5 CFU mL −1 of V. parvula and A. naeslundii , and 10 6 CFU mL −1 of F. nucleatum, A. actinomycetemcomitans and P. gingivalis was prepared.
HA discs covered with either NMs (NMs, Zn-NMs, Ca-NMs and Dox-NMs), positive (d-PTFE) and negative control discs were located in the wells of a 24-well tissue culture plate (Greiner Bio-one, Frickenhausen, Germany), and 1.5 mL of the mixed bacterial suspension were spilled in each well. Plates were incubated in anaerobic conditions at 37 °C for 12, 24, 48 and 72 h. To control for appropriate sterility, plates containing only culture medium were also incubated.
Biofilms structural analysis with scanning electron microscopy (SEM) and energy dispersion X-Ray (EDX)
The resulting 12–72 hours biofilms were studied by SEM. The SEM processing of the discs included, first, fixation with a solution of 4% paraformaldehyde and 2.5% glutaraldehyde for 4 h at 4 °C, then washing in PBS, then in sterile water for 10 min and dehydrated through a series of graded ethanol solutions (30, 50, 70, 80, 90 and 100%; immersion time per series 10 min). Once dried by critical point, the discs were sputter-coated with gold and analysed with a SEM with a back-scattered electron detector and an image resolution of 25 kV (JSM6400; JEOL, Tokyo, Japan). In selected specimens, Energy Dispersion X-Ray (EDX) (JSM6400; JEOL, Tokyo, Japan) was also used.
Bacterial analysis by quantitative polymerase chain reaction (qPCR)
Cultured HA discs, with or without membranes, were collected in sterile tubes with 1 mL of PBS. The bacterial samples were collected through vigorous shaking with vortex for 3 min in order to optimize DNA bacterial collection from the biofilm samples [ ]. DNA of 12, 24, 48- and 72 -hs growth biofilms was isolated using a commercial kit, following manufacturer’s instructions (MolYsis Complete5; Molzym GmgH & CoKG, Bremen, Germany), and then quantified using the hydrolysis probe 5´nuclease assay qPCR method with specific primers for the six bacterial species used. These primers and probes [obtained by Life Technologies Invitrogen (Carlsbad, CA, USA) and Roche (Roche Diagnostic GmbH; Mannheim, Germany)] were targeted against 16S rRNA gene [ ]. The qPCR amplification was performed in a total reaction mixture volume of 10 μL. The reaction mixtures contained 5 μL of 2x master mixture (LC 480 Probes Master; Roche), optimal concentrations of primers and probe (900, 900 and 300 nM for S. oralis ; 300, 300 and 300 nM for A. naeslundii ; 750, 750 and 400 nM for V. parvula ; 300, 300 and 200 nM for A. actinomycetemcomitans ; 300, 300 and 300 nM, for P. gingivalis and 600, 600 and 300 nM for F. nucleatum ), and 2 μL of DNA from samples. The negative control was 2 μL of sterile water [no template control (NTC)] (Water PCR grade, Roche). The samples were subjected to an initial amplification cycle of 95 °C for 10 min, followed by 45 cycles at 95 °C for 15 s and 60 °C for 1 min. Analyses were performed with a LightCycler® 480 II thermocycler (Roche). LightCycler® 480 Multiwell Plates 384 and sealing foils were used (Roche).
Each DNA sample was analyzed in duplicate. Quantification cycle (Cq) was determined using the provided software package (LC 480 Software 1.5; Roche). Quantification of cells by qPCR was based on standard curves. The correlation between Cq values and CFU mL −1 were automatically generated through the software (LC 480 Software 1.5; Roche).
Data were expressed as means and SD. The distribution and normality were tested by the Shapiro–Wilk goodness-of-fit tests. To compare the antimicrobial effects of the tested materials on the disc (bacterial load expressed on CFU mL −1 ), a general lineal model was constructed for each bacterial species ( S, oralis, A. naeslundii, V. parvula , A. actinomycetemcomitans , P. gingivalis and F. nucleatum ) and for the total bacteria using the method of maximum likelihood and Bonferroni corrections for multiple comparisons. Results were considered statistically significant at p < 0.05. A software package (IBM SPSS Statistics 22.0; IBM Corporation, Armonk, NY, USA) was used for all data analysis.
Structural analysis of biofilms by SEM/EDX
After 12 h of biofilm development, the HA discs not covered with a barrier membrane showed presence of bacteria on their surface as individual cells, bacterial chains, and multicellular mixed colonies (cell to cell coaggregation). Fusiform bacilli (corresponding to F. nucleatum ) could be observed forming a three-dimensional net, with different morphotypes of bacteria adhered to their surface ( Fig. 1 A-B). Twelve-hour biofilms on discs covered with NMs, Ca-NM S and Zn-NMs membranes, showed a similar bacterial biomass and structural organization as those formed on HA discs not covered with membranes (negative control) ( Fig. 1 C-H). The positive control discs, covered with d-PTFE membranes, showed similar biomass of bacteria although with a particular colonization pattern ( Fig. 1 K-L), forming a regular hexagon lattice architecture ( Fig. 2 ). After 12 h of biofilm growth, only the discs doped with doxycycline (Dox-NMs) demonstrated statistically significant differences, in terms of biomass and structure, since the presence of bacteria was hardly observed over the membrane surface ( Fig. 1 I-J).
After 24 h, the amount of the biofilm increased with the presence of bacterial colonization covering the entire disc area on test (NMs, Ca-NMs and Zn-NMS) and control discs, always with a similar biofilm structure. Only in the discs covered with Dox-NMs, biofilm formation was clearly reduced ( Fig. 3 A). In the positive control discs, the resulting biofilms maintained the influence of this membrane structure ( Fig. 3 B). After 48 h, all discs covered with NMs demonstrated a decreased bacterial mass when compared with negative (HA) and positive (d-PTFE) control discs ( Fig. 4 ).
After 72 h, a thick layer of bacteria covered all specimens, with F. nucleatum representing the most noticeable bacterial species, surrounded by multiple bacterial micro-colonies ( Fig. 5 ), acquiring the typical biofilm-structure on all tested membranes, with presence of bacterial organised in stacks and tunnels for nutrient circulation ( Fig. 5 ). These biofilms were completely mature in all tested groups excepted on Dox-NMs, with a significantly reduced bacterial load and a structure of a poorly structured biofilm ( Fig. 5 I-J). In these 72 -hs biofilms, EDX analysis confirmed the presence of mineral deposits of calcium and sodium salts on the membrane surface (Supplementary Fig. 1 ).
Quantitative evolution of the bacterial load and presence of specific bacteria
Table 1 depicts the effect of the different NMs on total bacterial counts, at different time points between 12 and 72 -hs biofilms. Until reaching the exponential phase (up to 48 h), biofilms on NMs demonstrated a lower total number of bacteria compared to control biofilms (HA discs). This effect was more noticeable in 48 -hs biofilms in contact to Ca-NMs (p = 0.003, as compared with negative HA control biofilms).
|Time of incubation||Number of bacteria [CFU / biofilm, expressed as mean (SD)]|
|12 h||3.69 × 10 8||1.28 × 10 8||2.53 × 10 8||1.19 × 10 8||2.98 × 10 6 *||1.26 × 10 8|
|(1.09 × 10 8 )||(7.15 × 10 7 )||(8.72 × 10 7 )||(4.30E x10 7 )||(2.62 × 10 6 )||(7.55 × 10 7 )|
|24 h||5.37 × 10 8||4.68 × 10 8||1.08 × 10 8||2.85 × 10 8||7.42 × 10 6 *||3.60E x10 8|
|(1.87 × 10 8 )||(2.46 × 10 7 )||(1.57 × 10 7 )||(1.17 × 10 8 )||(2.52 × 10 6 )||(1.38 × 10 8 )|
|48 h||1.14 × 10 9||3.17 × 10 8||2.01 × 10 8 *||6.52 × 10 8||2.82 × 10 6 *||3.33 × 10 8|
|(1.54 × 10 8 )||(1.40 × 10 8 )||(6.50 × 10 7 )||(1.57 × 10 8 )||(3.46 × 10 6 )||(1.79 × 10 8 )|
|72 h||2.59 × 10 8||4.71 × 10 8||6.62 × 10 8 †||8.70 × 10 8 †||1.28E x10 6 *†||1.75 × 10 8|
|(2.37 × 10 7 )||(3.00 × 10 8 )||(2.09 × 10 8 )||(5.57 × 10 8 )||(7.76 × 10 6 )||(6.47 × 10 7 )|
In mature biofilms (72 h), this effect was lost, with high numbers of bacteria present on the membrane nanofibers. Significantly higher numbers of bacteria were present on Ca-NMs and Zn-NMs when compared to positive control (d-PTFE) membranes (p = 0.050 and p = 0.015, respectively). Nevertheless, there was no statistically significant differences in the amounts of bacteria on Ca-NMs and Zn-NMs when compared with negative control HA discs (p ≤ 0.245). Similarly, bacterial load in contact with d-PTFE membranes and NMs without charge did not significantly differ with HA discs (p = 1.00 in both cases).
Dox-NMs maintained the antibacterial effect at all time points during biofilm formation and maturation. In contact to doxycycline, the total bacteria load was significantly lower in the biofilms after 12, 24, 48 and 72 h, when compared to the other groups (p < 0.001 in all cases).
The effect of the different tested membranes on each of the bacterial species during their biofilm growth is depicted in Table 2 . At 12 h, no statistically significant differences were observed among the different bacterial species growing in the tested membranes when compared with the negative control group (HA), excepted for Dox-NMs [p ≤ 0.005 in all cases, except for P. gingivalis (p ≤ 0.060)]. Within the NMs studied, statistically significant lower bacterial counts for the primary colonizers S. oralis , A. naeslundii and V. parvula and the secondary colonizers F. nucleatum and A. actinomycetemcomitans, were found in Dox-NMs, when compared with the other groups (p ≤ 0.005 in all cases), except for A. naeslundii when compared to Zn-NMs (p = 0.808).