Mini-review: Inhibition of biofouling by marine microorganisms

Abstract

Any natural or artificial substratum exposed to seawater is quickly fouled by marine microorganisms and later by macrofouling species. Microfouling organisms on the surface of a substratum form heterogenic biofilms, which are composed of multiple species of heterotrophic bacteria, cyanobacteria, diatoms, protozoa and fungi. Biofilms on artificial structures create serious problems for industries worldwide, with effects including an increase in drag force and metal corrosion as well as a reduction in heat transfer efficiency. Additionally, microorganisms produce chemical compounds that may induce or inhibit settlement and growth of other fouling organisms. Since the last review by the first author on inhibition of biofouling by marine microbes in 2006, significant progress has been made in the field. Several antimicrobial, antialgal and antilarval compounds have been isolated from heterotrophic marine bacteria, cyanobacteria and fungi. Some of these compounds have multiple bioactivities. Microorganisms are able to disrupt biofilms by inhibition of bacterial signalling and production of enzymes that degrade bacterial signals and polymers. Epibiotic microorganisms associated with marine algae and invertebrates have a high antifouling (AF) potential, which can be used to solve biofouling problems in industry. However, more information about the production of AF compounds by marine microorganisms in situ and their mechanisms of action needs to be obtained. This review focuses on the AF activity of marine heterotrophic bacteria, cyanobacteria and fungi and covers publications from 2006 up to the end of 2012.

Keywords:

• marine microorganism
• biofilm
• inhibition
• larval settlement
• biofouling
• antifouling defence

Introduction

In non-sterile seawater, any natural and artificial substrata quickly become fouled. Within minutes, organic molecules and particles are adsorbed onto the surface of any substratum, which is later colonised by bacteria, diatoms and protozoa (Wahl 1989) and which form a biofilm. Biofilms are composed of multiple species of microorganisms attached to the substratum (and to each other) and encased within a matrix of extracellular polymeric substances (EPS). Mature biofilms have complex, three-dimensional structures, which depend on the species composition of the biofilm, microbial activity and environmental conditions (Lewandowski 2000). Marine biofilms contain different species of heterotrophic bacteria (mainly proteobacteria), cyanobacteria, archaea and unicellular eukaryotes (eg diatoms and flagellates), while the densities of sarcodines, ciliates and fungi remain low (reviewed by Dobretsov 2010; Wahl et al. 2012). The diversity of marine organisms from biofilms has mainly been described using culture methods and microscopic observations; the number of molecular-based studies of marine biofilms remains small.

Bacteria can coordinate their adhesion, biofilm maturation, swarming, luminescence and toxin production in a process called quorum sensing (QS) (Waters & Bassler 2005; Steinberg et al. 2011). During this process, a class of small molecules named ‘autoinducers’ is produced and released into the environment. When the concentrations of these signals reach a threshold level in diffusion-limited environments, the signals trigger an expression of target genes that consequently change the behaviour of bacteria (Irie & Parsek 2008; Steinberg et al. 2011). Gram-positive and Gram-negative bacteria use a number of different, potentially overlapping QS signalling systems, but the best characterised of these is based on the production and perception of N-acyl homoserine lactones (AHLs) (reviewed by Dobretsov et al. 2009). Approximately 30% of the bacterial species within subtidal biofilms (Huang et al. 2009) and those associated with corals (Golberg et al. 2011) and sponges (Taylor et al. 2004) are able to produce AHLs. Some bacteria and marine organisms, such as algae, are able to respond to the QS signals of other bacteria or to compromise them (reviewed by Dobretsov et al. 2009).

Within biofilms, microorganisms interact with each other, with new microbial colonizers, as well as with larvae of invertebrates and algal spores (reviewed by Dobretsov 2010). Development of a biofilm on a substratum changes the attractiveness of the substratum to invertebrate larvae and algal spores through physical modification of surfaces and production and release of chemical compounds (reviewed by Qian et al. 2007; Hadfield 2010; Harder et al. 2012; Mieszkin et al 2012).

While being of relatively small size, microbial biofilms are extremely important in industry, as they can increase drag force and reduce heat transfer efficiency (Yebra et al. 2004). Biofilms can lead to corrosion of metals by increasing the open circuit potential toward positive values by metal deposition at the cathode, oxygen removal and acid production (Videla & Herrera 2005; Landousli et al. 2011). While biocidal and fouling-release coatings are generally free from macrofouling, they are still subject to colonisation by biofilms (Cassé & Swain 2006; Molino et al. 2009; Dobretsov & Thomason 2011; Zargiel et al. 2011; Briand et al. 2012). Biofilms on antifouling (AF) coatings increase roughness and hence drag, which results in increased fuel consumption of vessels (Yebra et al. 2006; Schultz et al. 2011). Thus, eradication of biofilms and inhibition of their growth are major industrial concerns.

The recent ban on toxic AF biocides (eg triorganotins) has underlined the need for the development of ‘environmentally friendly’ AF strategies and stimulated an active search for non-toxic natural marine AF compounds (reviewed by Qian et al. 2010). Most of these compounds have been isolated from eukaryotic organisms (Qian et al. 2010), with the disadvantage that such compounds are in limited supply. In contrast, marine microorganisms are providing a number of benefits for industries, which include the possibility of having an unlimited supply of compounds through fermenter cultivation, and the possibility of genetic and chemical modification of the source organisms and compounds, respectively (Dobretsov et al. 2006). Indeed, chemicals produced by marine microbes have been shown to be effective inhibitors of biofouling and their production to be suitable for scaling up in industrial applications (Qian et al. 2012).

This review covers the literature (including recent patents) on the inhibition of biofouling by marine bacteria, cyanobacteria and fungi from 2006 up to the end of 2012. Over this period of time, several comprehensive reviews have been published on the properties of AF compounds (Qian et al. 2010; Fusetani 2011), larval settlement (Hadfield 2010; Thiyagarajan 2010), properties of epibiotic bacteria (Egan et al. 2008, 2012; Penesyan et al. 2010; Harder et al. 2012; Wahl et al. 2012) and chemical ecology (Paul & Ritson-Williams 2008; Paul et al. 2011). Interference with microbial QS as a mechanism to control marine biofilms has been the subject of another recent review (Dobretsov et al. 2009); therefore, this topic has been excluded from this article. This review is organised by the bioactivity of the microorganisms, focusing on antibacterial, antialgal and antilarval activities and on microbial disruption of biofilms. These considerations are followed by concluding remarks.

Antibacterial activity

Heterotrophic bacteria

Bacteria with antimicrobial properties have been isolated frequently from surfaces of marine invertebrates and algae (Table 1). It is beneficial for the host to harbour epibiotic bacteria with AF properties because they provide protection against fouling. Although microbial epibiotic communities associated with marine organisms have been detected in many studies (reviewed by Wahl et al. 2012), it is far from clear whether hosts select the bacteria with AF properties, or whether such activities are simply coincidental.

Table 1. Antimicrobial activity of marine heterotrophic bacteria.

Bacteria	Effective against	Source (host)	Location	Inhibitive compound	Reference
92 Isolates	Environmental and pathogenic bacteria	Nine species of red algae	India, Indian Ocean	NI	Kanagasabhapathy and Nagata (2008)
210 Isolates belonging to α-, β-, γ-proteobacteria, Firmicutes, Actinobacteria and Bacteriodetes	Pathogens	Laminaria saccharina	Germany, North Sea	NI	Wiese et al. (2009)
325 Isolates most belonged to α- and γ-proteobacteria	Pathogens	Algae Ulva australis and Delisea pulchra	Australia, Pacific Ocean	NI	Penesyan et al. (2009)
Pseudovibrio sp. D323	Environmental bacteria	D. pulchra	Australia, Pacific Ocean	Tropodithietic acid	Penesyan et al. (2011)
10 Bacterial isolates	Environmental bacteria	Ulva lactuca	Fiji, South Pacific Ocean	NI	Kumar et al. (2011)
Shewanella irciniae, Roseivirga spongocola, Gillisia myxillae, Fabibacter halotolarans, Microbulbifer mediterranus, Stenothermonacter spongiae, Winogradskyella poriferorum	Environmental bacteria	Lissodendoryx isodictyalis	Caribbean, Atlantic Ocean	NI	Dash et al. (2009)
75 Isolates	Environmental bacteria	Sponges Anoxycalyx joubini and Lissodendoryx nobilis	Antarctica	NI	Mangano et al. (2009)
52 Isolates	Pathogens	Haliclona simulans	Ireland, North Sea	NI	Kennedy et al. (2009)
Bacillus licheniformis	Pathogens	Halichondria sp.	India, Indian Ocean	Indole, 3-phenylpropionic acid, 4,4′-oxybis[3-phenylpropionic acid]	Devi et al. (2010)
105 Isolates	Environmental and pathogenic bacteria	14 Invertebrates: sea urchins, brittle stars, tunicates, oysters, seaweed	Australia, Pacific Ocean	NI	Wilson et al. (2011)
Pseudoalteromonas haloplanktis	Environmental and pathogenic bacteria	Scallop hatchery	Chili, Pacific Ocean	Isovaleric acid (3-methylbutanoic acid) and 2-methylbutyric acid (2-methylbutanoic acid)	Hayashiga-Soiza et al. (2008)
14 Roseobacter bacteria	Environmental bacteria	Pfiesteria piscicida and seawater	Denmark, North Sea	NI	Bruhn et al. (2007)
90 Actinomyces spp.	Pathogens	11 Sponge species	Egypt, Red Sea	NI	Abdelmohsen et al. (2010)
20 Isolates	Pathogens	9 Sponge species	Brazil, Atlantic Ocean	NI	Santos et al. (2010)
2562 Isolates	Pathogens	10 Sponge species	Germany, North Sea	NI	Muscholl-Silberhorn et al. (2008)
36 Bacteria	Pathogens	Antipathes dichotoma	China, South China Sea	NI	Zhang et al. (2012a)
340 Isolates	Pathogens	14 Bryozoan species	Baltic, Mediterranean Seas	NI	Heindl et al. (2010)
28 Isolates	Pathogens	Balanus amphitrite	India, Indian Ocean	NI	Jebasingh and Murugan (2011)
78 Isolates	Pathogens	9 Species of corals	Israel, Red Sea	Evidence of cell bound proteinaceous compounds	Shnit-Orland and Kushmaro (2009)
1138 Bacteria	Pathogens	5 Species of corals	Taiwan, South China Sea	NI	Chen et al. (2012)
519 Bacteria	Pathogens	Seawater and surface of organisms	Global waters	Tropodithietic acid	Gram et al. (2010)
101 Strains of Pseudoalteromonas	Pathogens	Seawater and surface of organisms	Global waters	NI	Vynne et al. (2011)
110 Bacteria	Environmental bacteria	Seawater	Denmark, North Sea	NI	Bernbom et al. (2011)
Shewanella oneidensis SCH0402	Environmental bacteria	Seawater	Korea, South Sea	Cis-9-oleic acid	Bhattarai et al. (2007)
Pseudoalteromonas sp. 3J6	Environmental bacteria and pathogens	Seawater	France, Morbihan Gulf	NI	Dheilly et al. (2010)
Bacillus mojavensis, B. firmus	Environmental bacteria	Turtle grass and limestone	Mexico, Atlantic Ocean	Surfacin A, mycosubtilin, bacillomycin D	Ortega-Morales et al. (2008)
Pseudomonas rhizosphaerae	Environmental bacteria	Deep sea sediments	Pacific Ocean	Cyclo(Tyr-Pro), cyclo(Tyr-Ile), cyclo(Phe-Pro), cyclo(Val-Pro)3-phenyl-2-propenoic acid, uracil acid	Qi, Xu, Gao et al. (2009)
Note: NI = compound has not been identified.

Since the discovery of the first antimicrobial compound from the symbiotic bacterium Pseudomonas tunicata associated with the tunicate Ciona intestinalis (James et al. 1996), the number of studies on the antibacterial activity of epibiotic marine bacteria has increased substantially (Table 1). Of the 210 bacterial isolates from the brown alga Laminaria saccharina, 103 inhibited the growth of at least 1 of the pathogens tested (Wiese et al. 2009). By contrast, out of 92 isolates from Japanese red algae, only ca one-third of the isolates inhibited the growth of environmental and pathogenic bacteria (Kanagasabhapathy & Nagata 2008). Similarly, a low percentage (∼12%) of microbes with antibacterial properties belonging to the α- and γ-proteobacteria were isolated from the green alga Ulva australis and the red alga Delisea pulchra, respectively (Penesyan et al. 2009). All these studies suggest that the proportion of antibacterial epibiotic bacteria is different for different species of algae and may vary with different geographical locations. The latter assumption is supported by the study of Heindl et al. (2010), which showed that the presence of epibiotic, cultivable bacteria with antibacterial activity was site-specific, but not species-specific.

Bacteria associated with sponges, corals, echinoderms and bryozoa have also been demonstrated to have antibacterial activity (Table 1). Four out of 7 bacteria isolated from sponges from the Caribbean inhibited the growth of bacteria isolated from Hong Kong-derived biofilms (Dash et al. 2009). Antagonistic interactions among epibiotic bacteria isolated from the sponges Anoxycalyx joubini and Lissodendoryx nobilis have been demonstrated (Mangano et al. 2009), and the microbial isolates associated with these sponges have been characterised. More than 50% of the isolates from the sponge Haliclona simulanas inhibited the growth of pathogenic bacteria and yeast (Kennedy et al. 2009). Screening of actinomycete isolates for the presence of genes encoding polyketide synthase and non-ribosomal peptide synthase (typically involved in the synthesis of antibiotics), revealed that these genes were common in active isolates. In contrast, another study with a high number of bacterial isolates (2562) from 10 Baltic sponge species showed that only a few of them had inhibitory properties (Muscholl-Silberhorn et al. 2008). One of the bioactive isolates related to Pseudovibrio denitrificans was present at high densities in all of the sponges investigated, and hence it was speculated that this strain might be responsible for AF protection of these sponge species. Four out of 28 isolates from the shells of the barnacle Balanus amphitrite inhibited growth of 10 human pathogens (Jebasingh & Murugan 2011). Seventy percent of the bacterial strains isolated from massive and solitary corals had antibacterial activity, while only a few isolates from branching and soft corals had such activity (Shnit-Orland & Kushmaro 2009).

Some bacterial strains isolated from seawater inhibited the growth of marine and human pathogens (Table 1; Gram et al. 2010; Bernbom et al. 2011; Vynne et al. 2011). The highest antibacterial activity was observed among Pseudoalteromonas and Ruegeria isolates. The attachment of Pseudoalteromonas S91 was prevented by biofilms of Pseudoalteromonas piscicida, P. tunicata and P. ulvae (Bernbom et al. 2011). Liquid culture of Pseudoalteromonas sp. 3J6 isolated from French waters inhibited the formation of a biofilm and reduced the number of viable cells of Paracoccus sp. 4M6, Vibrio sp. D01 and the pathogens Pseudomonas aeruginosa, Salmonella enterica and Escherichia coli (Dheilly et al. 2010).

Only a small number of natural products have been isolated from epibiotic bacteria (Table 1). Pseudovibrio sp. D323, isolated from the red alga D. pulchra, produced tropodithietic acid, which inhibited the growth of other marine bacteria (Penesyan et al. 2011). The authors speculated that tropodithietic acid prevents growth of competitors and inhibits growth of pathogens on the surface of the alga. This hypothesis was supported by the observation of Webster et al. (2008), who noticed that the loss of species producing tropodithietic acid coincided with heavy microbial colonisation of the sponge Rhopaloeides odorabile. Interestingly, tropodithietic acid was identified as an antimicrobial compound and was isolated from Ruegeria strains obtained from all ocean areas except the Arctic and the Antarctic waters during the Danish marine research expedition Galathea 3 (Gram et al. 2010). Three antibacterial compounds were isolated from methanol extracts of the bacterium Bacillus licheniformis associated with the sponge Halichondria sp. (Table 1; Devi et al. 2010). One of these compounds, indole, had significant activity against bacterial pathogens, while others, 3-phenylpropionic acid and 4,4′-oxybis[3-phenylpropionic acid], had moderate and weak antibacterial activity, respectively. Lipopeptides (sufracin A, mycosubtilin and bacillomycin D) produced by Bacillus mojavensis and B. firmus associated with biotic or abiotic surfaces inhibited the growth of the bacterium Halomonas marina, and were toxic to the crustacean Artemia salina (Ortega-Morales et al. 2008). Several diketopiperazines (DKPs), uracil and 3-phenyl-2-propenoic acid produced by the deep sea bacterium Pseudomonas rhizosphaerae inhibited the growth of fouling bacteria (Qi, Xu, Gao et al. 2009). Isnansetyo and Kamei (2009) reviewed several of the antibacterial substances from marine Pseudomonas species, including pyroles, quinoline and quinolone.

The data presented in Table 1 suggest that the majority of bacterial strains isolated from marine organisms demonstrate some antimicrobial activity. The production of antibacterial compounds by bacteria is not constitutive and is not attributed to a specific taxonomic group of bacteria or specific geographical location. The specific culture conditions and currently unknown host cues affect the production of antimicrobial compounds by marine isolates cultured under laboratory conditions (Bruhn et al. 2007). For example, a strain of B. licheniformis lost its bioactivity when it was cultured in a liquid medium instead of on agar surfaces (Yan & Boyd 2002; Matz et al. 2008). Similarly, Pseudoalteromonas isolates attached to biotic or abiotic surfaces are more likely to have antibacterial properties than planktonic isolates (Vynne et al. 2011). Mucus of the coral Acropora palmata inhibited antibiotic and pigment production by the ‘visitor’ isolate Pseudoalteromonas sp. (Ritchie 2006). This provides evidence that marine hosts have the ability to manipulate the secondary metabolism and colonisation behaviours in the microorganisms they encounter.

Cyanobacteria

Marine cyanobacteria (blue-green algae) are a diverse group of unicellular microorganisms that exhibit a range of morphologies and have the ability to perform a range of metabolic processes such as oxygenic and anoxygenic photosynthesis, fermentation and nitrogen fixation (Stahl 1995). Both planktonic and benthic species are known, and the latter are actively involved in the formation of biofilms and microbial mats. Cyanobacteria produce chemicals in order to facilitate aggregation to form biofilms and to compete with other organisms for space and nutrients. Therefore, marine cyanobacteria have been recognised as a rich source of secondary metabolites with AF properties (Burja et al. 2001; Pulz & Gross 2004; Abed et al. 2009; Dobretsov et al. 2010; Kwan et al. 2011). Although there are more than 300 nitrogen-containing secondary metabolites with different biological activities reported from marine cyanobacteria, only a few compounds are known for their AF properties (Tan & Goh 2009). Most of these compounds, which are produced by 15 different cyanobacterial genera, have been chemically identified as lipopeptides, polyketides, amides, alkaloids, indoles and fatty acids (reviewed by Dahms et al. 2006).

The marine cyanobacterium Lyngbya majuscula has been shown to be a rich source of bioactive compounds including antibacterials (Table 2). Several cyclic depsipepetides (eg pitipeptolides A–F) were purified from this cyanobacterium and were shown to inhibit the growth of Mycobacterium tuberculosis besides having other cytotoxic activities (Montaser et al. 2011). In another study, extracts from the same cyanobacterium were shown to inhibit the growth of E. coli, Bacillus subtilis, B. cereus, Staphylococcus epidermis and Enterococcus faecalis (Kaushik et al. 2009). A methanolic extract of L. majuscula had significant activity against B. subtilis with a MIC value of 512 g ml⁻¹. A new antibacterial brominated indole alkaloid, identified as 6-bromo-3hydroxy-methyl-indole-2-one, has been isolated and purified from the marine cyanobacterium Anabaena constricta (Volk et al. 2009). Similarly, 2 new metabolites with antibacterial activity have been isolated from the marine cyanobacteria Nodularia harveyana and Nostoc insulare. The two compounds were identified as norharmane (9H-pyrido(3,4-b)indole) and 4,4′-dihydroxybiphenyl, and both inhibited the growth of E. coli, P. aeruginosa, S. aureus and B. subtilis with a MIC between 16 and 160 g ml⁻¹.

Table 2. Antimicrobial compounds from cyanobacteria.

Species of cyanobacteria	Effective against	Source (host)	Location	Inhibitive compound	Reference
Family Oscillatoriaceae
Lyngbya majuscula	Mycobacterium tuberculosis	Seawater	Guam, Pacific Ocean	Depsipepetides (ie pitipeptolides A–F)	Montaser et al. (2011)
L. majuscula	Escherichia coli, Bacillus subtilis, B. cereus, Staphylococcus epidermis	Culture collection	India	Methanol extract	Kaushik et al. (2009)
Family Nostocaceae
Anabaena constricta	B. cereus, Arthrospira laxissima, Nostoc carneum, Chroococcus minutus, and Synechosystis aquatilis	Seawater	Kuwait, Persian Gulf	6-Bromo-3hydroxy-methyl-indole-2-one	Volk et al. (2009)
Nodularia harveyana and Nostoc insulare	E. coli, P. aeruginosa, S. aureus and B. subtilis, Arthrospira laxissima	Seawater	Germany, Baltic Sea	Norharmane (9H-pyrido(3,4-b)indole) and 4,4’-dihydroxybiphenyl	Volk and Furkert (2006)
Family Chroococcaceae
Synechocystis spp., Synechococcus spp.	Gram-positive bacteria	Rocky shores	Portugal, Atlantic Ocean	Dichloromethane extracts	Martins et al. (2008)

Twenty strains of cyanobacteria isolated from hard corals infected with black band disease (BBD) and other marine sources have been tested against heterotrophic bacteria isolated from the surface mucopolysaccharide layer of healthy and infected corals (Gantar et al. 2011). Fifteen out of the 20 isolates exhibited antibacterial activity against at least 1 of the bacterial strains tested and inhibition was higher for BBD cyanobacteria. Synechocystis spp. and Synechococcus spp. had antibacterial activity against Gram-positive bacteria, but not against Gram-negative ones (Martins et al. 2008). Similarly, cyanobacteria isolated from benthic mats of Antarctic lakes have been shown to have antimicrobial activity (Biondi et al. 2008). Seventeen cyanobacteria out of the 48 isolates inhibited the growth of the bacterium S. aureus, the filamentous fungus Aspergillus fumigatus and the yeast Cryptococcus neoformans. The antimicrobial activity of the cyanobacterium Geitlerine-ma strain Flo1 was found to depend on the growth conditions (Caicedo et al. 2011).

Fungi

Many marine fungal isolates demonstrate promising antibacterial activity against pathogens (Table 3). In a recent study, more than half of the 24 fungal strains belonging to the orders Eurotiales, Hypocreales, Pleosporales and Botryosphaeriales associated with the black coral Antipathes dichotoma, exhibited antimicrobial activity against bacterial and fungal pathogens (Zhang et al. 2012a). Eighteen fungal isolates belonging to the Ascomycota from the gorgonian coral Echinogorgia rebekka were screened for antimicrobial activity against 5 pathogenic bacteria (Wang et al. 2011). All isolates showed high to moderate antimicrobial activities, with the genera Penicillium and Cladosporium being the most active (Table 3). In another study, 121 fungal isolates were obtained from 6 species of healthy gorgonian corals from the South China Sea (Zhang et al. 2012b), and ∼38% of the isolates inhibited the growth of environmental bacteria and pathogenic fungi. These facts suggest that epibiotic fungi potentially help their host to prevent microbial fouling.

Table 3. Antifouling activity of marine fungi.

Fungal strain	Effective against	Source (host)	Location (country)	Inhibitive compound	Reference
Eurotiales Hypocreales Pleosporales Botryosphaeriales	Pathogenic microbes	Antipathes dichotoma	China	NI	Zhang et al. (2012a)
18 Fungal isolates	Pathogenic microbes	Echinogorgia rebekka	China	NI	Wang et al. (2011)
121 Isolates belonging to 20 genera	Environmental and pathogenic bacteria	6 Gorgonian corals	China	NI	Zhang et al. (2012b)
Aspergillus sp.	Environmental bacterial strains	Xestospongia testudinaria	China	Aspergiterpenoid A, (-)-sydonol, (-)-sydonic acid, and (-)-5-(hydroxymethyl)-2-(2′,6′,6′-trimethyltetrahydro-2H-pyran-2-yl)phenol	Li et al. (2012)
Cladosporium sp. F14	Environmental bacteria and Bugula neritina	Seawater	China	3-phenyl-2propenoic acid, cyclo(Phe-Pro), cyclo(Val-Pro), bis(2-ethylhexyl)phthalate	Xiong et al. (2009); Qi, Xu, Xiong et al. (2009)
Ampelomyces sp.	Environmental bacteria, Balanus amphitrite and Hydroides elegans	Seawater	China	3-chloro-2,5-dihydroxybenzyl alcohol	Kwong et al. (2006)
Cochliobolus lunatus	The B. amphitrite	Dichotella gemmmacea	China	14 New resorcylic acid lactones	Shao, Wang (2011)
Aspergillus sp.	B. amphitrite	D. gemmmacea	China	Aspergilones A	Shao, Wu (2011)
Penicillium sp. OUCMDZ-776	Bugula neritina	Hypersaline sediments	China	Penispirolloid A	He et al. (2012)
Note: Fungal strains are classified by their activity against bacteria and invertebrate larvae. NI = compound has not been identified.

Several antimicrobial compounds have been isolated from marine fungi (Table 3). The fungus Ampelomyces sp. produced 3-chloro-2,5-dihydroxybenzyl alcohol, which inhibited not only the growth of marine bacterial species, but also prevented settlement of Hydroides elegans and larvae of Balanus amphitrite (Kwong et al. 2006). The fungus Aspergillus sp. isolated from the sponge Xestospongia testudinaria produced 4 bisabolane-type sesquiterpenoids that were effective against environmental bacteria (Li et al. 2012). Similar to bacteria, the cultivation conditions of fungi seem to be an important factor in the production of bioactive compounds (Xiong et al. 2009).

Microbial disruption of biofilms

Microbial biofilms form in a multi-step process of signalling and regulatory events, all of which lead to the development of sessile microbial communities (in a physiological state that is distinct from planktonic cultures) encased in an extracellular polymer matrix. Therefore, microbial biofilms can be disrupted via interference with the signal exchange that leads to biofilm formation or through degradation of the various extracellular polymers that make up the matrices of the biofilm. In this review, the focus is on the N-acyl homoserine lactone (AHL)-based signalling system, as it is among the most intensively studied.

Enzymes that degrade biofilm-forming QS signals

Under laboratory conditions and in natural environments, the formation of a biofilm has been inhibited by enzymatic degradation of the signals involved in biofilm formation (Rajamani et al. 2011; Oh et al. 2012). Turn-over of bacterial AHLs is one of the best studied examples of the enzyme-mediated manipulation of bacterial signalling and biofilm formation. AHL-degrading microorganisms are taxonomically diverse and include true fungi (Ascomycetes and Basidiomycetes), Firmicutes and Gram-negative bacteria (Teplitski et al. 2011). Some bacteria can completely degrade AHLs through the combined action of several enzymes (Huang et al. 2003; Uroz et al. 2003, 2005). AHL-degrading enzymes include lactone hydrolases (lactonases), acylases and oxidoreductases. The most studied AHL-lactonases belong to 2 families: Zn-dependent metallo--lactamases (eg the autoinducer inactivating enzyme (AiiA) of Bacillus sp. as a prototypical member (Dong et al. 2000)) and metallo-dependent phosphotriesterases (Uroz et al. 2008). Since the isolation of the first AHL-degrading enzyme, AiiA and the cloning of the gene aiiA encoding it from Bacillus sp. 240B1 (Dong et al. 2001), over 220 AiiA sequences from various Bacillus spp. have been deposited into the NCBI database.

A novel AHL-acylase was identified and isolated from the marine nitrogen-fixing filamentous cyanobacterium Anabaena sp. PCC 7120 (Romero et al. 2008). This enzyme, named ‘autoinducer inhibitor from cyanobacteria’ (AiiC), showed broad acyl chain length specificity and was produced during the exponential growth phase, but disappeared after 24 h. The AiiC enzyme could be used by the cyanobacterium to control the response of its own QS signals, but can potentially be used to interfere with signalling within mixed microbial communities. The search for the acylase sequences revealed the presence of similar genes in other cyanobacteria, such as Nostoc punctiforme, N. violaceus and Synechocystis sp., which indicate that these enzymes may be widespread among cyanobacteria (Romero et al. 2008).

Soon after AiiA was first identified, this enzyme or its various forms were expressed in other bacteria, plants and unicellular algae to test its applications for controlling virulence, disrupting biofilm formation and manipulating host-associated microbiota (Dong et al. 2001; Rajamani et al. 2011; Oh et al. 2012). In the unicellular alga Chlamydomonas reinhardtii, expression of the codon-optimised AiiA altered the composition of the alga-associated microbiota (Rajamani et al. 2011). When AiiA was expressed in V. cholera, it completely eradicated biofilm formation (Augustine et al. 2010). Transgenic aiiA-expressing bacteria have also been used to disrupt industrial biofilms: the introduction of micro-encapsulated transgenic E. coli strongly reduced biofouling of membranes in wastewater treatment facilities over an extended period of time (Oh et al. 2012). Several naturally occurring bacteria producing AHL-degrading enzymes have been identified as promising agents in reducing biofouling of membranes in wastewater treatment facilities (Oh et al. 2012).

Aside from their function in disrupting biofouling, AHL-degrading bacteria, isolated from healthy fish and aquatic invertebrates, have been shown to be promising biocontrol agents capable of disrupting the virulence of pathogenic Vibrio spp. (reviewed by Tinh et al. 2008; Cao et al. 2012). Thus, the identification of biological functions in marine microorganisms could have wide-reaching applications.

Disruption of biofilms with polymer-degrading enzymes

The extracellular matrix encasing monospecific and multispecies biofilms includes complex mixtures of polysaccharides, proteins and nucleic acids (Kaplan 2009). Marine microbes have been shown to disrupt biofilms through the production of lytic enzymes capable of degrading components of the biofilm matrix. For example, α-amylase from a marine isolate of B. subtilis reduced Vibrio cholera biofilms by 20–80% (Kalpana et al. 2012). Treatment of biofilms formed by human pathogens, eg P. aeruginosa, with α-amylase did not effectively reduce the ability of the pathogen to form a biofilm (Kalpana et al. 2012) as the major components of the biofilm matrix of pathogens (including Vibrio spp.) consist of glucose, galactose, N-acetylglucosamine and mannose (Fong et al. 2010). Treatment of biofilms containing poly-β-1,6-N-acetyl-D-glucosamine with Dispersin B (a -N-acetyl-hexosaminidase) was effective in detaching pre-formed biofilms (Kaplan 2009). However, it is not known how common Dispersin B is in marine bacteria. A NCBI BLAST search revealed only one DspB homologue in the marine strain Photobacterium sp. SKA34 (Accession Number: SKA34_07164). On the other hand, the ability to degrade acetylated hexose amines and polymers containing them is widespread in marine bacteria (Riemann & Azam 2002; Krediet et al. 2009), but it is not yet known whether these N-acetyl-glycosaminidases have a role to play in biofilm dispersal. Furthermore, nucleic acids are becoming recognised as an important component of biofilms, including those formed by V. cholerae (Nijland et al. 2010; Seper et al. 2011). Self-produced and exogenous nucleases reduced biofilm formation by at least twofold (Kaplan 2009; Nijland et al. 2010; Seper et al. 2011). Therefore, a better understanding of the enzymatic activities by which marine bacteria dislodge existing biofilms will likely have further applications in the medical field and probably also for industry (Kaplan 2009).

Disruption of biofilms through inhibition of bacterial signalling and regulatory cascades

In addition to QS signals, such as AHLs, many other small molecules are known to trigger regulatory cascades leading to biofilm formation or dispersal (reviewed by Irie & Parsek 2008; Dobretsov et al. 2009). Inhibition or interference with QS regulation has been explored since the mid-1990s when the first mimics of bacterial AHLs were characterised in the red alga D. pulchra, which is itself mostly free of bacterial biofilms (Manefield et al. 1999). Screens of natural and synthetic products for QS inhibitory activities typically relied on biosensors, which are semi-synthetic constructs in which a reporter, typically genes involved in pigment production, fluorescence or luminescence, are inserted downstream from the promoter responsive to AHLs. Such screens led to the identification of dozens of compounds with presumed QS-inhibitory activities. However, further work with 1 of the reporters (based on the lasRI QS system of P. aeruginosa) revealed that a number of small molecules produced by marine organisms can interfere with the activity of the reporter, although not only through interference with the receptor–AHL interactions (Dobretsov et al. 2010; Kwan et al. 2011).

In addition to compounds capable of modulating QS, several other small molecules with biofilm-inhibitory properties have been purified from marine bacteria. For example, 4-phenylbutanoic acid was isolated from a marine Bacillus sp., which was capable of strongly inhibiting the formation of biofilm by marine bacteria as well as by human pathogens (Nithya et al. 2011). Although the exact mechanism responsible for inhibiting the formation of biofilm remains unknown, the activity of the compound appears not be linked to an inhibition in the production of EPS (Nithya et al. 2011). Using Vibrio vulnificus EPS-defective mutants, it was found that D-glucosamine is responsible for inhibiting the formation of biofilm. It was also shown that glucosamine is inhibitory to the type-2 autoinducer system, yet glucosamine retained its inhibitive properties even after immobilisation on polymeric nanofibers (Kim et al. 2011).

Recently, indole has been identified as a signal molecule involved in the formation of biofilm by V. cholerae (Mueller et al. 2009), controlling the vps genes which regulate the production of Vibrio polysaccharides that make up the bulk of a biofilm matrix. Several natural products (containing indole moieties) extracted from a Mediterranean gorgonian coral were shown to be effective at inhibiting the formation of biofilms by 3 marine bacteria (Penez et al. 2011). Indole-producing bacteria also can inhibit the growth of pathogens (see antibacterial activity). As the genes controlled by indole are now characterised in V. cholerae (Mueller et al. 2009), it will be of interest to test whether the alkaloids produced by marine organisms are capable of interfering with bacterial indole-mediated signalling.

Cyanobacteria produce siderophores to enable them to obtain iron from the environment. Several types of cyanobacterial siderophores are known, such as schizokinen, synechobactins and anachelins (Goldman et al. 1983; Ito et al. 2004; Ito & Butler 2005). Recently, the catecholate fragment of the siderophore anachelin was obtained from the cyanobacterium Anabaena cylindrica and was used to immobilise polyethylene glycol (PEG) on surfaces in order to prevent the formation of biofilms (Gademann et al. 2009). This fragment was further modified by merging the anachelin chromophore with the antibiotic vancomycin through a PEG linker. This mixture was used for the generation of antimicrobial surfaces on implants, catheters and stents, and thus has important medical applications.

Antialgal activity

Heterotrophic bacteria

In this review, antialgal activity is defined as the prevention of settlement of algal spores and diatom cells as well as inhibition of growth of diatoms and microalgae. Bacteria associated with the green alga Ulva lactuca have been shown to have antibacterial and antialgal activity (Table 4; Rao et al. 2007; Kumar et al. 2011). Eighty percent of epibiotic isolates from U. lactuca inhibited growth of the diatom Cylindrotheca fusiformis (Kumar et al. 2011). All isolates had differential activity; Pseudoalteromonas isolates were active against bacteria and diatoms, while members of the genera Bacillus, Vibrio and Shewanella demonstrated antidiatom activity only.

Table 4. Inhibition of the settlement of algal spores and diatoms by marine heterotrophic bacteria.

Bacteria	Effective against	Source (host)	Location	Inhibitive compound	References	Comments
10 Bacterial isolates	Cylindrotheca fusiformis	Ulva lactuca	South Pacific Ocean, Fiji	NI	Kumar et al. (2011)	Species specific anti-diatom and anti-bacterial activity
Alteromonas sp.	Eight benthic diatoms and spores of Ulva sp.	Water column	South Pacific, Chile	Theromostable hydrophilic peptide	Silva-Aciares et al. (2008)
Leucothrix mucor	Navicula annexa and spores of Ulva pertusa	Seawater	South Sea, Korea	17-(1,2-Dihydroxyl-5-methyl-hexane)-2,3-dihydroxyl-cholest-4-en-6-one [compound and 13-acetate-17-(1,5-dimethylhexane)-cholest-7-en-3,5,6,15-tetraol	Cho (2012)
Shewanella oneidensis SCH0402	Spores of Ulva pertusa; barnacles; mussels	Seawater	Yellow Sea, Korea	2-Hydroxymyristic acid and cis-9-oleic acid	Bhattarai et al. (2007)	Field experiments showed good AF effect after 1.5 years
22 Bacterial isolates	Spores of Ulva sp.	Water column	Baltic Sea	NI	Bernbom et al. (2011)	Strongest activity among Pseudoalteromonas strains
Pseudoalteromonas haloplanktis CI4	Spores of U. pertusa	Seawater	South China Sea	Heat sensitive 3–10 kDa, not protein	Ma et al. (2010)
Pseudoalteromonas tunicata and Phaeobacter sp. strain 2.10	Polysiphonia sp.	Ulva australis	Pacific Ocean	NI	Rao et al. (2007)
Note: Bacterial strains are classified by their activity. NI = compound has not been identified.

Bacterial isolates from seawater have been shown to have antialgal activities (Table 4). Biofilms of the bacterium Alteromonas sp. Ni1-LEM inhibited settlement of nine benthic diatoms and spores of Ulva sp. more efficiently than the reference strains H. marina and Pseudolateromonas tunicata (Silva-Aciares & Riquelme 2008). It was shown that the AF compound is thermostable and a hydrophilic protein >3500 Da is produced extracellulary by the bacterium. The bacterium Pseudoalteromonas haloplanktis CI4 was shown to produce a large (3–10 kDa), heat sensitive, non-protein compound that inhibited germination of zoospores of U. pertusa (Ma et al. 2010). Two AF steroids were isolated from the filamentous bacterium Leucithrix mucor (Cho 2012). These compounds inhibited settlement of zoospores of U. pertusa and the diatom Navicula annexa. The marine bacterium Shewanella oneidensis SCH0402 produced 2-hydroxymyristic acid and cis-9-oleic acid, which inhibited germination of spores of U. pertusa (Bhattarai et al. 2007). In field bioassays, these compounds were as efficient as the toxic AF compound tributyltin oxide.

Antilarval activity

Heterotrophic bacteria

Biofilms of many epibiotic bacteria inhibit settlement of invertebrate larvae in the laboratory (Table 5), but it is uncertain if the same bacteria exhibit this activity under natural conditions. Recent investigations provided evidence that some epibiotic biofilms could indeed prevent larval settlement on the surface of algae in natural conditions. For instance, Pseudoalteromonas tunicata and Phaeobacter sp. strain 2.10 (formerly Roseobacter gallaeciensis) isolated from the alga U. australis inhibited larval settlement of the bryozoan Bugula neritina at densities of 10³ to 10⁵ cells cm⁻², which are similar to the densities of these bacteria under natural conditions (Rao et al. 2007). Multispecies biofilms with similar composition to those from the alga Fucus vesiculosus inhibited the settlement of cyprid larvae of the barnacle Amphibalanus improvisus (Nasrolahi et al. 2012).

Table 5. Inhibition of the settlement of invertebrate larvae by marine bacteria.

Bacteria	Effective against	Source (host)	Location	Inhibitive compound	Reference	Comments
Heterotrophic bacteria
Pseudoalteromonas tunicata and Phaeobacter sp. strain 2.10	Bugula neritina	Ulva australis	Pacific Ocean	NI	Rao et al. (2007)
Pseudoalteromonas issachenkonii UST041101–043	B. neritina	Deep sea sediments	Pacific Ocean	Protease	Dobretsov et al. (2007)	Active in laboratory and field experiments
Natural communities, Photobacterium halotolerans and Ulvibacter litoralis, Shewanella basaltis Pseudoalteromonas arctica, Shewanella baltica and Bacillus foraminis	Amphibalanus improvisus	Fucus cerratus, F. vesiculosus and Polysiphonia stricta	Baltic Sea	NI	Nasrolahi et al. (2012)
Streptomyces fungicidicus	Balanus amphitrite	Deep sea sediments	West Pacific Ocean	Five Diketopiperazines	Li et al. (2006)
Pseudomonas rhizosphaerae	B. amphitrite; B. neritina	Deep sea	Pacific Ocean Province A	Bis(2-ethylhexyl)phthalate, cyclo(Tyr-Ile), cyclo(Phe-Pro), cyclo(Val-Pro) 3-phenyl-2-propenoic acid	Qi, Xu, Gao (2009)	Both antimicrobial and antilarval activities
Pseudoalteromonas sp. sf57	Hydroides elegans	Natural biofilms	South China Sea	NI	Huang et al. (2011)	Inhibition correlates with bacterial pigmentation
Streptomyces sp. UST040711–290	H. elegans	Deep sea sediments	Pacific Ocean Province A	Fatty acid 12-methyltetradecanoid acid	Xu et al. (2009)	Demonstrated effect of this compound on larval gene expression
Winogradskyella poriferorum	H. elegans; B. amphitrite	Lissodendoryx isodictyalis	Caribbean, Atlantic Ocean	Poly-ether A-E	Dash et al. (2009); Dash et al. (2011)	Inhibitive concentrations of 70 and 60 g ml^−1
Streptomyces sp.	H. elegans; B. neritina and B. amphitrite	Deep sea sediments, seawater	West Pacific Ocean	Five compounds belonging to α, β unsaturated lactones	Xu et al. (2010)	Based on structure relations antifouling compound was synthesised
Macrococcus sp., Bacillus sp., Pseudoalteromonas sp.	Aulacomya maoriana	Biofilms	South-West Pacific Ocean, New Zealand	NI	Alfaro et al. (2011)	Highest inhibition for bacterial exudates
Shewanella oneidensis SCH0402	Barnacles, mussels	Seawater	Yellow Sea, Korea	2-Hydroxymyristic acid and cis-9-oleic acid	Bhattarai et al. (2007)	Field experiments showed good AF effect after 1.5 years
Cyanobacteria
Oscillatoria sp. and Hormoscilla sp.	Biomphalaria glabrata	Seawater	Palmyra Atoll, Northern Pacific Ocean	Thiopalmyrone and palmyrrolinone	Pereira et al. (2011)	LC50 = 6–8.3 M
Lyngbya bouillonii	B. glabrata	Seawater	Papua New Guinea, South Pacific Ocean	Cyanolide A	Pereira et al. (2010)	LC50 = 1.2 M
L. majuscula	Amphibalanus (Balanus) amphitrite	Seawater	Caribbean, Atlantic Ocean	Isolmajusculamide, majusculamide A and hanutupeptin C	Tan and Goh (2009)
Note: Bacterial strains are classified by their activity. NI = compound has not been identified.

Microbial biofilms affect adhesion of larvae and spores of fouling organisms. In the study of Zardus et al. (2008) settlement of the ascidian Phallusia nigra, the polychaete H. elegans, the bryozoan B. neritina and the barnacle Balanus amphitrite was investigated in the laboratory. Only adhesion of B. neritina larvae was unaffected, while adhesion of the other species was facilitated. In contrast, the bacterium Cobetia marina inhibited the adhesion strength of spores of Ulva linza (Mieszkin et al. 2012). It is not clear how bacteria facilitate or inhibit adhesion of other species but it is possible that multiple biochemical, behavioural and physical mechanisms may be involved (Zardus et al. 2008).

The AF activity of certain bacteria has been found to correlate with bacterial colour. This unexpected fact was described initially by Egan et al. (2002) for the bacterium P. tunicata. Purple mutants of P. tunicate lost AF activity because of a disruption of the gene wmpR, which is involved in the synthesis of AF compounds. Recently, a correlation between bacterial colour and AF activity has been shown for another bacterium, Pseudoalteromonas sp. sf57, which inhibits settlement of H. elegans larvae (Huang et al. 2011). Several mutants of this bacterium were made by transposon mutagenesis. White mutants formed thicker biofilms and became inductive to settlement by larvae. Further analysis suggested that the type II secretion pathway, the LysR transcriptional regulator, NAD(P)-binding proteins, flagella and cell membrane processes could play a role in the regulation of bioactivity in Pseudoalteromonas sp. sf57.

The ability of bacteria incorporated in a non-toxic matrix to inhibit larval settlement has been shown previously (Holmström et al. 2000). This ‘living paint’ contained the bacterium P. tunicata and was able to inhibit barnacle fouling for 14 days. A recent study with the same bacterial species demonstrated its ability to remain viable for up to 12 months when incorporated into a kappa-carrageenan matrix (Yee et al. 2007). Field experiments in Sydney harbour demonstrated the ability of these ‘living paints’, with P. tunicata able to inhibit fouling for up to 7 weeks. This suggests that by using the right matrix, it is possible to create a ‘living paint’ with a longer period of AF defence.

During the last 6 years, several AF compounds have been isolated and identified from marine bacteria (Table 5). These include 6 poly-ethers A–E isolated from the epibiotic bacterium Winogradskyella poriferorum (Dash et al. 2009, 2011). These natural products inhibited the settlement of the barnacle B. amphtrite, the polychaete H. elegans and the bryozoan B. neritina. The compounds did not have any toxic effect on the development of zebra fish embryos, rendering them promising candidates for future AF applications.

Bacteria living in extreme environments, such as the deep sea, hot springs, hydrothermal vents and hypersaline lakes, may prove to be a rich source of novel AF compounds. The first AF compound from deep sea microorganisms was isolated from the bacterium Streptomyces fungicidicus originating from sediments 5000 m deep in the Western Pacific Ocean (Li et al. 2006). This bacterium produced 5 DKPs, which significantly inhibited the settlement of the barnacle B. amphitrite. One of the DKPs, viz. cyclo-(L-Val-L-Pro) had the highest activity and therapeutic ratio, and this makes it a good candidate for future AF applications. Analogously, Pseudomonas rhyzoshaerae from the Pacific Ocean produced DKPs that inhibited the settlement of B. amphitrite and B. neritina (Qi, Xu, Gao et al. 2009). Pseudoalteromonas issachenkonii UST041101–043 was isolated from a depth of 3000 m in the Aleutian margin of the Pacific Ocean and produced a protease that inhibited the settlement of the larvae of the bryozoan B. neritina at a concentration of 1 ng ml⁻¹ (Dobretsov et al. 2007). This protease and commercially available trypsin incorporated in a water-soluble paint matrix significantly reduced the settlement of bryozoans and the barnacle B. amphitrite (Dobretsov et al. 2007). Eleven Streptomonas strains isolated from deep sea sediments were screened for antilarval activity (Xu et al. 2009). An AF compound identified as 12-methyltetradecanoid acid (12-MTA) was isolated from the bacterium Streptomonas sp. UST040711–290 (Table 3). Treatment of competent larvae of H. elegans with 12-MTA down-regulated the expression of GTPase-activating protein and up-regulated the expression of ATP synthase, thus preventing settlement (Xu et al. 2010). This observation suggests that certain genes are important for larval settlement and metamorphosis. Genetic methods can be important tools to analyse the mechanisms of action of AF compounds and to find targets for new compounds.

Cyanobacteria

Among the previously known molluscicidal compounds isolated from cyanobacteria are diterpenoid comnostin B produced by Nostoc commune (Jaki et al. 2000) and maculalactone A isolated from the marine heterocystous cyanobacterium Kyrtuthrix maculans (Brown et al. 2004). Comnostin B inhibited the growth of the mollusc Biomphalaria glabrata whereas maculalactone A inhibited the growth of larvae from various species of barnacle including Tetraclita japonica, Ilba cumingii and B. amphitrite. Recently, new molluscicidal metabolites were isolated from extracts of environmental assemblages of Oscillatoria and Hormoscilla spp. (Table 5; Pereira et al. 2011). These compounds were identified as thiopalmyrone and palmyrrolinone, and they were shown to inhibit the growth of the snail B. glabrata (LC₅₀ = 8.3 and 6.0 M, respectively). Another molluscicidal agent against B. glabrata identified as Cyanolide A was isolated from Lyngbya bouillondii, a cyanobacterium originating from Papua New Guinea (Pereira et al. 2010).

L. majuscula is a marine cyanobacterium with a wide distribution. The species has recently been studied for the production of several bioactive compounds including those with AF activity (Paul et al. 2011). More than 113 bioactive compounds were isolated from this organism, out of which 10% possessed antifungal and antimicrobial activity (Burja et al. 2001). Extracts of L. majuscula isolated from the western lagoon of Pulau Hantu, Singapore were incorporated into a Phytagel^TM matrix and the development of fouling communities was followed over 4 weeks (Tan & Goh 2009). The data showed that extracts from L. majuscula significantly inhibited the settlement of barnacles. Based on a settlement assay using the barnacle Amphibalanus (Balanus) amphitrite, a number of AF products were purified and identified as isolmajusculamide, majusculamide A and hanutupeptin C (Tan & Goh 2009).

Fungi

Several AF compounds have been isolated recently from marine fungi (Table 3). Two fungal species, Cochliobolus lunatus (Shao, Wang et al. 2011) and Aspergillus sp. (Shao, Wu et al. 2011), were isolated from the gorgonian coral Dichotella gemmmacea. The fungus C. lunatus was shown to produce 14 new resorcylic acid lactones that inhibited the settlement of cyprid larvae of the barnacle B. amphitrite. The fungus Aspergillus sp. produces aspergilones A that inhibited the settlement of B. amphitrite with an EC₅₀ value of 7.68 μg ml⁻¹ (Shao, Wu et al. 2011). Penispirolloid A produced by Penicillium sp. OUCMDZ-776 isolated from hypersaline sediment in China inhibited the settlement of larvae of B. neritina with an EC₅₀ of 2.40 μg ml⁻¹ (He et al. 2012). The marine fungus Cladosporium sp. F14 produced 3-phenyl-2propenoic acid and bis(2-ethylhexyl)phthalate, which effectively inhibited the settlement of B. neritina and B. amphitrite (Qi, Xu, Xiong et al. 2009).

Looking ahead

Epibiotic microorganisms associated with marine algae, sponges, corals and other invertebrates have high AF potential. Whilst the question of whether these epibiotic microorganisms protect their hosts from biofouling remains open, without doubt, epibiotic microorganisms remain a potential target for the isolation of novel AF compounds. Since the review of Dobretsov et al. (2006), a number of AF compounds have been isolated from deep sea bacteria and marine fungi. These groups of microorganisms deserve further investigation, especially considering the fact that several patents based on their AF properties have been filed (Melde & Markowitz 2012; Qian et al. 2012).

It is clear that some compounds produced by microorganisms can have multiple functions. For example, indole is involved in the formation of biofilms by V. cholerae (Mueller et al. 2009) and has significant activity against bacterial pathogens (Devi et al. 2010). DKPs produced by many bacteria can work as signal molecules (Tommonaro et al. 2012), inhibit the formation of biofilms (de Carvalho & Abraham 2012), the growth of bacteria (Qi, Xu, Xiong et al. 2009) and the settlement of larvae (Li et al. 2006; Qi, Xu, Gao et al. 2009). Compounds which inhibit both microfouling and macrofouling should be investigated further as future potential AF agents.

Isolation of compounds from multispecies communities, such as microbial mats and complex biofilms, not from single isolates, is a potential new way of obtaining novel AF compounds. This approach is based on the fact that some bioactive compounds are only produced by microorganisms when in mixed cultures or consortia. The challenges of this method include difficulties in obtaining AF compounds in sufficiently large quantities and changes in the production of compounds due to changes in the microbial communities. Additionally, antifoulants might not be effective against microbes present in the same biofilm. In spite of these drawbacks, it was shown that cyanobacterial mats from hot springs and biological desert crusts produce antibacterial and QS inhibitory compounds under in situ conditions (Abed et al. 2011; Dobretsov, Abed et al. 2011). It is of note that bioactivity in the microbial mats was correlated with the diversity of the microbial communities; extracts of less diverse microbial mats from hot springs had higher antibacterial activity than the extracts of more diverse mats (Dobretsov, Abed et al. 2011).

The AF properties of dozens of marine compounds are currently being characterised and it is apparent that many more are still to be discovered. Even though many compounds can inhibit biofouling in laboratory conditions, very little is known about their performance in the field (Dobretsov, Teplitski et al. 2011). From an ecological point of view, it is important to learn whether these compounds are produced in situ and whether they are produced in quantities that are sufficient to trigger meaningful responses in marine organisms. The question regarding the production of bioactive compounds in situ by bacteria is especially relevant, since, unlike macro-organisms and filamentous bacteria, collecting ‘wild’ samples of bacteria in the volumes suitable for chemical isolation and identification is not a trivial undertaking. It is also commonly accepted that the production of bioactive compounds by organisms is greatly affected by the presence of specific neighbouring species, nutrients, signals and environmental factors (eg temperature, salinity and light). Therefore, establishing whether AF compounds are produced in situ seems especially important for understanding their functions. The deployment of ‘chemical traps’ (essentially dialysis pouches filled with hydrophobic resins) has helped to demonstrate that compounds inhibitory to biofilms are present within the mucus-associated microbial communities on the surface of corals (Alagely et al. 2011), and although the chemical nature of these activities was not determined, this was the first report of the presence of such bioactivity on the surface of healthy corals.

Recent advances in molecular tools will facilitate the analysis of the diversity and function of fouling microbial communities, which will help in designing strategies to reduce their development. Pyrosequencing has the advantage of creating thousands of sequences read at a very low cost and thus provides a comprehensive overview of the structure of fouling communities. When this technique is combined with other metabolomic, transcriptomic and proteomic techniques, more information will be obtained on the metabolism and interaction among different groups of organisms within fouling communities. The identification of transcripts (transcriptomic) and polypeptides (proteomics) differentially accumulated in response to a particular AF compound will help restrict the potential molecular targets of the chemicals of interest. Once it is established that the activities of interest are produced in situ and have an effect at meaningful concentrations, it should be feasible to test how the composition of microbial communities changes in the presence of the AF activities. Other techniques, such as Nano-SIMS and Raman spectroscopy, can be used to study the interaction between organisms in field samples as well as in model cultures. Recent developments in single-cell technologies including sorting individual cells and sequencing their genomes should be useful in exploring the genetic makeup of chemically rich microorganisms. Relevant genes that code for certain AF compounds may be cloned and optimised for commercial production.

Several studies have demonstrated the usefulness of natural compounds and enzymes in preventing and disrupting biofilms and dozens of patents have been filed. However, to date these compounds are not widely used in controlling biofouling on industrial scales. Clearly, extraction of marine micro- and macro-organisms cannot provide a sustainable and cost-effective method of obtaining pure compounds suitable for industry. Therefore, the production of natural products derived from microbes has to be through scale-up in bioreactors or by chemical synthesis. Questions regarding the toxicity of compounds to other marine organisms also need to be addressed prior to wider commercial application. Overall, this review clearly shows that marine microorganisms are an important source of antibacterial, antialgal and antilarval compounds, which can potentially be used for solving the biofouling problems of industry and for other biotechnological applications.

Acknowledgements

The authors thank Professor S. Smith (University of Miami, USA) for her constructive comments and for proof-reading the manuscript. The work of SD was supported by a Sultan Qaboos University internal grant IG/AGR/FISH/12/01 and by a HM Sultan Qaboos Research Trust Fund SR/AGR/FISH/10/01. SD acknowledges the help of Professor R. Coutinho (IEAPM, Arraial do Cabo, Brazil) and the programme Science without Frontiers (CNPq).