The search for new antifouling (AF) coatings that are environmentally benign has led to renewed interest in the ways that micro-organisms colonize substrates in the marine environment. This review covers recently published research on the global species composition and dynamics of marine biofilms, consisting mainly of bacteria and diatoms found on man-made surfaces including AF coatings. Marine biofilms directly interact with larger organisms (macrofoulers) during colonization processes; hence, recent literature on understanding the basis of the biofilm/macrofouling interactions is essential and will also be reviewed here. Overall, differences have been identified in species composition between biofilm and planktonic forms for both diatoms and bacteria at various exposure sites. In most studies, the underlying biofilm was found to induce larval and spore settlement of macrofoulers; however, issues such as reproducibility, differences in exposure sites and biofilm composition (natural multispecies vs. monospecific species) may influence the outcomes.
Marine biofilms, often termed microfouling, are organized communities of mixed micro-organisms, typically surrounded by a matrix of extrapolymeric substances (EPS) (reviewed by Dobretsov and Thomason, 2011). Along with diatoms, bacteria constitute the major components of biofilms occurring in the marine environment, where the prevalence of both is dependent on geographical and seasonal variations. The moment a clean surface is submerged in the sea, biofilm-forming micro-organisms will rapidly colonize it and subsequently form highly complex, dynamic three-dimensional (3D) surface structures (Davey and O'Toole, 2000; Huggett et al., 2009; Molino et al., 2009a). Bacteria in biofilms are well known to exhibit enhanced resistance against antibiotics and many other types of stress compared with their planktonic forms (e.g. Costerton et al., 1999; Hall-Stoodley et al., 2004). A change in the bacterial phenotype (i.e. from planktonic to biofilm and vice versa) has been suggested to occur in a cell density-dependent manner, commonly known as quorum sensing (QS) (e.g. reviewed by Waters and Bassler, 2005). QS regulates a range of different processes, serving as a simple communication network via the secretion, accumulation and recognition of low-molecular mass signalling compounds, leading to the expression of phenotypes for improved access to nutrients, more rigid colonization and greater resistance of the community to hostile environments (Reading and Sperandio, 2005; Waters and Bassler, 2005). Diatoms (Bacillariophyceae and Ochrophyta) are a diverse group of eukaryotic algae, usually dominating the phytoplankton community of nutrient-rich marine waters. Diatoms are major primary producers and play an important role in marine biogeochemical cycles. In addition to their planktonic form, diatoms may occupy a wide variety of habitats such as sands in coastal beaches, the interstices on the bottom side of ice, as well as living surfaces and can live heterotrophically in deep-sea sediments. The attachment of diatoms on surfaces has made them important fouling organisms that adhere on man-made substrates and that, along with bacteria, constitute a major problem on artificial structures immersed in the marine environment.
Man-made structures as well as natural surfaces (both inanimate and living) are affected by biofilm attachment and growth. Marine substrates including aquaculture nets, oil and gas installations, as well as ship hulls, often demonstrate a great diversity of accumulated microfoulers and macrofoulers (e.g. algae, barnacles and mussels). In particular, on a single ship's hull, there exists a wide range of materials and structures providing a variety of fouling niches favourable to diverse organisms (see Fig. 1). The economic impact of hull biofouling has been examined by Schultz and colleagues (2011) who showed that the primary cost associated with fouling was due to increased fuel consumption linked to frictional drag (estimated to be $56 million per year for the entire class of DDG-51 naval ships). Several papers have described the effect of biofilms on the hydrodynamic performance of ship hull surfaces (reviewed by Howell, 2009). For instance, Schultz, Swain and colleagues (Schultz and Swain 1999, 2000; Schultz, 2007; Schultz et al., 2011) observed penalties in local skin friction coefficients of between 33% and 187% on flat plates fouled with biofilms. This describes how, despite the fact that macrofouling organisms are primarily responsible for degrading ship hydrodynamic performance, microfouling organisms also play an important role, particularly in the colonization processes.
Recent reviews into fouling control using textured and biomimetic surfaces have identified the numerous challenges of developing antifouling (AF) technologies (Salta et al., 2010; Scardino and de Nys, 2011). Importantly, it was highlighted that these technologies usually target specific taxonomic groups, for instance coatings that inhibit macrofoulers may not be active against biofilms. While Scardino and de Nys primarily focused on the effect of surface modifications on macrofoulers, they noted that microbial biofilms can modify and/or mask the surface topographies/properties, and they went on to state the need to control these complex biofilm communities.
Overall, marine biofilms are now recognized to be a significant issue for a wide range of engineered structures; however, there is very limited published information about marine biofilm composition and how it varies with substrates and AF coatings. Marine biofilms often form the initial surface accumulation, which culminates with macrofouling colonization. Thus, understanding the development of microfouling communities is an important aspect in the selection and management of AF coatings for marine operations. This paper aims to review marine biofilm composition on man-made materials with an additional focus on the positive or negative effects that biofilms may exert on macrofoulers.
Impact of local environment on biofilms
Comparative studies on early-stage biofilm composition dynamics is challenging because different immersion sites are associated with differences in environmental conditions (such as nutrient status, temperature, hydrodynamics and water chemistry), all of which influence both planktonic and biofilm ecophysiology (e.g. Hall-Stoodley et al., 2004). Although few studies have reported the environmental details of the immersion sites, it should be underlined that biofilm community composition is directly driven by its ecosystem, i.e. biogeochemical and physical interactions, seasonal and geographical variations. Biofilm dynamics should then preferably be evaluated in this context. For instance, temperature (Lau et al., 2005), as well as nutrients (Chiu et al., 2008; Briand et al., 2012), photosynthetically available radiation, pH and dissolved oxygen (Nayar et al., 2005) all have been shown to affect the structure of biofilm microbial communities whereas salinity appeared to be less influential (Lau et al., 2005). Diatom community structure in the biofilms illustrated significant differences between seasons that were attributed to combined physicochemical and biological changes in the water column (Patil and Anil, 2005).
In an attempt to link changes in biofilm community composition with anthropogenic contamination in Antarctica, glass slides were immersed at three contaminated sites (scaling from low to highly impacted) (Webster and Negri, 2006). Denaturing gradient gel electrophoresis (DGGE) analysis revealed that the bacterial sequences retrieved from the glass surfaces were affiliated to γ-Proteobacteria, Cytophaga/Flavobacteria of Bacteroidetes (CFB), Verrucomicrobia and Planctomycetales. Biofilm community composition was found to be more diverse between the different sites than within them, with α-Proteobacteria, γ-Proteobacteria and CFB dominating the least impacted site. Interestingly, sulphate-reducing bacteria were found in higher proportions at the more contaminated Antarctic sites. Highly diversified deep-sea biofilm bacterial communities were described based on terminal restriction fragment length polymorphism (T-RFLP) in the Eastern Mediterranean (Ionian Sea). Following 155 days of submersion down to 4500 m depth, it was found that biofilm diversity increased with depth while communities appeared to be less affected by substrate type (Bellou et al., 2012). Interestingly, recent work by Burke and colleagues (Burke et al., 2011a,b) has demonstrated that epiphytic (on Ulva sp.) biofilm community structures appear to share common functional genes rather than taxa. External shear force will not only change biofilm architecture and structure (Valiei et al., 2012; Kumar et al., 2013; M. Salta, L. Capretto, D. Carugo, J.A. Wharton and K.R. Stokes, submitted), but it can also influence biofilm detachment (Horn et al., 2003), which in turn can influence biofilm formation and the microbial ecology within it. Biofilms grown at low velocities (0.15 knots) exhibit low density and high effective diffusivity, whereas biofilms grown at high flow velocities (0.54 knots) have high density and low effective diffusion (Horn et al., 2003). Tsai (2005) investigated the impact of flow velocity on the dynamic behaviour of biofilm bacteria and found that maximum biofilm biomass did not change when flow velocity was increased from 0.38 to 0.77 knots, but was significantly affected when flow velocity was further increased to 1.16 knots. Evidence is emerging that multispecies communities that develop under high shear stress are less diverse than those that have developed at lower shear stress (reviewed by Howell, 2009). Rochex and colleagues (2008) assessed the effect of shear stress (0.055–0.27 Pa) on biofilm diversity using the Polymerase Chain Reaction (PCR)-single-strand conformation polymorphism fingerprinting method and concluded that shear stress affected biofilm composition. They showed that biofilm species composition decreased with increasing shear stresses, suggesting reduction of biofilm maturation and maintenance of biofilms in early developmental stage. Biofilm morphology and colony structure appears to be greatly affected with increasing shear stresses (Salta et al., submitted) while microstructures within flow channels that produce secondary flows resulted in pronounced changes in biofilm dynamics (Kumar et al., 2013).
Marine biofilm diversity
Biofilm organisms are mainly represented by sessile bacteria, microalgae including diatoms, microscopic fungi, heterotrophic flagellates and sessile ciliates (heterotrophic protists). The abundance ratio between these organisms varies, with bacteria and diatoms often being the dominant taxa. In the White Sea (Russia), microfouling communities that developed on polymer plates had a bacteria/diatoms/heterotrophic flagellates cell (107 cells ml−1) ratio of 640:4:1 (Railkin, 2004) while the proportion of other unicellular organisms (yeast, autotrophic flagellates and ciliates) was only about 0.15% of the total number of cells. Similar proportions of the main microfouler groups were reported for the US Pacific Coast communities (Railkin, 2004).
Biofilm communities on artificial surfaces without AF properties
It has been reported that bacterial communities on two dissimilar surfaces, stainless steel and polycarbonate (immersed in Delaware Bay, USA), would evolve to a similar pattern over time (Jones et al., 2007b). Likewise, biofilm community composition, as determined by a combined approach of DGGE and fluorescence in situ hybridization (FISH), was reported as similar across all surfaces, regardless of initial substrate wettability (10 chemical treatments of glass slides, immersed at Hawaii, USA), and all these surfaces had distinct temporal shifts in community structure over a 10-day period (Huggett et al., 2009). Similar results have been obtained using polystyrene and glass (Hung et al., 2008) or granite rocks (Chung et al., 2010) immersed at the same site in Hong Kong waters (China). It is therefore reasonably clear that bacteria have evolved mechanisms that allow them to adapt and colonize surfaces that possess a range of physico-chemical properties (Hung et al., 2008; Chung et al., 2010; Wahl et al., 2012).
The early-stage biofilms were dominated by the same major classes of bacteria that were most abundant in planktonic communities, with the latter demonstrating a higher diversity when compared with that of biofilm bacteria (Jones et al., 2007b; Lee et al., 2008). The Rhodobacterales (α-Proteobacteria), especially the Roseobacter clade members, are the dominant and ubiquitous primary surface colonizers in temperate coastal waters (Pacific and Atlantic coast) (Dang et al., 2008), with their success being possibly attributed to their ability to react to a low level of nutrient enrichment faster than other bacteria. Bacteria in the Alteromonas (γ-Proteobacteria) and Roseobacter (α-Proteobacteria) groups were identified as the main primary colonizers (9–72h). Acidobacteria, Actinobacteria, Bacteroidetes, Chloroflexi, Cyanobacteria, Firmicutes, Planctomycetes, β-,δ- and ε-Proteobacteria and Verrucomicrobia were identified as minor groups also belonging to these biofilms dependent on the nature of the surfaces (glass, polycarbonate, polystyrene and steel) and immersion sites (Great Barrier Reef, Australia, Delaware bay, South Carolina and Hawaii, USA, Sacheon harbour, Korea and Hong Kong, China) (Dang and Lovell, 2000; 2002; Webster et al., 2004; Jones et al., 2007a; Lee et al., 2008; Huggett et al., 2009; Chung et al., 2010). During the early developmental stage, between 24 h to 6 days, Archaea have not been detected on glass and polystyrene surfaces (Dang and Lovell, 2002; Chung et al., 2010) while in other studies they have been found only at low densities on glass (Webster et al., 2004; Huggett et al., 2009). Pooled replicates from 6- and 12-day-old biofilms developed on polystyrene Petri dishes were analysed using PhyloChip (DNA Microarray for Rapid Profiling of Microbial Populations) (Chung et al., 2010). This analysis showed a much higher species richness than previously described: 100/123 taxa in the 6- and 12-day-old biofilm were affiliated to 10/9 known phyla respectively (see Fig. 2). The global predominance of Proteobacteria (especially α-Proteobacteria) was confirmed.
Further studies should reveal whether these changes can be associated with the variation of environmental conditions through the seasons or whether they only reflect the capacity of microbial populations to maintain their functions in an ecosystem beyond phylogenetic diversity. In addition, when looking at the genus or species level within the same phylum, the dominance would be expected to change over time. It is of interest that human pathogens like Escherichia coli or Vibrio cholerae were also identified in marine biofilms on submerged surfaces including ship's hulls, which raises the question of their role in pathogen dissemination (Inbakandan et al., 2010; Shikuma and Hadfield, 2010).
It has been shown that, as with bacterial biofilms, the planktonic diatom community structure differs from the biofilm community, with pennate diatoms dominating the biofilms and centric diatoms the water column (Patil and Anil, 2005 and included citations). Frequently identified fouling diatoms include the pennate genera Navicula, Nitzschia, Cocconeis, Licmophora, Synedra, Amphora, Achnanthes, Bacillaria, Biddulphia and the centric genera Melosira, Fragilaria, Grammatophora, Rhabdonema, Berkeleya (Railkin, 2004). Settlement of marine periphytic algae on glass substrata in a tropical estuary in Singapore showed that the periphytic algal community comprised 30 microalgal species, dominated by diatoms (78%), followed by Cyanobacteria (19%), green flagellates (1%), dinoflagellates (1%), with other taxa accounting for the remaining 1 % of the total cell counts. Diatoms such as Skeletonema costatum and Thalassiosira rotula dominated the assemblages, together with the marine Cyanobacteria Synechococcus sp. (Nayar et al., 2005). Dominant biofilm diatoms found on fibreglass and glass coupons in a monsoon-influenced tropical estuary were pennate species belonging to the genera Navicula, Amphora, Nitzschia, Pleurosigma and Thalassionema (Patil and Anil, 2005). The biofilms formed on these two substrata revealed significant differences in density and diversity; however, the species composition was almost constant. An aquaculture study in Malaysia (Khatoon et al., 2007) showed that diatoms consisting of Amphora, Navicula and Cymbella were found to be most abundant on all the test substrates (bamboo pipe, plastic sheet, polyvinylchloride pipe, fibrous scrubber and ceramic tile) despite differences in the overall settlement densities. In addition, Oscillatoria (Cyanobacteria) were the next most important group while green algae were the least abundant. Nitzschia, Cylindrotheca, Navicula and Amphora have also been identified in 10-day-old biofilms developed on polystyrene Petri dishes immersed at the low intertidal zone in Port Shelter, Hong Kong, China (Chiu et al., 2008).
Biofilm communities on artificial surfaces with AF properties
It is now reasonably well established that different AF surfaces behave in contrasting ways with regard to the attachment and growth of marine fouling. The main focus of current research into AF surfaces or coatings that hinder fouling is associated with either self-polishing coatings (SPCs) with booster biocides, or silicone fouling release coatings (FRCs) with hydrophobic properties that operate by physicochemical and mechanical effects (Lejars et al., 2012). Although FRCs inhibit most macrofouling, they fail to prevent colonization of biofilms (Cassé and Swain, 2006; Molino and Wetherbee, 2008). There are a number of reports in the literature showing that diatom-dominated biofilms adhere tenaciously to hydrophobic surfaces and do not release from FRCs even on vessels operating at high speed (> 30 knots) (Anderson et al., 2003) and have been shown to significantly impact the performance of FRCs (Molino et al., 2009a,b).
Static and hydrodynamic seawater immersion testing of four commercial AF coatings, of which three were biocide based [tributyltin (TBT), self-polishing, copper self-polishing and copper ablative] and one was a biocide free-FRC, were performed in Florida (Cassé and Swain, 2006). Following static immersions for 60 days, both Gram-positive (Actinobacteria and Firmicutes) and Gram-negative (α- and γ-Proteobacteria) bacteria were found on all coated surfaces. Following static immersions for 60 days, both Gram-positive (Actinobacteria and Firmicutes) and Gram-negative (α- and γ-Proteobacteria) bacteria were found on all coated surfaces. These findings were based on culture-dependent methods that may not allow relevant conclusions on bacterial taxa dominance and diversity as it is estimated that less than 1 % of the bacterial species, in any ecosystem, are culturable (Amann et al., 1995). The main two genera reported during the static trials were Micrococcus and Pseudomonas, while bacterial total cell counts were found to be similar on all test surfaces. Following hydrodynamic immersion testing of the coatings, the biofilm diversity decreased and the dominant cultivable bacterium reported was Micrococcus (Cassé and Swain, 2006). Bacterial counts on all surfaces showed a substantial decrease after the dynamic testing, and although the greatest cell count decrease was observed for the FRC, bacterial diversity was highest on that coating.
When the pioneering colonization of bacterial biofilms on the FRC was investigated, the coating displayed the quickest colonization by bacteria, resulting in major modification of these coated surfaces within 2–4 days following immersion in the ocean (Molino et al., 2009b). It should be noted that Molino and colleagues (2009b) identified their sampling technique (i.e. scraping the biofilm off the coated surface) to be limiting for taxonomic analysis, especially for surfaces with low diatom colonization.
Polystyrene, Polytetrafluoroethylene and four AF paints were immersed for 2 weeks at two contrasting sites near Toulon on the French Mediterranean coast. All pioneer bacterial communities (analysed by PCR-DGGE) showed related structure (more than 25% of common Operational Taxonomic Units, OTU), controlled both by the sites and the type of substrata (Briand et al., 2012).
A study has been performed on the bacterial composition of concretions on the USS Arizona, a national naval memorial located in Pearl Harbor (Hawaii, USA), where the microbial community may be at a stable late stage. 16S rDNA clones indicated that the biofilms consisted of bacteria related to only three phyla: Firmicutes, Bacteroidetes and Proteobacteria (α and γ). Among the α-Proteobacteria, two of the three clones clustered with sequences from the genus Roseobacter. These data could be coherent with dominance on non AF surfaces but with a lower diversity that may result from residual AF paint on the hull (either copper or mercuric oxide based paints) (McNamara et al., 2009).
The majority of the studies have reported static immersions with the dominant diatom populations found on all coatings (SPC and FRC) during static trials being mainly Amphora (Cassé and Swain, 2006; Pelletier et al., 2009; Zargiel et al., 2011) in addition to Cylindrotheca, Licmophora and Nitzschia (Molino et al., 2009a; Dobretsov and Thomason, 2011; Briand et al., 2012). Short-term immersions (Molino et al., 2009a; Dobretsov and Thomason, 2011; and Briand et al., 2012) lead to low diatom densities that limited the conclusions in terms of dominance and diversity. In addition, Briand and colleagues (2012) reported that short immersion time corresponded to the initial phase of pioneer colonization, with a relatively high level of similarity between communities and two dominant species, Cylindrotheca closterium and Licmophora gracilis. Eight commercial marine ship hull coatings were exposed at three immersion sites during 60 days along the east coast of Florida. Among the 127 species comprising 44 genera identified, copper coatings were primarily fouled by Amphora delicatissima and Entomoneis pseudoduplex whereas copper-free coatings were colonized by Cyclophora tenuis, A. delicatissima, Achnanthes manifera and Amphora bigibba (Zargiel et al., 2011). Amphora also appears to be resistant to the organic biocides Copper Omadine™ (AkzoNobel, Amsterdam, the Netherlands) and Sea Nine™211 (The DOW Chemical Company, Midland, Michigan, USA) (Pelletier et al., 2009) or zinc pyrithione (Zargiel et al., 2011). In addition, FRC exhibited higher average diatom abundance and were typified by C. tenuis and several species of Amphora that was similar in diversity to the copper and copper-free coatings (Zargiel et al., 2011). A second comparison between SPC and FRC during 16 days conducted over three seasons in both temperate and tropical marine environments in Australia showed that diatom biofilms colonized the FRC (Intersleek 700®, AkzoNobel, Amsterdam, the Netherlands) more rapidly and exhibited a higher diversity when compared with biocidal SPC paints (including oxide and zinc pyrithione or TBT) (Molino et al., 2009a, Table 1). Considering only FRCs (Intersleek 900® and Intersleek 700®), Cylindrotheca was the dominant genus found on both coatings that were immersed in a marina in Oman, while overall the amphiphilic Intersleek 900® performed better, displaying the lowest biofilm colonization when compared with the hydrophobic Intersleek 700® and the control (tie coat) (Dobretsov and Thomason, 2011).
Table 1. Biofilm formation on AF coatings constructed using data from Molino et al., 2009b (it should be noted that where no diatoms were observed this may be as a consequence of the sampling technique).
For the only article that reported dynamic exposure, Amphora was the dominant genus that remained attached to most surfaces, with the exception of the copper ablative coating (Cassé and Swain, 2006). In contrary to static exposure, for FRC, diversity appeared to be significantly reduced. Achnanthes was present on both the TBT self-polishing and copper ablative surfaces after static immersion but remained only on TBT self-polishing after hydrodynamic immersion. This probably reflects the resistance of Achnanthes to TBT, but not to copper (Callow, 1986).
Considering diatom communities found on in-service ships, a study on 20 vessels coated with SPC paints containing TBT and copper showed that Amphora predominately colonized all ocean-going vessels, followed by Achnanthes, Amphiprora, Navicula and Stauroneis that were also observed in some cases (Callow, 1986).
Methodological approaches for biofilm community studies
Microscopy is typically used to assess marine biofilms, such as epifluorescence or scanning electron microscopy (SEM), while confocal laser scanning microscopy is rarely used despite its versatility (Shikuma and Hadfield, 2010; Faÿ et al., 2011). A typical example of marine biofilm formed in a temperate estuary (Southampton, UK) over 3 weeks during the summer period was imaged using an optical focus variation microscope (Alicona G4 Infinite Focus System, Alicona Imaging GmbH, Raaba, Austria) (Fig. 3). The principles of focus variation have been described by Helmli (2011). This microscope employs an optical 3D micro coordinate system and has long working distance objectives, which enables finely detailed microscopic imaging even of fully hydrated biofilmed surfaces. We believe that this is the first time a 3D, true colour image (Fig. 3a) of a marine biofilm in situ has been presented. Quantitative information can be acquired on the biofilm surface topography, roughness and structure (Fig. 3b) as well as biofilm thickness (Fig. 3c). This rapid and non-invasive (no fixation or staining required) optical imaging technique is ideal for the assessment of marine biofilms on artificial surfaces where samples can be processed within minutes after retrieval and examined while still wet.
In biofilm work, the use of flow cytometry is rarely reported (e.g. only in two studies Mitbavkar et al., 2012; Toupoint et al., 2012), although this is a frequently used method in oceanography (e.g. Collier, 2000). The main problem with using flow cytometry for biofilm work is the observed difficulties in separating the bacterial cells from the EPS.
Diversity studies have mainly been based on culture-independent techniques. Most of these studies use molecular fingerprinting approaches (e.g. DGGE and T-RFLP) or FISH, and consequently comparisons between the structures of biofilm communities is possible. However, species level of identification is not reached, and relative abundance cannot be generally assessed. A few cloning-sequencing approaches were performed using only 16S rDNA as a target (e.g. Jones et al., 2007a; Dang et al., 2008). To date, only one metagenomic study has been published. Such an approach based on metagenome DNA sequencing or PCR-amplicon sequencing will significantly improve the level of taxonomic but also functional diversity identification, especially for rare taxa that appear to be increasingly more informative (if not more so) than major ones. Next-generation pyrosequencing has been applied to analyse the microbial metagenome of the marine biofilm community that colonizes and degrades insoluble polysaccharides in situ in the Irish Sea. The identification of 211 gene sequences revealed functional genes predicted to be involved in cellulose utilization (Edwards et al., 2010). In addition, a proteomic approach has been recently developed for biofilms on ship hulls that could initiate functional studies (Leary et al., 2012).
Molecular techniques are less used to study eukaryotic micro-organisms from biofilms, especially the autotrophic communities of diatoms (Webster and Negri, 2006), which are mainly identified to the genus level using optical microscopy or SEM. Lack of a universal molecular target such as 16S for prokaryotes limits such approaches. For example, 18S cloning-sequencing was performed to identify eukaryote members of biofilms in order to study their relationship with mussels. Surprisingly, 18S cloning-sequencing failed to detect diatoms and dinoflagellates despite their identification through SEM (Toupoint et al., 2012). Thus, metagenomic approaches should be developed in order to better describe and understand the entire process of biofilm development in the marine environment (Burke et al., 2011a,b).
Finally, metabolomic approaches that will allow not only the whole diversity, but also the identity and sometimes the quantity of metabolites to be taken into account would be of great interest to characterize biofilms. First, investigations were performed using gas chromatography mass spectroscopy (Chung et al., 2010), but tools like desorption electrospray ionization mass spectroscopy (Espy et al., 2011) would be much more powerful for mapping artificial surfaces.
Biofilms and macrofouling: a love/hate affair
Marine organisms themselves suffer pressures from the settlement of foulers – a phenomenon often referred to as epibiosis (from the Greek epi = on top and bio = life) (see Fig. 4). Fouling by secondary organisms can be detrimental to the settled organism, restricting access to light and nutrition, and increasing the probability of tissue damage and disease (Bryan et al., 1996). Therefore, marine organisms have evolved defences against epibiosis. Substantial evidence suggests that chemical cues from marine organisms affect the colonization processes of secondary colonizers reviewed by Harder and Yee (2009). Marine colonization processes are broadly described as sequential steps with the formation of microbial biofilms preceding the attachment of larger organisms. Biofilms are therefore considered to be key mediators for the subsequent colonization by macro-organisms, with observed attractive (love) or repulsive (hate) influences. However, detailed information for any one system on the nature of such cues, their distribution in situ, and their effects on the demography of colonizers is rare (Steinberg et al., 2002). Understanding the role of microbial biofilms in chemical ecology would be substantially enhanced by greater knowledge of the bacterial species composition in the environment and their relative abundance (Steinberg et al., 2002). The laboratory and field methodologies used to assess the effects of biofilms on macrofoulers can be found in an extensive review by Wieczorek and Todd (1998). Table 2 summarizes the recently published literature on the effect of biofilms on settlement by macrofoulers and/or larval metamorphosis, with emphasis on the experimental substrate used.
Table 2. Review of selected papers on effect of marine biofilms on macrofouler's settlement and metamorphosis
+/− signifies the negative or positive effect of the biofilm on macrofouler's settlement and metamorphosis.
Biofilm composition and larval settlement are known to be affected by the physical properties of a substrate, such as colour (Satheesh and Wesley, 2010), surface roughness (Kerr and Cowling, 2003), wettability (Maki et al., 1992) as well as topography (Magin et al., 2010). However, only limited information is currently available on the interactions between the physical properties of a substrate, biofilm formation and subsequent larval settlement. It should be noted that the relationships between biofilms and macrofoulers have usually been studied by performing bioassays with a limited selection of algal or invertebrate species that are commonly cultured in the laboratory and are of particular commercial interest. A comprehensive recent review by Hadfield (2011) attempts to address the chemo-relationship between larvae and bacteria concluding that, despite 60 years of research, little is known about their interactions. Several studies report positive effects of marine biofilms on macrofoulers' settlement; however, to a lesser extent there is also evident to support the opposite, i.e. settling inhibition of larvae due to biofilms; the following sections will separately cover the recent findings on both positive and negative effects.
Biofilms and macrofoulers: positive effect
Overall, natural biofilms appear to significantly increase adhesion strength of macrofoulers in most of the studies: for example, the ascidian Phallusia nigra (reported from the Mediterranean to Micronesia) (Zardus et al., 2008), the polychaete tubeworm Hydroides elegans (e.g. Huang and Hadfield, 2003; Hung et al., 2005; Shikuma and Hadfield, 2006; Zardus et al., 2008; Huggett et al., 2009), the serpulid Pomatoceros lamarkii (Hamer et al., 2001), several mussels including Mytilus edulis (Bao et al., 2007; Campbell et al., 2011; Toupoint et al., 2012), the barnacle Amphibalanus (Balanus) at one or more developmental stages (e.g. Qian et al., 2003; Chiu et al., 2008; Hung et al., 2008; Zardus et al., 2008) and the algal genus Ulva (Dillon et al., 1989; Joint et al., 2000; Mieszkin et al., 2012).
Maki and Mitchell (1985) presented evidence that the larvae of another small tube-dwelling polychaete, Janua brasiliensis, are stimulated to settle and metamorphose in the presence of biofilm EPS, produced by species such as Pseudomonas marina. Hydroides elegans is an early colonizer of newly submerged surfaces in the succession of macrofouling invertebrates, with a global distribution found mainly in tropical and subtropical bays and harbours (Shikuma and Hadfield, 2006). The presence of a biofilm is a prerequisite for larval settlement of H. elegans (e.g. Lau et al., 2005), and it was found that larval metamorphosis of this species was not predominately affected by nutrient enrichment, but rather by the age of biofilms (Chiu et al., 2008). When clean surfaces were tested (i.e. no biofilm present), Balanus amphitrite cyprid attachment was greatly affected by surface wettability with preferential settlement on hydrophilic surfaces (glass). However, the presence of aged biofilms (6 days old) overrode the effect of surface wettability on cyprid attachment (Hung et al., 2008). For H. elegans, several studies report that biofilm age is the main attractant with older biofilms inducing higher larval recruitment (e.g. Shikuma and Hadfield, 2006; Huggett et al., 2009; Chung et al., 2010), which appears to be commonly observed for a number of macrofoulers, e.g. for ascidian larvae (Wieczorek and Todd, 1998), coral larvae (Webster et al., 2004), mussel larvae (Bao et al., 2007; Toupoint et al., 2012; Wang et al., 2012), temperate or tropical barnacle cyprids (Wieczorek et al., 1995; Thompson et al., 1998; Hung et al., 2008) as well as algal spores (Joint et al., 2000; Patel et al., 2003). Therefore, it would be more insightful to extend test surface submersion times for a sufficient duration to allow the biofilm and the macrofoulers to mature and reproduce (Shikuma and Hadfield, 2006; Huggett et al., 2009; Chung et al., 2010). Also, metamorphosis of H. elegans was positively correlated with increased bacterial densities (Huang and Hadfield, 2003).
One of the best defined QS system molecules are the acyl-homoserine lactones (AHL), which are the most common autoinducer class for Gram-negative bacteria (e.g. Waters and Bassler, 2005; Galloway et al., 2011). The positive correlation between Ulva zoospore settlement and biofilm AHLs has been reported (e.g. Wheeler et al., 2006; Tait et al., 2009). Interestingly, Wheeler and colleagues (2006) identified a chemoresponse through chemokinesis according to which, zoospores decreased their swimming speed and increased their settlement when close to the AHL sources. Chung and colleagues (2010) highlighted that although chemical cues are produced by several microbes in natural biofilms, the chemical profiles are likely to vary quantitatively and qualitatively over the period of biofilm development. A recent study by Huang and colleagues (2012) identified four genes (from the bacterium Pseudoalteromonas luteoviolacea HI1 and mutants lacking one of the genes) that are necessary for metamorphic induction for H. elegans larvae; these genes encode functions that may be related to cell adhesion and bacterial secretion systems.
Biofilms and macrofoulers: negative and neutral effect
Relatively few studies indicate a negative or neutral effect of biofilms on macrofouling settlement (see Table 2). In a laboratory-based study, Faimali and colleagues (2004) found that biofilms inhibited cyprid attachment in an age-dependent manner, i.e. older biofilms hindered settlement (Faimali et al., 2004). The widespread temperate and tropical bryozoan Bugula neritina appeared to be a relatively indiscriminate settler on natural or laboratory (including diatoms) biofilms (Dahms et al., 2004; Zardus et al., 2008). Chung and colleagues (2010) and to a lesser extent Lau and colleagues (2005) found no evidence that correlate settlement of H. elegans with bacterial densities. Similarly, the larval settlement of two Balanus sp. (B. amphitrite and Balanus trigonus) was unrelated to bacterial density or biomass (Lau et al., 2005). Semibalanus balanoides cyprids metamorphosis and settlement was enhanced by mature biofilms in field trials (UK); however, laboratory-based tests did not support these findings (Thompson et al., 1998).The settlement was reported to be predominantly influenced by the proximity of conspecifics and by traces of previous barnacle colonization.
In general, very few studies report data that deal with the interactions between biofilm-forming diatoms and macrofoulers. Among diatom strains isolated from natural biofilms, only 10% induced the larval settlement of H. elegans compared with 31% and 59% that showed either weak or no inducement respectively. The variability in larval settlement was not correlated with the density of diatoms in monospecific films (Harder et al., 2002). Contrary to marine bacteria, diatom-induced larval settlement cues are more likely to be composed of polymeric components, such as EPS and not secondary metabolites (Lam et al., 2003). Intraspecies and interspecies (or taxon) communication in diatoms is known to occur through pheromones (reviewed by Amin et al., 2012) that could potentially serve as inductive molecules for larval settlement; however, this remains to be investigated.
Overall, the presence of a biofilm has been shown to demonstrate contrasting effects on macrofouler settlement associated with both inhibitory and stimulating effects (see Table 2). These contradictory results may be due to differences in experimental procedures: (i) the low reproducibility inherent in experimental settlement assays for invertebrate larvae (Wieczorek and Todd, 1998); (ii) field vs. laboratory-based experiments, where laboratory conditions provide a controlled environment whereas field experiments are subjected to dynamic conditions such as seasonality (e.g. Lau et al., 2005); (iii) natural multispecies vs. monospecific biofilms, where again it has been shown that biofilm community composition is season and time dependent (e.g. Bao et al., 2007); therefore, it may induce different attachment responses to macrofoulers; and (iv) different macrofouling species used within the various studies that do not facilitate direct data comparisons.
Although several studies report that biofilm presence may increase the macrofoulers' attachment strength on substrates, the exact interactive mechanisms are yet to be discovered, e.g. whether larvae penetrate through the biofilm to adhere onto the surface or integrate their own adhesive with that of the biofilm in order to achieve a more persistent attachment (Zardus et al., 2008). A recent study by Mieszkin and colleagues (2012) revealed that naturally occurring sea water bacterial biofilms mainly increased attachment strength of algal spores, sporelings and diatoms, while surfaces biofilmed with a single species (Cobetia marina) reduced adhesion strength of the same organisms. They concluded that organisms' settlement (and adhesion) on artificial surfaces is largely affected by bacterial biofilms; however, deconvoluting the exact mechanisms is indeed a difficult task.
In this review, we have attempted to summarize the literature concerning marine biofilms, the influence of composition and dynamics from two points of view: (i) substrates with and without AF capabilities and (ii) their influence on macrofouling. Marine biofilm studies should be coupled with the surrounding environmental parameters as it has been clearly demonstrated that they can affect both biofilms and macrofoulers. In addition, biofilm studies on AF coatings, and especially FRC, should be mainly studied in dynamic conditions as hydrodynamics have an effect on biofilm formation.
Recent studies using novel molecular techniques have illustrated that functional genes may be necessary for biofilm processes, which illustrates the importance of including the latest technological developments (i.e. meta ‘omic’ approaches) as marine biofilms are dynamic communities with extensive interactions between different species. Thus, a range of approaches need to be combined in biofilm research for better understand these complex communities.
Most of the studies on biofilm-macrofouling interactions identify a positive correlation with biofilms inducing larval settlement and metamorphosis; the majority of the reviewed work here relates this positive effect on biofilm age. Although there have been some attempts to identify phylogenetic relationships between biofilm and macrofouling taxa, no clear evidence supports a strong correlation. Interestingly, recent work employing mutational analysis revealed that specific genes were necessary for larval metamorphosis (H. elegans). The role of diatoms on larval recruitment is fairly unexplored, which could potentially be as important as the bacterial one.