Experimental Assessment of the Performance of Two Marine Coatings to Curb Biofilm Formation of Microfoulers

Biofilms formed on submerged marine surfaces play a critical role in the fouling process, causing increased fuel consumption, corrosion, and high maintenance costs. Thus, marine biofouling is a major issue and motivates the development of antifouling coatings. In this study, the performance of two commercial marine coatings, a foul-release silicone-based paint (SilRef) and an epoxy resin (EpoRef), was evaluated regarding their abilities to prevent biofilm formation by Cyanobium sp. and Pseudoalteromonas tunicata (common microfoulers). Biofilms were developed under defined hydrodynamic conditions to simulate marine settings, and the number of biofilm cells, wet weight, and thickness were monitored for 7 weeks. The biofilm structure was analyzed by confocal laser scanning microscopy (CLSM) at the end-point. Results demonstrated that EpoRef surfaces were effective in inhibiting biofilm formation at initial stages (until day 28), while SilRef surfaces showed high efficacy in decreasing biofilm formation during maturation (from day 35 onwards). Wet weight and thickness analysis, as well as CLSM data, indicate that SilRef surfaces were less prone to biofilm formation than EpoRef surfaces. Furthermore, the efficacy of SilRef surfaces may be dependent on the fouling microorganism, while the performance of EpoRef was strongly influenced by a combined effect of surface and microorganism.


Introduction
In marine environments, aquatic micro and macroorganisms-such as bacteria, algae, and invertebrates-spontaneously colonize submerged surfaces in a process known as biofouling [1,2]. This natural phenomenon has severe economic and environmental implications all over the world [1][2][3]. Fouling organisms attached to marine vessel hulls promote surface corrosion and increase frictional drag, leading to higher dry-docking periods and fuel consumption, respectively [1]. Additionally, biofouling can

Surface Preparation
Two commercial marine coatings, a silicone-based paint (SilRef) and an epoxy resin (EpoRef) were used in this study. Glass coupons (1 cm × 1 cm; Vidraria Lousada, Lda, Lousada, Portugal) were cleaned and disinfected [23] before being used as a substrate for coating. The silicone-based paint was provided by Hempel A/S (HEMPASIL X3+ 87500, Copenhagen, Denmark) and it is a two-component system obtained from blending a base resin (Hempasil Base 87509) and curing agent (Hempasil Crosslinker 98951) in a ratio of 17.8/2.2 (v/v). SilRef-coated glass surfaces were prepared using conventional brush painting, following the recommendations of the manufacturer: a prior glass coupons coating with a universal tie-coat (base resin XA17-17310 and curing agent XA18-RD003) and a curing step at room temperature (about 23 • C) for 8 h, followed by a second coating with the HEMPASIL system, and curing step at room temperature (about 23 • C) for 24 h. Finally, the cured SilRef-coated glass coupons were sterilized with UV radiation for 30 min before use.
The epoxy resin was produced by HB Química (Matosinhos, Porto, Portugal) and consisted of a mixture of HB Eposurf 2 resin and HB Eposurf hardener in a ratio of 10:3 (v/v) [23]. For the production of EpoRef-coated glass surfaces, 70 µL of epoxy resin was deposited on top of the glass coupons using spin coating (Spin150 Polos TM , Paralab, Porto, Portugal) at 6000 rpm, with increments of 1000 rpm, for 40 s. Then, the surfaces were dried in two steps: (i) 12 h at room temperature (approximately 25 • C), and (ii) 3 h at 60 • C, according to the instructions from the manufacturer [23], and sterilized by using immersion in 70% (v/v) ethanol (VWR International S.A.S., Fontenay-sous-Bois, France) for 20 min. Before the biofilm formation experiments, the initial weight of each coupon was determined.

Atomic Force Microscopy (AFM)
AFM studies were carried out using a PicoPlus scanning probe microscope interfaced with a Picoscan 2500 controller (both from Keysight Technologies, Santa Rosa, CA, USA). Each sample was imaged with a 100 × 100 µm 2 piezo-scanner. The surface roughness was determined in 40 × 40 µm 2 scanned areas in three randomly chosen locations per sample (total of three replicates) at room temperature. The SilRef and EpoRef surfaces were analyzed through contact mode, with a v-shape silicon tip, with a spring constant of 0.085 N/m and 0.284 N/m (AppNano, Mountain View, CA, USA), respectively. The scan speed was set at 1.0 L/s. The WSxM5.0 software (Nanotec Electronica, Feldkirchen, Germany) was used to perform the roughness surface measurements and to obtain the 2D images [24]. The roughness height parameter calculated was the average roughness (R a ).

Hydrophobicity
Surface hydrophobicity was evaluated according to the approach developed by Oss et al. [25]. Contact angles were determined at 25 ± 2 • C by the sessile drop method through a contact angle meter (Dataphysics OCA 15 Plus, Filderstadt, Germany), as fully described by Gomes et al. [26] in three independent assays. For each assay, at least 25 measurements were performed for each material.

Biofilm Formation
Biofilm experiments were performed using 12-well microtiter plates (VWR International, Carnaxide, Portugal) under controlled hydrodynamic conditions. Sterilized SilRef and EpoRef coupons were fixed to the microplate wells with double-sided adhesive tape [23]. A P. tunicata suspension at an optical density (OD 610 nm) of 0.1 (which corresponds to 10 8 cells/mL) was prepared from an overnight culture, while the cyanobacterial culture was centrifuged to remove any traces of the supplemented Z8 medium and the pellet was resuspended in fresh VNSS medium to a final concentration of 10 8 cell/mL. Subsequently, 3 mL of each cell suspension was added to the wells, and the plates were incubated at 25 • C in an orbital shaker with 25 mm diameter (Agitorb 200ICP, Norconcessus, Ermesinde, Portugal) at 185 rpm under alternate light cycles of 14 h light (10-30 mol photons/m 2 /s)/10 h dark. According to previous computational fluid dynamic studies performed by the group [23,32], the selected shaking frequency of 185 rpm produces an average shear rate of 40/s and a maximum of 120/s at the plate bottom, including for instance the shear rate estimated for a ship in a harbor (50/s) [33]. Biofilm formation was monitored for seven weeks (49 days) and sampled every seven days. During the incubation period, the culture medium was replaced twice a week. Three independent biofilm formation experiments (biological replicates) were performed with two technical replicates each (two coupons of SilRef or EpoRef).

Biofilm Analysis
Every seven days, the culture medium was carefully removed, and the coupons were gently washed with 3 mL of 0.85% (v/v) sterile saline solution to remove loosely attached microorganisms. Two coupons from each experimental condition were analyzed regarding (i) the number of biofilm cells, (ii) biofilm wet weight, and (iii) biofilm thickness. Additionally, the biofilm structure was analyzed at day 49 using confocal laser scanning microscopy (CLSM).

Biofilm Cell Counting and Wet Weight
For cell counting, coupons were taken out of the wells, dipped in 2 mL of 0.85% (v/v) sterile saline solution and vortexed for 3 min at maximum power to release biofilm cells. Then, 10 µL of each cell suspension was placed on each side of a Neubauer chamber and observed in a brightfield microscope (Nikon Eclipse LV100 microscope, Nikon Corporation, Tokyo, Japan).
To assess the wet weight of biofilms, coupons were removed from the wells with a tweezer and weighted on an analytical balance. The biofilm wet weight was determined by the difference between the initial weight of coupon (before inoculation) and the weight measured on the sampling day.

Biofilm Thickness
Biofilm thickness was assessed through optical coherence tomography (OCT) (Thorlabs Ganymede Spectral Domain Optical Coherence Tomography system, Thorlabs GmbH, Dachau, Germany) with a central wavelength of 930 nm. Before biofilms were imaged, the culture medium was carefully removed from the microplate wells, the coupons were gently washed, and the wells were filled with 3 mL of 8.5 g/L sodium chloride sterile solution. Since biofilms are essentially composed of water [34], the established refractive index was 1.40, close to the refractive index of water (1.33). At each sampling day, a minimum of five different fields of view (2D images) per material and Coatings 2020, 10, 893 5 of 13 microorganism were analyzed and captured. Image analysis was performed using a routine developed in the Image Processing Toolbox from MATLAB 8.0 and Statistics Toolbox 8.1 (The MathWorks, Inc., Natick, MA, USA), as described by Romeu et al. [35].

Confocal Laser Scanning Microscopy (CLSM)
At day 49, coupons containing the biofilms were removed from the microplates, washed with saline solution, and stained with 6 µM Syto9 (Thermo Fisher Scientific, Waltham, MA, USA)-a green cell-permeant nucleic acid marker-for 10 min at room temperature. Each stained sample was mounted on a microscopic slide and image acquisition was performed using a Leica TCS SP5 II Confocal Laser Scanning Microscope (Leica Microsystems, Wetzlar, Germany). All biofilms were scanned at 400 Hz using a 40× water objective lens (LEICA HCX PL APO CS 40.0×/1.10 WATER UV) with a 488-nm argon laser set at 25% intensity. The emitted fluorescence was recorded within the range of 500-600 nm. A minimum of five stacks of horizontal-plane images (512 × 512 pixels, corresponding to 387.5 µm × 387.5 µm) with a z-step of 1 µm were acquired per sample.
Three-dimensional (3D) projections of the biofilms were constructed from the CLSM acquisitions using the "Easy 3D" function of the IMARIS 9.1 software (Bitplane, Zurich, Switzerland). Quantitative structural parameters (biovolume and surface coverage) were extracted from confocal image series with the plug-in COMSTAT2 [36] run in ImageJ 1.48v software [37]. While the biovolume provides an estimate of the biomass in the biofilm (µm 3 /µm 2 ), the surface coverage corresponds to the percentage of surface area covered on the biofilm base.

Data Analysis
Descriptive statistics were used to calculate the mean and standard deviation for the different biofilm parameters (number of biofilm cells, wet weight, thickness, biovolume, and surface coverage).
Data analysis was performed using the GraphPad Prism ® for Windows, version 8 (GraphPad Software, Inc., San Diego, CA, USA). Since the variable distribution was normal, Student's t-test was used to compare biofilm formation between the two marine surfaces tested (SilRef and EpoRef) for both P. tunicata and Cyanobium sp. LEGE 10375 strains. The same statistical test was used to compare the biovolume and surface coverage of P. tunicata and Cyanobium sp. LEGE 10375 biofilms between the two tested surfaces at the end of the experiment (day 49). Significant results were considered for p-values < 0.1.
In order to ascertain the performance of SilRef and EpoRef marine coatings to curb biofilm formation of main microfoulers, the surface and microorganism influence were estimated for each parameter monitored over time (number of biofilm cells, wet weight, and thickness) and represented in radar charts. Radar charts were split into four quadrants and each one represents the average values obtained in each sampling point (days) for the following assays: P. tunicata biofilm formation on SilRef (Ps/SilR, top right quadrant); P. tunicata biofilm formation on EpoRef (Ps/EpoR, bottom right quadrant); Cyanobium sp. biofilm formation on EpoRef (Cya/EpoR, bottom left quadrant); and Cyanobium sp. biofilm formation on SilRef (Cya/SilR, top left quadrant). The influence of the microorganism on biofilm development was calculated by subtracting the values obtained for the two microfoulers for both SilRef (Ps/SilR vs. Cya/SilR) and EpoRef (Ps/EpoR vs. Cya/EpoR) surfaces, whereas the influence of the surface on biofilm development was determined by subtracting the values obtained for two different surfaces for P. tunicata (Ps/SilR vs. Ps/EpoR) and Cyanobium sp. (Cya/SilR vs. Cya/EpoR) microfoulers. All positive differences were considered as increments resulting from the microorganism or surface effect and represented by a colored area (microorganism effect-orange area; surface effect-red area). The combined effect (blue area) has been plotted whenever the surface effect overlapped the microorganism effect.

Results
In this study, the antibiofilm efficacy of two commercial marine coatings, a silicone-based paint (SilRef) and an epoxy resin (EpoRef), was determined through the analysis of Cyanobium sp. and P. tunicata biofilms developed on SilRef-and EpoRef-coated glass substrates for 49 days under hydrodynamic conditions that mimic the marine environment.
The SilRef and EpoRef surfaces were first analyzed concerning their hydrophobicity and roughness (Table 1). Both water contact angles (θ w ) and degree of hydrophobicity (∆G) indicated that the SilRef is more hydrophobic than the EpoRef surface (θ w SilRef = 108.4 • vs. θ w EpoRef = 69.4 • , Figure S1; ∆G SilRef = −55.8 mJ/m 2 vs. ∆G EpoRef = −26.7 mJ/m 2 ). The surface topography of SilRef and EpoRef surfaces was examined by AFM in contact mode ( Figure 1). The topography images revealed that both coatings are homogeneous and very smooth. In fact, the roughness of both surfaces is at a nanoscale, suggesting that there are no features in the surfaces to act as niches for the microfoulers. Nevertheless, the SilRef surface displayed higher roughness with an average value (R a ) of 49.7 nm, in opposition to R a = 12.9 nm for the EpoRef surface (Table 1). Table 1. Contact angles with water (θ w ), formamide (θ F ) and α-bromonaphthalene (θ B ), hydrophobicity (∆G) and roughness (R a ) determined for the SilRef and EpoRef surfaces. Values are presented as means ± standard deviations.

Roughness
Coatings 2020, 10, x FOR PEER REVIEW 6 of 14 The SilRef and EpoRef surfaces were first analyzed concerning their hydrophobicity and roughness (Table 1). Both water contact angles (θw) and degree of hydrophobicity (∆G) indicated that the SilRef is more hydrophobic than the EpoRef surface (θw SilRef = 108.4° vs. θw EpoRef = 69.4°, Figure S1; ∆G SilRef = −55.8 mJ/m 2 vs. ∆G EpoRef = −26.7 mJ/m 2 ). The surface topography of SilRef and EpoRef surfaces was examined by AFM in contact mode (Figure 1). The topography images revealed that both coatings are homogeneous and very smooth. In fact, the roughness of both surfaces is at a nanoscale, suggesting that there are no features in the surfaces to act as niches for the microfoulers. Nevertheless, the SilRef surface displayed higher roughness with an average value (Ra) of 49.7 nm, in opposition to Ra = 12.9 nm for the EpoRef surface (Table 1).  Results of biofilm analysis-including the number of biofilm cells, wet weight, and thicknessare presented in Figure 2. In general, regardless of the surface material, the cyanobacterial strain has higher biofilm-forming capacity than P. tunicata. This is particularly evident in the last two weeks of the experiment (between days 35 and 49), in which the wet weight and mean thickness values obtained for the cyanobacterial biofilms (Figure 2b,c) were about 26% higher than those of bacterial biofilms (Figure 2e,f). On the other hand, it is possible to globally observe that, while the biofilm parameters for cyanobacteria (in particular, the wet weight and thickness) followed an increasing tendency over time, there was a certain stagnation of the biofilm values for the marine bacteria, regardless of the surface (Figure 2). Results of biofilm analysis-including the number of biofilm cells, wet weight, and thickness-are presented in Figure 2. In general, regardless of the surface material, the cyanobacterial strain has higher biofilm-forming capacity than P. tunicata. This is particularly evident in the last two weeks of the experiment (between days 35 and 49), in which the wet weight and mean thickness values obtained for the cyanobacterial biofilms (Figure 2b,c) were about 26% higher than those of bacterial biofilms (Figure 2e,f). On the other hand, it is possible to globally observe that, while the biofilm parameters for cyanobacteria (in particular, the wet weight and thickness) followed an increasing tendency over time, there was a certain stagnation of the biofilm values for the marine bacteria, regardless of the surface ( Figure 2). Looking at the cell number of Cyanobium sp. biofilms (Figure 2a), there was a growing trend for EpoRef surfaces from day 7 (with 8.39 × 10 7 cells/cm 2 ) to the end of the experiment (with 1.25 × 10 9 cells/cm 2 ). The beneficial effect of the SilRef surfaces was particularly evident towards the end of the experimental period, when a reduction in the cell number of about 26% was registered compared to EpoRef surfaces for day 49 (p < 0.05, Figure 2a). For P. tunicata biofilms (Figure 2d), the number of cells increased until day 21 and then kept more or less constant for both surfaces (on average, 7.66 × 10 8 cells/cm 2 ). Again, at day 49, the SilRef-coated surfaces presented 17% less cells than the EpoRef surfaces (p < 0.1, Figure 2b).
The wet weight of Cyanobium sp. biofilms developed either on SilRef or EpoRef surfaces ( Figure  2b) increased until day 35. In all sampling points, statistically significant differences were found between the wet weight of biofilms developed on EpoRef and SilRef surfaces. Indeed, on SilRef surfaces, Cyanobium sp. biofilms weighted, on average, 33% less than the biofilms formed on EpoRef surfaces. The wet weight of P. tunicata biofilms (Figure 2e) increased on both surfaces until day 35 and then remained constant, with exception of biofilms formed on SilRef surfaces whose wet weight decreased on day 49. An average wet weight reduction of 35% was obtained for biofilms developed on SilRef compared to those formed on EpoRef surfaces (Figure 2e).
The mean thickness of Cyanobium sp. biofilms generally increased until day 49 ( Figure 2c) and biofilms formed on SilRef surfaces had, on average, 27% lower thickness than those developed on EpoRef surfaces throughout the experiment. This difference was particularly visible on day 49, when the biofilm formed on SilRef had almost half the mean thickness of the biofilm grown on EpoRef surfaces (p < 0.01, Figure 2c). The thickness of P. tunicata biofilms also increased over time (Figure 2f), stabilizing on day 42 for biofilms developed on SilRef surfaces. On average, P. tunicata biofilms developed on SilRef surfaces were approximately 15% thinner than those developed on EpoRef surfaces.
The relative importance of surface properties and microorganism type on biofilm formation was estimated for each analyzed parameter ( Figure 2) and graphically represented in Figure 3 where only increases in the assayed parameters are depicted. For Cyanobium sp. biofilms developed on SilRef surfaces (top left quadrant), the microfouler organism had a strong influence on the increase of biofilm cells, wet weight, and thickness (Figure 3a-c), as represented by the orange area. Looking at the cell number of Cyanobium sp. biofilms (Figure 2a), there was a growing trend for EpoRef surfaces from day 7 (with 8.39 × 10 7 cells/cm 2 ) to the end of the experiment (with 1.25 × 10 9 cells/cm 2 ). The beneficial effect of the SilRef surfaces was particularly evident towards the end of the experimental period, when a reduction in the cell number of about 26% was registered compared to EpoRef surfaces for day 49 (p < 0.05, Figure 2a). For P. tunicata biofilms (Figure 2d), the number of cells increased until day 21 and then kept more or less constant for both surfaces (on average, 7.66 × 10 8 cells/cm 2 ). Again, at day 49, the SilRef-coated surfaces presented 17% less cells than the EpoRef surfaces (p < 0.1, Figure 2b).
The wet weight of Cyanobium sp. biofilms developed either on SilRef or EpoRef surfaces (Figure 2b) increased until day 35. In all sampling points, statistically significant differences were found between the wet weight of biofilms developed on EpoRef and SilRef surfaces. Indeed, on SilRef surfaces, Cyanobium sp. biofilms weighted, on average, 33% less than the biofilms formed on EpoRef surfaces. The wet weight of P. tunicata biofilms (Figure 2e) increased on both surfaces until day 35 and then remained constant, with exception of biofilms formed on SilRef surfaces whose wet weight decreased on day 49. An average wet weight reduction of 35% was obtained for biofilms developed on SilRef compared to those formed on EpoRef surfaces (Figure 2e).
The mean thickness of Cyanobium sp. biofilms generally increased until day 49 ( Figure 2c) and biofilms formed on SilRef surfaces had, on average, 27% lower thickness than those developed on EpoRef surfaces throughout the experiment. This difference was particularly visible on day 49, when the biofilm formed on SilRef had almost half the mean thickness of the biofilm grown on EpoRef surfaces (p < 0.01, Figure 2c). The thickness of P. tunicata biofilms also increased over time (Figure 2f), stabilizing on day 42 for biofilms developed on SilRef surfaces. On average, P. tunicata biofilms developed on SilRef surfaces were approximately 15% thinner than those developed on EpoRef surfaces.
The relative importance of surface properties and microorganism type on biofilm formation was estimated for each analyzed parameter ( Figure 2) and graphically represented in Figure 3 where only increases in the assayed parameters are depicted. For Cyanobium sp. biofilms developed on SilRef surfaces (top left quadrant), the microfouler organism had a strong influence on the increase of biofilm cells, wet weight, and thickness (Figure 3a-c), as represented by the orange area. Additionally, a subtle combined effect was observed between the surface and microorganism (blue area) that contributed to a higher number of cells at an early stage of Cyanobium sp. biofilm formation (between day 7 and 28) Coatings 2020, 10, 893 8 of 13 ( Figure 3a). For Cyanobium sp. biofilms formed on EpoRef surfaces (bottom left quadrant), a combined effect between the surface and microorganism (blue area) was responsible for a higher number of biofilm cells, wet weight, and thickness (Figure 3a-c). This effect was more noticeable from day 35 of biofilm formation. Although the microorganism type affected biofilm formation, the combined effect was stronger (blue area > orange area) (Figure 3a-c). Moreover, the surface effect (red area) contributed to an increase of Cyanobium sp. biofilm wet weight at an early stage of its development (Figure 3b).
for a higher number of biofilm cells, wet weight, and thickness (Figure 3a-c). This effect was more noticeable from day 35 of biofilm formation. Although the microorganism type affected biofilm formation, the combined effect was stronger (blue area > orange area) (Figure 3a-c). Moreover, the surface effect (red area) contributed to an increase of Cyanobium sp. biofilm wet weight at an early stage of its development (Figure 3b).
For P. tunicata biofilms formed on SilRef surfaces (top right quadrant), an increase in the assayed parameters is not seen due to the effect of the microorganism type (orange area) as only a slight increase in wet weight can be seen in Figure 3b. In turn, for P. tunicata biofilms formed on EpoRef surfaces (bottom right quadrant), a strong influence of the surface (red area) was visible, contributing to a higher number of biofilm cells, wet weight, and thickness (Figure 3a-c). Additionally, the combined effect between the surface and microorganism had a small influence on the biofilm wet weight at an early development stage (days 7 and 14) (Figure 3a). The structural differences between the Cyanobium sp. and P. tunicata single-species biofilms formed on the two tested surfaces at the end of the experiment (49 days) were evaluated using CLSM. Examples of 3D biofilm reconstructions are presented in Figure 4. It is possible to observe that denser and thicker biofilms grew on the epoxy resin, regardless of the microorganism (Figure 4a,c), which confirms the results of the biofilm cell number (Figure 2a,b) and thickness (Figure 2c,d) obtained by brightfield microscopy and OCT, respectively. On the contrary, biofilms formed on the silicone-based coating were thinner and did not cover the entire surface area (Figure 4b,d) due to the formation of large cell aggregates, which are particularly visible in the P. tunicata biofilm (Figure 4d). In fact, while in the EpoRef surfaces, both Cyanobium sp. and P. tunicata biofilms displayed a surface coverage above 70%, it was only around 45% in the SilRef surfaces (p < 0.01, Figure 5b,d). With regard to For P. tunicata biofilms formed on SilRef surfaces (top right quadrant), an increase in the assayed parameters is not seen due to the effect of the microorganism type (orange area) as only a slight increase in wet weight can be seen in Figure 3b. In turn, for P. tunicata biofilms formed on EpoRef surfaces (bottom right quadrant), a strong influence of the surface (red area) was visible, contributing to a higher number of biofilm cells, wet weight, and thickness (Figure 3a-c). Additionally, the combined effect between the surface and microorganism had a small influence on the biofilm wet weight at an early development stage (days 7 and 14) (Figure 3a).
The structural differences between the Cyanobium sp. and P. tunicata single-species biofilms formed on the two tested surfaces at the end of the experiment (49 days) were evaluated using CLSM. Examples of 3D biofilm reconstructions are presented in Figure 4. It is possible to observe that denser and thicker biofilms grew on the epoxy resin, regardless of the microorganism (Figure 4a,c), which confirms the results of the biofilm cell number (Figure 2a,b) and thickness (Figure 2c,d) obtained by brightfield microscopy and OCT, respectively. On the contrary, biofilms formed on the silicone-based coating were thinner and did not cover the entire surface area (Figure 4b,d) due to the formation of large cell aggregates, which are particularly visible in the P. tunicata biofilm (Figure 4d). In fact, while in the EpoRef surfaces, both Cyanobium sp. and P. tunicata biofilms displayed a surface coverage above 70%, it was only around 45% in the SilRef surfaces (p < 0.01, Figure 5b,d). With regard to biovolumes (Figure 5a,c), they were also significantly higher for the EpoRef surfaces when compared to the SilRef surfaces. In the case of cyanobacteria, the biovolume determined for the EpoRef surfaces was about 3 times higher than that found for the SilRef surfaces after 49 days of biofilm development. This is Coatings 2020, 10, 893 9 of 13 correlated to the higher cell density and weight of cyanobacterial biofilms on the EpoRef coating at the experimental end-point (Figure 2a,b).
Coatings 2020, 10, x FOR PEER REVIEW 9 of 14 biovolumes (Figure 5a,c), they were also significantly higher for the EpoRef surfaces when compared to the SilRef surfaces. In the case of cyanobacteria, the biovolume determined for the EpoRef surfaces was about 3 times higher than that found for the SilRef surfaces after 49 days of biofilm development. This is correlated to the higher cell density and weight of cyanobacterial biofilms on the EpoRef coating at the experimental end-point (Figure 2a,b).   Coatings 2020, 10, x FOR PEER REVIEW 9 of 14 biovolumes (Figure 5a,c), they were also significantly higher for the EpoRef surfaces when compared to the SilRef surfaces. In the case of cyanobacteria, the biovolume determined for the EpoRef surfaces was about 3 times higher than that found for the SilRef surfaces after 49 days of biofilm development. This is correlated to the higher cell density and weight of cyanobacterial biofilms on the EpoRef coating at the experimental end-point (Figure 2a,b).

Discussion
In this study, the long-term performance of two commercial marine coatings, a silicone-based paint and an epoxy resin, to prevent biofilm formation by common microfoulers was demonstrated through the kinetic evaluation of different biofilm parameters, including the number of biofilm cells, wet weight, and thickness. For this purpose, Cyanobium sp. and P. tunicata biofilms were developed on both surfaces for seven weeks under controlled hydrodynamic conditions that mimic those found in some marine settings. As the hydrodynamic forces bring a strong impact on marine biofilm development [23,32,38,39], the experiments were performed under specific shear forces that include the shear rate estimated for a ship in a harbor (50/s) [33] in order to increase the predictive value of the results.
At an initial development stage (until day 28), Cyanobium sp. biofilms formed on SilRef-coated surfaces presented, on average, a higher number of cells compared to biofilms formed on EpoRef surfaces. However, from day 35 onwards, the number of biofilm cells stabilized on SilRef surfaces, while on EpoRef surfaces this number increased up to day 49. A similar biofouling response was detected for P. tunicata biofilms. Until day 28, the number of cells of biofilms formed on SilRef surfaces was slightly higher than those obtained for biofilms formed on EpoRef. However, on SilRef surfaces, the biofilm cell number was maintained, while on EpoRef surfaces, it increased up to day 49. Based on this parameter, the EpoRef surfaces effectively inhibited biofilm formation at initial stages (until day 28), while SilRef surfaces showed higher antifouling activity during biofilm maturation (from day 35 onwards). Indeed, surface characterization analysis demonstrated that the EpoRef surface is less hydrophobic and rough, which may explain its higher performance at initial stages of biofilm formation. In opposition, the higher hydrophobicity and roughness of SilRef may have hindered their initial antifouling activity. However, despite several studies reporting a correlation between the surface hydrophobicity and roughness and cell attachment [40,41], there is evidence that the biofilm formation induces changes in the surface properties [42,43]. These findings corroborate the higher antifouling activity of SilRef surfaces during the biofilm maturation stages.
SilRef is a commercial coating, and accordingly with its technical properties, is based on a dual-mode of action, a 'non-stick' ability and a foul-release effect, allied to its relative higher elasticity (Young's modulus). These properties may contribute to decreasing fouling settlements and cell cohesion interactions [8,10,33], as confirmed by using confocal microscopy for both studied marine strains. Besides, this dual-action has improved the FRC original properties by the hydrogel contribution in its singular hydrogel-PDMS polymeric matrix, which acquires a more hydrophilic character upon contact with water. Thus, it can prevent either weaker hydrophobic or hydrophilic interactions, which can delay the adsorption of proteins, bacteria, and subsequent fouling [11,12,15].
Recently, Faria et al. demonstrated the impact of the surface properties on marine biofilm formation under controlled hydrodynamic conditions [23]. Indeed, surface properties may promote cell-surface interactions that, even in a low magnitude, may facilitate microfouler retention and contribute to biofilm development [44].
Although there was a shift between the number of cells of biofilms formed on SilRef and EpoRef surfaces that may be dependent on the biofilm age, wet weight and thickness analysis revealed that, in all development stages, Cyanobium sp. and P. tunicata biofilms formed on SilRef surfaces showed lower weight and were thinner than those developed on EpoRef surfaces. These results should be interpreted considering that a biofilm is an organized community of aggregated cells, living at an interface between a surface and a liquid phase, and embedded in a self-produced organic polymeric matrix [9] whose behavior is driven by cell-cell, cell-extracellular polymer, and cell-surface interactions [45]. As a consequence of these dynamic interactions, biofilm size, matrix composition, architecture, and cellular physiology can change during biofilm development [46,47]. Thus, the discrepancy initially observed between the number of biofilm cells and biofilm weight and thickness may be an indirect effect of the cell-surface interactions. Nevertheless, both biofilm wet weight and thickness analysis also suggested that SilRef-coated surfaces had a better performance in preventing Cyanobium sp. and P. tunicata biofilm formation than EpoRef surfaces. Moreover, CLSM analysis revealed that the biovolume and surface coverage of 49-old-biofilms grown on SilRef surfaces were lower than on EpoRef surfaces. These results can be explained by the foul release properties of this coating that prevent fouling settlements and provide extremely smooth and self-cleaning surfaces [8,10,11,15,48].
The properties of the fouling surfaces, together with the features of the fouling microorganisms, are the main factors that govern the fouling impact and strength of the biofouling interface [3]. The results demonstrated that the efficacy of SilRef-coated surfaces may be compromised by the microorganism.
Indeed, for Cyanobium sp. biofilms, the increase observed in all biofilm parameters was mainly due to the microorganism features. Accordingly, SilRef surfaces displayed higher activity against P. tunicata than Cyanobium sp. biofilms. For Cyanobium sp. biofilms formed on EpoRef surfaces, a combined effect between the surface and microorganism features played a considerable role in biofilm development at the maturation stage. The microfouler features also determined, to a lower extent, the biofilm formation. In addition, the properties of EpoRef surfaces contributed to a higher biofilm weight at the initial biofilm development stage. Lastly, the surface properties exerted a significant effect on P. tunicata biofilm development, contributing to a higher number of cells, wet weight, and thickness, especially in the biofilm maturation stage.

Conclusions
This work suggests that the dynamic behavior of biofilms developed on the different marine coatings is affected by the properties of the coatings, the microfoulers' features, and the interactions that are established between them in the different stages of biofilm development. Furthermore, it was shown that the SilRef marine coating is an effective, non-toxic commercial alternative to toxic biocide antifouling paints to control biofilm formation of common microfoulers.