Unravelling the sponge microbiome as a promising source of biosurfactants

Abstract Microbial surfactants are particularly useful in bioremediation and heavy metal removal from soil and aquatic environments, amongst other highly valued uses in different economic and biomedical sectors. Marine sponge-associated bacteria are well-known producers of bioactive compounds with a wide array of potential applications. However, little progress has been made on investigating biosurfactants produced by these bacteria, especially when compared with other groups of biologically active molecules harnessed from the sponge microbiome. Using a thorough literature search in eight databases, the purpose of the review was to compile the current knowledge on biosurfactants from sponge-associated bacteria, with a focus on their relevant biotechnological applications. From the publications between the years 1995 and 2021, lipopeptides and glycolipids were the most identified chemical classes of biosurfactants. Firmicutes was the dominant phylum of biosurfactant-producing strains, followed by Actinobacteria and Proteobacteria. Bioremediation led as the most promising application field for the studied surface-active molecules in sponge-derived bacteria, despite the reports endorsed their use as antimicrobial and antibiofilm agents. Finally, we appoint some key strategies to instigate the research appetite on the isolation and characterization of novel biosurfactants from the poriferan microbiome.


Introduction
Since ancient times, ocean environments have been severely impacted by human action. Following the discovery of crude petroleum and gas deposits beneath the rocky seabed at the end of the nineteenth century, offshore oil drilling has been posing a critical form of environmental risk (Cordes et al. 2016). The latest dramatic example of oil spill came in 2019 as a mysterious occurrence of crude oil on the Northeast Coast of Brazil (Escobar 2019). This puzzling spill caused a considerable decrease in the fish sale by 53.7%, tourism interest in the region, affected the health of the local population, given the consumption and handling of contaminated seafood (Ramalho 2019;de Oliveira Estevo et al. 2021), and disrupted the local structure of some key members of the benthic community (Craveiro et al. 2021).
Although continued transportation, manipulation, and exploration can result in these accidental spills, oil is still a central resource for the world economy and there is a long path until its complete replacement for renewable energy sources (Singh and Singh 2012). In this framework, techniques to mitigate the possible impacts of oil and its derivatives have been gaining greater attention in the last decades. Bioremediation, the application of ex situ or in situ (micro-)biological agents for the breakdown and natural removal of polluting contaminants, emerges as one of the mainstream strategies for the restoration of oil-contaminated areas (Azubuike et al. 2016). Essentially, petroleum-derived hydrocarbons are enzymatically converted into nontoxic compounds in an overall eco-friendly and costeffective process in comparison with physical and chemical remediation approaches (Xue et al. 2015;Prince and Atlas 2019). For the efficient uptake of oil fractions, hydrocarbon-degrading microorganisms have evolved a versatile arsenal of surface-active molecules, which can increase the water solubility of oil-deriving hydrocarbons, enhancing the bioavailability of hydrophobic substrates and, consequently, the rate of their microbial catabolism. These molecules are broadly named biosurfactants (Ron and Rosenberg 2019).
Biosurfactants are defined as amphiphilic surfactants consisting of hydrophilic and hydrophobic portions that can reduce surface and interfacial tension, allowing mixing between different polar and non-polar phases of a solution (Banat et al. 2010;Uzoigwe et al. 2015). These surfactants can therefore associate with oil molecules to transform them into smaller particles, which can potentialize the biodegradation by indigenous microorganisms (Karlapudi et al. 2018). Microbial surfactants can be part of the microbial cell envelope or be released extracellularly and belong to a variety of chemical classes (Mulligan 2005;Dell'Anno et al. 2018). However, they are mainly divided into: (a) biosurfactants with a high capacity for the formation of emulsions, also referred to as bioemulsifiers, including lipopolysaccharides, polysaccharides, proteins, or a complex mixture of these compounds; (b) and surface-active agents more effective on surface tension reduction, generally categorized as biosurfactants, encompassing fatty acids, glycolipids, phospholipids, lipopeptides, and lipoamino acids (Jahan et al. 2020;Nikolova and Gutierrez 2021).
As abovementioned, biosurfactants are generally of microbial origin, in contrast to commercially synthetic surfactants used in the bioremediation process, which come mostly from petrochemicals and are directly associated with harmful changes in the environment, notably cytotoxic effects to the dwelling fauna and microbiome (Francis and Diemand 2011;Zheng et al. 2014;Varjani and Upasani 2017). In this regard, the influence of synthetic surfactants has been already evaluated in comparison with the overall effects caused by biosurfactants in natural microbial communities.
Unsurprisingly oil spills and all the activities from the petroleum and gas industries can provoke dramatic effects on macrobenthic fauna structure (Zhou et al. 2019), including marine sponges (Vad et al. 2018). Natural engineers of the benthic ecosystems, sponges (phylum Porifera) are amongst the remarkable filterfeeding animals in aquatic environments, with a seminal estimative of around daily 24 000 L filtered seawater per 1.0 kg of sponge biomass (Vogel 1977;Taylor et al. 2007). The sessile nature of these invertebrates exposes them to the increasing accumulation of different substances present in the surrounding water, including particles, pollutants, and microorganisms. A natural consequence of this striking filter-feeding activity was the establishment of a long-term, stable, and successful association between poriferans and microorganisms from all three life domains (de Oliveira et al. 2020b). Currently, sponges are recognized to house more than 60 prokaryotic phyla, which are the main drivers of a myriad of metabolic, defensive, and adaptive functions of crucial importance for the homeostasis of the invertebrate host (Pita et al. 2018).
Considering the key ecological roles played by sponges in marine ecosystems and their susceptibility to several environmental harms imposed by anthropogenic activity, biosurfactants released by their inhabiting microbial consortium are likely to be essential in guarantying the survival and maintaining the resilience of these invertebrates in their habitats (Santos-Gandelman et al. 2014c). Under the conceptual classification of functions played by the sponge microbiome (Pita et al. 2018), we here hypothesize that sponge microbial biosurfactants operate in two different ways within the sponge holobiont. First, as a metabolic function, in which the production of biosurfactants is induced in sponge-associated microorganisms for the emulsification of free hydrocarbon-based compounds. Once emulsified and solubilized by the surfactant action, their catabolism is facilitated by both the sponge and the environmental microbiomes. Another likely scenario is a defensive function, where these surfactants are released by sponge-derived microorganisms to inhibit the growth and fouling or kill opportunistic microorganisms and predators. Hence, these biosurfactants would protect both the sponge host and its microbial symbiotic consortia against eventual external invasions. Unfortunately, neither of these ecophysiological roles of sponge microbial biosurfactants have been tested by in vivo experiments.
Biotechnologically, sponge-bacterial symbioses have been particularly exploited for a plethora of biologically active substances (Santos-Gandelman et al. 2014c), mostly antimicrobial substances (Indraningrat et al. 2016), cytotoxic compounds (Zhang et al. 2017), and industrial biocatalysts (de Oliveira et al. 2020a). While there are a growing number of reports on these biotechnologically relevant bioproducts from the sponge microbiome, little effort has been performed on reviewing the biosurfactant molecules produced by members of the sponge-associated microbial communities. Recently, marine host-microbe associations have been estimated as promising but still-neglected biosurfactant reservoirs (Kubicki et al. 2019). In addition to their functional roles in this holobiont model, biosurfactants from sponge-associated microorganisms could therefore also represent potential alternatives for bioremediation purposes and petroleum, food, detergent, and pharmaceutical industries.
This review concerns biosurfactants exclusively from the sponge-derived bacteria and their biotechnological applications primarily in environment, but also in the biomedical area and other noteworthy industrial sectors. Different strategies are finally pointed out for the future isolation, characterization, and multiple uses of surfactants from sponge-dwelling microorganisms. We hope to place sponge microbial biosurfactants in a higher level of importance and, finally, instigate research for prospecting these bioproducts in this exemplar metazoan-microbe association.
Topic 1: biosurfactant substances from spongeassociated bacteria: an overview All the studies on biosurfactant substances from sponge-associated bacteria are chronologically displayed in Supplementary Table S1. Figure 1 illustrates the number of publications in the years 1995-2021 obtained from the available literature in PubMed, ScienceDirect, Web of Science, Google Scholar, Google Patents, Latipat-Espacenet, Lens, and INPI (National Institute of Industrial Property, free translation from Portuguese) databases (search filters used were the words "sponge", "sponge-associated bacteria", "spongeassociated fungi", "fungi", "yeast", "bacteria", "surfactant", "emulsifier", "biosurfactant", and/or "bioemulsifier"). During the search, 29 studies were found, with a study from 2016 using the same strain (NIOT-06) from other published work (Anburajan et al. 2015(Anburajan et al. , 2016; however, no patent was filed. Although about 26-year interval, it was not possible to observe a consistent rise trend in the number of studies, even though in some years there was a sudden increase compared to the previous one followed by a decay of the total sum of reports. 2014 was the year with the highest number of studies. Furthermore, studies with extraction and chemical characterization of the biosurfactant substances represent 81.5% of all publications. Screening reports, which were only restricted for the evaluation of the biosurfactant-producing capacity of bacterial strains, summed the remaining one-fifth of works. In this context, the drop collapsing and oil displacement assays, the determination of the emulsification index (EI 24 ) with distinct hydrocarbon compounds, and detection of lipolytic and haemolytic activities were the most recurrent screening methodologies used to detect the biosurfactant production ability of the sponge-associated bacterial strains. On the other hand, some studies accomplished the full chemical identification of the biosurfactant substances extracted from spongederived bacterial cultures in the supernatant extract by high-performance liquid chromatography (HPLC) (Kalinovskaya et al. 1995;Anburajan et al. 2016). The detection of the presence of biosurfactant genes in sponge-derived bacteria was conducted just in six reports (Kiran et al. 2010d;Anburajan et al. 2015Anburajan et al. , 2016Meenatchi et al. 2020). Moving to the chemical nature of biosurfactants revealed in the characterization studies, two main classes were reported, with lipopeptide constituting 39% of all studies, followed by glycolipid, which totalized 29% of the reports (Figure 2(A)). Amongst the proposed applications for these surfactant substances, bioremediation stands out as the most endorsed, succeeded by enhanced oil recovery (EOR) and antibiofilm potential (Figure 2(B)).
Considering the taxonomic affiliation of the spongederived bacterial producers of biosurfactants, Firmicutes is the dominant phylum, with the majority of the members included in the family Bacillaceae. Remarkably, the bacterial producers were from the Bacillus genus in one-third of all studies. Actinobacteria come in the second place, with Brevibacterium and Nocardiopsis as the most reported genera, and Proteobacteria in the third top phylum, with most of the representative producers affiliated to the Alphaproteobacteria class (Figure 2(C)).
The search in the databases (with the search filters abovementioned) found solely two reports about the eukaryotic surfactant producers recovered from sponge sources, of which, two were subsequently characterized as rhamnolipid and glycolipoprotein (Kiran et al. , 2010c. Accordingly, four marine fungi previously isolated from Fasciospongia cavernosa ) showed positive results for biosurfactant production. In particular, Aspergillus ustus MSF3 exhibited the highest emulsification activity (42.8%), 78.50 mm 2 of oil displacement. Optimization of biosurfactant production with A. ustus MSF3 was successfully achieved under 20 C, 3% salt concentration, and the chemical analysis partially characterized the biosurfactant as a glycolipoprotein. Notably, this fungal surfactant displayed a broad-spectrum antimicrobial activity and was successfully applied for microbially enhanced oil recovery (MEOR) even in temperatures as high as 50 C ). Ultimately, there is only one report demonstrating biosurfactant bacterial producers isolated from nondemosponge specimens, from the Calcarea class, Paraleucilla magna (Santos-Gandelman et al. 2014a). This outcome comes as no surprise, since Demospongiae is the main class of the phylum Porifera, covering up to 82.8% (7831) of the total accepted sponge species (de Voogd et al. 2021). Moving to the sponge genera, Dendrilla nigra is the leading source from which biosurfactant-producing bacteria have been hitherto isolated. This may have been motivated by the fact that most of the reports of sponge microbial biosurfactants are from the same Indian research group (Kiran et al. , 2010a(Kiran et al. , 2010b(Kiran et al. , 2010e, 2014Hema et al. 2019) who were able to identify these molecules and unveil several of their biotechnological applications Lipton 2004a, 2004b). Therefore, the geographic origins of the source sponge specimens are mostly restricted to the coastlines and sea zones of the Indian subcontinent (Table S1).
Topic 2: bioremediation potential of biosurfactants from sponge-associated bacteria About 33% of the application proposals of biosurfactants produced by sponge-associated bacteria correspond to bioremediation of environments contaminated with hydrocarbons and heavy metals (Figure 2(B); Table  S1). These data are of particular interest since it shows a reply to the ecological claim in the face of the impacts caused by the oil spills and leakages. Foreign biosurfactants are usually sought for application in marine habitats. This happens because when remediation in oil-contaminated environments is based exclusively on indigenous microorganisms, the potential for remediation is expected to be rather limited due to their extremely low cell concentrations in the aquatic environment. Sufficient release of the surfactant molecules is an important requirement for their effective and quick interaction with free hydrocarbons, allowing posterior biodegradation by microbial metabolism. Thus, a higher cellular density from the microbial producers is highly desirable to achieve decent bioremediation rates (Ron and Rosenberg 2002;Nikolova and Gutierrez 2021). Once their biosurfactant potential is confirmed, cultivable marine sponge-associated microorganisms might represent an interesting option as sources of these surface-active molecules for bioremediation in marine environments, especially if their natural growth in saline conditions is optimized for the obtainment of superior surfactant yields for further uses (Kiran et al. 2010d).
Given the proposed use of biosurfactants from sponge-derived bacteria in bioremediation and considering the potential of these symbiotic microorganisms to reaching the higher biosurfactant concentrations, Gandhimathi et al. (2009) were the first research group to perform the optimized production and characterization of a lipopeptide biosurfactant by Nocardiopsis alba strain MSA10, an actinobacterium recovered from the sponge Dendrilla nigra. Five different screening methods were initially used to select potential surfactant producers, including: the measurement of the emulsification index, as if a bacterial culture produces biosurfactant, then it will emulsify the hydrocarbon present in the solution; and indirect tests, such as drop collapsing and oil displacement tests, which were properly applied for the determination of the surface tension activities. Moreover, paddy straw was the cheapest raw material used as a carbon source during the optimization of the lipopeptide production by the N. alba MSA10 showed the highest emulsifying activity. Thus, this spongederived actinobacterial was particularly praised as a sustainable lipopeptide producer for bioremediation in oil-polluted and heavy metal-contaminated environments ).
Envisaging the reduction of biosurfactant production costs, Kiran et al. (2010e) conducted a study based on the solid-state cultivation method, in which microorganisms grow on non-soluble material or moist organic solid substrates (Krieger et al. 2010). Moreover, solidstate cultivation can allow cost reduction during biosurfactant production, when compared to the widely used submerged fermentation. Overall, a smaller size bioreactor can be used, energy requirements for agitation can be diminished, and superior product recovery can be obtained from the solid-state culture (SSC) (Pandey 2003;Banat et al. 2021). In this context, Nocardiopsis lucentensis MSA04, a strain also isolated from D. nigra, was demonstrated as a potential glycolipid producer using SSC. The results of the established hydrocarbonsurfactant system indicated that the cellular hydrophobicity with kerosene was 85%, which was the highest index amongst the five other tested potential surfactant bacterial producers selected in the study. Following optimization of the surfactant production by response surface methodology and its complete extraction and purification, the MSA04 glycolipids displayed broad thermostability, even after autoclaving, invariable stability over a pH 5.0-9.0 and 1.0-4.0% NaCl concentrations. In addition to the superior emulsifying indexes when compared to synthetic surfactants, the emulsion formed by the actinobacterial glycolipids was stable for more than three months. These attracting results led the authors to support the use of this other sponge-associated actinobacterial strain as an outstanding biosurfactant source for bioremediation processes (Kiran et al. 2010e).
In this sense, the report with a surfactin-producing Bacillus licheniformis NIOT-AMKV06, isolated from the marine sponge Acanthella spp., provides a good case in point (Lawrance et al. 2014). An increase (3.0 mg/L) in surfactin production was achieved using a newly optimized fermentation medium, particularly employing 2.5% of crude oil. However, a threefold rise (11.78 g/L) was accomplished through heterologous expression of the biosurfactant-encoding genes sfp, sfpO, and srfA in an Escherichia coli host. By using a myriad of hydrophobic substrates (kerosene, olive oil, vegetable oil, crude oil, and waste engine oil), the recombinant surfactin displayed emulsification index values comparable to those extracted from the wild-type sponge-derived Bacillus and were as proficient as that of the non-biogenic surfactants (Lawrance et al. 2014). The use of a heterologous host strain significantly leveraged up the concentration of the final expressed biosurfactant substance and can constitute an alternative approach to optimization studies with wild-type sponge-derived bacteria. In fact, this is applicable since other bacterial species can express selected target biosurfactant biosynthetic genes. Despite these advantages, the heterologous expression approach has some constraints, including the need for previous knowledge of surfactant-encoding genetic repertoire and the choice of the appropriate host system. This last one is particularly challenging when it is considered the production of polymeric biosurfactants by the host cell, which should be efficiently released by the host cell.
Still under the realm of the environmental applications, Dhasayan et al. (2015) suggested that a lipopeptide biosurfactant produced by Bacillus amyloliquefaciens MB-101 could be potentially used for bioremediation processes in marine environments. For that claim, they relied on the high oil displacement ability (11 ± 0.2 mm) and low surface tension (28 ± 0.6 mN/ m) displayed by the sponge-derived Bacillus, especially when compared with other bacteria isolated from the same sponge specimen (C. diffusa). Amongst these isolates, MB-30 had a high cell hydrophobicity (79.2%), a result that glimpsed the potential use of this spongederived bacteria for enhancing biodegradation of hydrocarbon in oil-contaminated areas (Dhasayan et al. 2015). In another study by the same research group, Halomonas MB-30 was confirmed as a glycolipid producer. A remarkable oil displacement ability together with a considerable surface tension reduction (30 mN/m) in kerosene and diesel-supplemented medium was observed. This was accompanied by the maximum glycolipid production following crude oil addition to the culture media and notable stability of the surfactant at temperatures superior to 100 C. The authors suggested that Halomonas sp. MB-30 could be turned into an interesting microbial resource for enhanced oil recovery supported by in situ biosurfactant and bioremediation of oil-derived hydrocarbons (Dhasayan et al. 2014).
Although most of the reports target hydrocarbon bioremediation, biosurfactants are capable of being applied in heavy metals bioremediation. Many reports associate biosurfactant-producing bacteria with their potential in treating sites contaminated with several heavy metals. Basically, biosurfactant molecules are capable to form different interactions with these metals, removing them and enabling a quick and efficient reduction of their toxic concentrations from the environment (Akbari et al. 2018).
Previous reports by our research group (Santos-Gandelman et al. 2014a, 2014b showed the potential application for the mercury bioremediation by Bacillus cereus Pj1, a strain isolated from Polymastia janeirensis sampled in the Brazilian Southeast coast (Santos-Gandelman et al. 2014a). B. cereus Pj1 was resistant to HgCl 2 (mercury chloride), MeHg (methylmercury), CdCl 2 (cadmium chloride), and Pb(NO 3 ) 2 (lead nitrate). Notably, the strain Pj1 was able to reduce the toxic mercuric cation (Hg 2þ ) to non-toxic elemental mercury (Hg 0 ). Surfactant production was confirmed for B. cereus Pj1 given the positive results in the drop-collapsing test, and the emulsifying activity around 65.8 ± 1.32% in n-hexadecane, which was superior to two tested synthetic surfactants, Triton X-100 (60.4 ± 0.73%) and Tween 80 (62.6 ± 1.43%). Finally, the biosurfactants likely produced by B. cereus Pj1 were suggested as one of the main drivers for the successful mercury remediation potential of this sponge-derived bacterium (Santos-Gandelman et al. 2014a).
A representative application in heavy metal remediation was the report on a lipopeptide analogue isolated and completely chemically characterized from Bacillus sp. MSI 54, a strain associated with the sponge Agelas clathrodes. The substance removed heavy metals from the surface of several vegetables (cabbage, carrot, and lettuce) with rates of 75.5% Hg, 97.73% lead (Pb), 89.5% manganese (Mn), and 99.93% cadmium (Cd). Notably, the MSI 54 lipopeptide also showed an outstanding thermostability over a temperature range from 4 to 121 C, with a 79% emulsification index at 121 C (Ravindran et al. 2020). Such example of potential use in the food industry illustrates that these sponge microbial surfactants can be useful in several application fields outside the widely-studied bioremediation umbrella.

Topic 3: other applications
Despite that the efforts have been concentrated on bioremediation and environmental protection, the uses of microbial surfactants encompass a whole plethora of fields, especially in biomedicine and in the pharmaceutical industry (Giri et al. 2020). Given their membranedisrupting and disassembling properties, biosurfactants can be potentially applied as antimicrobial and antibiofilm agents. Therefore, they can be powerful weapons in the control and treatment of clinically concerning microbial pathogens and in solving fouling issues in aquaculture and the maritime industry (Banat et al. 2014;Naughton et al. 2019). In that regard, biosurfactant molecules reported from the sponge microbiome have been also investigated for such biological activities, in particular, the inhibition of formation and disruption of microbial biofilms. Kiran et al. (2010a) were the pioneers in assessing the antibiofilm potentiality from surface-active agents from sponge-dwelling microorganisms. They isolated a biosurfactant-producing strain, Brevibacterium casei strain MSA19, from the sponge D. nigra. In addition to its broad-spectrum antimicrobial activity and bacteriostatic action, the purified glycolipid was able to interfere in the biofilm-forming ability of mixed and individual bacterial strains, including clinical isolates, human and fish pathogens and marine isolates (Kiran et al. 2010a). In a following report from the same research group , Nocardiopsis sp. MSA13A, an actinobacterium strain also recovered from D. nigra, had the biosurfactant production significantly increased following the addition of iron nanoparticles (FeNPs). The glycolipid had an EI 24 of 35% upon the supplementation of treated molasses in the fermentation medium and completely disrupted the pre-formed biofilm of Vibrio alginolyticus, a common pathogen of marine animals and humans, even at lower biosurfactant concentrations (<300 mg/mL) . Lipopeptides from the other three sponge-derived actinobacterial strains were inhibited by 80% of the biofilm formation and reduced by 70% the cell viability in the biofilm of Vibrio harveyi, another concerning pathogen (Selvin et al. 2016). These reports illustrate well the potential of these biosurfactants from cultivable sponge-associated bacteria for the control of biofouling issues from mariculture to hospital settings.
In line with the abovementioned cases, the antifouling applications have been supported in a recent report. After confirming the antibacterial, antiadhesive, and non-toxic nature of biosurfactants released by a Mycale ramulosa-derived Bacillus, Alem an- Vega et al. (2020) assayed the strain's ability to control natural biofouling in a field assay under controlled conditions. They tested the crude biosurfactant supernatant of this Bacillus strain on formulated paints, which are commonly applied to reduce the deleterious effects of cumulative fouling. The surfactants produced by this strain were as effective as the specific antifouling paint in mitigating the initial microbial settlement and the settlement of both barnacles and ascidia. Despite this evident high value-added potential as antifoulings for the maritime industry and bivalve cultures, further optimization of production and chemical characterization would still be required for these sponge-associated Bacillus' biosurfactants (Alem an- Vega et al. 2020). Both these issues were addressed in another recently published work. Potentially novel surfactin isomers and lipoaminoacids derivatives were discovered in the metabolome of Pantoea cf. eucrina D2 (Vitale et al. 2020). This strain was isolated from Chondrosia reniformis, a sponge found in volcanic acidified soil in South Italy. Indeed, this wide range of surfactants produced by this sponge-derived Pantoea exhibited antimicrobial activity against Staphylococcus spp. and Listeria monocytogenes.
In the food industry, biosurfactants are valued because of their antimicrobial, antiadhesive, antibiofilm, and antioxidants properties. Furthermore, they are particularly attractive due to the stabilization of fat/oil emulsions, improvement of dough stability, texture, volume, conservation of bakery products and as enhancers of creaminess and viscosity (Nitschke and Silva 2018;Ribeiro et al. 2020). In this perspective, a lipopeptide surfactant produced by Nesterenkonia sp. MSA31, an actinobacterium initially isolated from Fasciospongia cavernosa, was studied with a focus on their direct use as food additives (Kiran et al. 2017). The lipopeptide was remarkably stable over a broad temperature range (4-120 C) and retained an EI 24 of 85% over a pH interval from 6.0 to 9.0 and at 10% NaCl, physicochemical features of special interest for food and cosmetic industries. In a well-designed experimental design, the authors incorporated the lipopeptide to muffin preparations and observed a reduction in hardness, chewiness and gumminess and a higher level of springiness and increased cohesiveness. Moreover, the muffins supplemented with the lipopeptide displayed a light-yellow colour, which was strikingly similar to the positive control. Thus, the obtained results were highly favourable for the use of this halo-alkali and thermal tolerant biosurfactant as a food additive (Kiran et al. 2017).
After the selective isolation of copper-reducing sponge-associated bacteria, Rozas et al. (2017), tested the bioleaching ability of the strains over copper wires (CuW) in a sweater-based medium. One isolate, Bacillus sp. Hyhel-1, leached copper efficiently by a precipitation mechanism, visible by the initial deposition of the metal over the bacterial surface followed by the formation of intracellular copper nanoparticles and vesicles. It was later confirmed that an iturin-like surfactant was the key mediator in the copper leaching capacity by Hyhel-1. Further bioleaching assays using grounded printer motherboards (e-waste) confirmed that the iturin-like molecule likely enhanced the bacterial bioleaching by trapping copper ions in solution while the sponge-derived Bacillus colonized the e-waste surface (Rozas et al. 2017). The copper nanoparticles formed in the bioleaching process were further characterized in a subsequent work by the same research group (Rozas et al. 2019). Stable and compact supramolecular structures were naturally formed in solution due to the chemical nature of the iturin lipopeptide produced by Bacillus sp. Hyhel-1. In addition to the recovery of copper in an alternative and greener biosorption process, these copper lipopeptide crystals could be the basis for further production of copper cyanide (CuCN) blocks. These blocks could be then investigated as templates for the manufacturing of superconductor and computer devices. These works showcase the wideness of applications for sponge microbial surfactants, which therefore deserves proper research attention (Rozas et al. 2019).

Future avenues and concluding remarks
Biosurfactants emerge as exemplary molecules in the current global demand for greener and eco-friendly alternatives to petrochemical-based surfactants. In the case of microbial surfactants, rapid biodegradability, bioavailability, biocompatibility, generally low toxicity, high selectivity and specificity, and outstanding stability under extreme conditions, and long shelf-life turn them ideal not merely for bioremediation processes. Food, detergent, and pharmaceutical industries are also particularly keen on such high-value-added and innovative products (Jimoh and Lin 2019;Adetunji and Olaniran 2021). This is well-illustrated by the expected growth at 5.3% CAGR (compound annual growth rate) for the biosurfactant market between 2020 and 2027, with a value projection of USD 6.5 billion by 2027 (GBM 2021). It has also been verified that a number of patents of microbial biosurfactants and bioemulsifiers over 250 patents were issued worldwide, encompassing a myriad of application fields, such as petroleum industry, cosmetics, antimicrobial, and medical applications, and in uses related to bioremediation processes (Randhawa and Rahman 2014;de Almeida et al. 2016).
In the present review, the sponge microbiome was uncovered as a prospective source of biosurfactant compounds. However, the amount of current work on this biotechnological potentiality in this microbial symbiotic consortium remains hitherto incipient. To foster future exploitation of sponge-associated microbial assemblages for potentially novel biosurfactants, we decided to emphasize the following key approaches below and summarized in Figure 3.

Selective isolation of biosurfactant-producing microorganisms
Having cultivable sponge-derived microorganisms in hand is crucial to fully harness the biotechnological potential of the sponge microbiome (Laport 2017). Redirecting the discovery of surface-active substances from the sponge microbiota at the earliest possible stage, the microbial isolation from the sponge specimens, will be fruitful in finding potential producers. This can simply be achieved by incorporating hydrocarbon-based compounds (e.g. crude oil, kerosene, gasoline, n-hexadecane, or a mixture of them) as the sole carbon sources to minimal mineral media. These enrichment cultures will increase the odds of isolating oildegrading microorganisms which are likely surfactant producers (Walter et al. 2010;Karlapudi et al. 2018).
Up to now, there is just a single report on the application of selective isolation for the straightforward recovery of biosurfactant-producing microorganisms from sponge samples. In addition to classical isolation, a pre-treatment with enrichment cultures composed of the Bushnell Haas medium added with 1.0% crude oil were applied for the bacterial isolation from sponges sampled in a brackish lake (Rizzo et al. 2018). Pilot studies committed to the formulation of selective mineral basal media supplemented with various hydrocarbon substrates are therefore required to unveil the potential of this approach. Specimens from different sponge taxonomic groups should be tested in this preliminary assessment for further validation of this isolation strategy. If the sponge specimens are originally sampled from oil-polluted environments, selective isolation will probably be rewarding in the recovery of promising biosurfactant-producing microorganisms. Independently, representative hydrocarbonoclastic bacterial groups have been shown to inhabit a plethora of undisturbed and non-polluted marine habitats (Gutierrez 2019). Potential hydrocarbon-degrading and biosurfactant-releasing bacteria are certainly amongst the sponge inhabitants, considering the complex and widely diverse profile of the sponge microbiome (Moitinho-Silva et al. 2017). Thus, selective isolation complemented by cutting-edge culturomics approaches will aid in finding these valuable sponge microbial resources for the bioremediation field.

Mining -omics datasets for the biosurfactant biosynthetic repertoire
Sponge microbiology has been completely revolutionized since the ascension of culture-independent methods in the early 2000s. From the initial and conventional 16S rRNA-based surveys to the recent widespread application of meta(-omics) approaches, the knowledge on the essential functions played by sponge-associated microbial assemblages has been expanded in a way that also allowed their exploitation in the biotechnological arena (Kiran et al. 2018;Liu and Li 2019). With the exponential growth of genomic data from sponge-derived microorganisms in the last few years, the direct in silico assessment of their biosynthetic repertoire has been significantly facilitated, resulting in successful examples, including potentially novel antimicrobials (well-illustrated by the recent report of Rust et al. 2020)   . Key approaches for the successful uncovering of novel sponge microbial biosurfactants. From sampling and microbial isolation, some approaches can be taken to minimize efforts, time save and increase the odds of discovering new surface-active substances from the sponge microbiota: (I) selective isolation of biosurfactant-producing microorganismsincorporating hydrocarbons as sole carbon sources to minimal mineral media; (II) mining -omics datasets for the biosurfactant biosynthetic repertoire targeted-mining for biosurfactants in the sponge microbial (meta)genomic data; (III) construction of micro-/mesocosmsthe design of a constantly monitored mesocosms system to mimic environmental factors close to natural conditions; finally, (IV) scaling-up the production processoptimization of the biosurfactant production at the downstream or upstream stages to reach an industrial scale. et al. 2016). This exemplifies how far we are from comprehending the molecular basis underlying biosurfactant production in this symbiotic consortium.
Several biosurfactant molecules are members of the microbial secondary metabolism and, consequently, the biosurfactant genetic repertoire detection can be easily achieved with the same set of computational tools generally applied for the virtual screening of secondary metabolites biosynthetic gene clusters (smBGCs). From these, the most well-known and extensively used is antiSMASH (Medema et al. 2011), currently in its sixth version (Blin et al. 2021). It classifies several of these gene clusters into chemical classes in which normally several of these surfactant molecules are usually classified, such as non-ribosomal synthesized peptides (e.g. lipopeptides). Thus, we can extend the use of these computational pipelines and additional in silico structural analyses for the direct identification of surfaceactive agents in sponge microbial genomes instead of simply assigning them as putatively encoded metabolites with antimicrobial and other pharmacologically relevant activities. In this framework, targeted-mining for biosurfactants in the sponge microbial (meta)genomic data may provide uncharacterized surfactant hits, with unique chemistry and, consequently, a likely diverse array of interesting bioactivities. This is also going to help in accelerating and improving the costefficiency of the whole biodiscovery process, which is advantageous for their introduction in the industrial and market sectors (Tripathi et al. 2018).
To some extent, the report by Selvin et al. (2016) is the only one in the current review in which the genomic data from sponge microbial symbionts was explored envisioning production and bioactivity of biosurfactants. The authors correlated the biosynthetic machinery of the produced biosurfactants and their respective biological activities through the identification of biosurfactant gene clusters in five different sponge-derived actinobacterial strains, complemented by phylogenetic analyses and reconstruction of the 3D key biosynthetic enzyme domains. Following the bioinformatic analyses, the predicted chemical nature of the potential surfactants was further confirmed by mass spectrometry. In particular, lipopeptides from two actinobacterial strains demonstrated high antibiofilm activity against V. harveyi (Selvin et al. 2016).
The previous knowledge on this biosurfactant genetic repertoire can be therefore valuable in: (i) guiding the heterologous expression of the biosurfactant molecules, such as carried out for the surfactin genes from sponge-associated B. licheniformis strains (Lawrance et al. 2014;Anburajan et al. 2015); (ii) defining assays for assessing the most likely activities of the surfactants given their predicted chemical nature; (iii) and allowing the application of genetic-based techniques (e.g. sitedirected mutagenesis) for molecular modification of the surfactant structure and overall improvement or adaptation of the aimed biological activity. In this sense, it should be also considered what has been recently discussed on the importance of -omics-based strategies to boost up the discovery of biosurfactants with potentially novel chemical nature and bioactivities (Trindade et al. 2021;Gaur et al. 2022). An integrated bioinformatic-experimental platform will constitute a powerful and time-saving path for the successful uncovering of novel sponge microbial biosurfactants.
In particular, functional-based metagenomics screening approaches have been well-established for the search of new biosurfactant molecules in a highthroughput fashion (Williams and Trindade 2017). Notably, the use of functional metagenomics for the isolation of biosurfactants from the marine microbiome has already been praised (Kennedy et al. 2011;Jackson et al. 2015;Gudiña et al. 2016), with the adaptation of diverse screening regimes with a higher number of environmental samples. The main advantage of this strategy is that there is no previous knowledge of the genetic repertoire associated with the surfactant production and, consequently, it increases the odds of finding new genes likely to be associated with chemically novel surfactants (Kennedy et al. 2011;Jackson et al. 2015). Once these surfactant-encoding genes have been revealed by the -omics mining strategy from the sponge microbial metagenomic clones, these functional-based phenotypic assays could be readily adapted for further expression, recovery, quantification, and characterization of the biosurfactants in a timely, straightforward and cost-effective manner, such as described in the recent literature (Kubicki et al. 2020).

Construction of micro-/mesocosms and evaluation of their performance
In bioremediation trials with biosurfactant molecules, in vitro experiments have been indicated to produce unrealistic data that cannot be exported to natural environments (Hassanshahian et al. 2014). Additionally, these results are not easy to interpret due to the complexity of the analysed biological processes (Cappello and Yakimov 2010) and that the most relevant experiments are carried out in situ (Reilly 1999;Hassanshahian et al. 2014). Moreover, these surfactants must be at a higher concentration to compete on an industrial level and stable to be considered as a good counterpart to synthetic surfactants. For these reasons, the detection of ecologically efficient production methodologies, which performance is attested in situ, is highly envisaged for these surfactants of biological origin (Rizzo et al. 2018).
Simulating the natural conditions of the original ecosystem may provide another avenue for the inducement of biosurfactant production from sponge-derived microorganisms. In a first stage, the capacity of in vitro production and interactions of the selected producer strains can be performed in microcosm experiments. This can be further scaled up to mimic the original habitat under controlled conditions in well-designed and constantly monitored mesocosms. Both systems have been long-termed used in studying the response of microbial communities to oil spills and in artificial bioremediation set-ups in aquatic environments (Snape et al. 2008;Cappello and Yakimov 2010;Genovese et al. 2014). Thus, a mixed consortium with potential sponge microbial strains with promising biosurfactant-producing potential might be designed especially for environmental applications.
Importantly, a number of factors for the implementation of these experimental ecosystems with these sponge-derived microbial isolates must be considered. First of all, the scope of the study should be clearly oriented for the application of these surfactants as enhancers or facilitators of bioremediation processes. Second, potential antagonistic interactions between the selected strains to compose the micro-/mesocosm community must be assessed prior to experiment design. Growth inhibition and competition have been already demonstrated between sponge bacterial isolates from the same and different sponge species, which is expected considering the complex nature of these associated microbial communities (Laport et al. 2016;Esteves et al. 2017). Third, the induced effects of the released biosurfactants by the sponge microbial artificial consortia in the naturally dwelling marine microbial communities should be investigated. Either 16S rRNA gene-based diversity analysis or metagenomic approaches can be performed to address this point. From a sustainability point-of-view, this assessment is fundamental since it will reveal the suitability of broader use of these surfactants in a real-life scenario of oil spill or contamination (Lawniczak et al. 2013;Olasanmi and Thring 2018;Nikolova et al. 2021aNikolova et al. , 2021b.

Scaling-up the production process
A major bottleneck for the wider replacement of chemically synthesized surfactants by biosurfactants is still their higher production costs. Especially, when working with wild-type microbial producers, the biosurfactant yields are not sufficient for the direct commercial usage of these molecules in several application fields. Therefore, optimization of biosurfactant production at the downstream or upstream stages is mandatory to attend to the industrial demands and enable their faster introduction in the market (Mukherjee et al. 2006;de Kronemberger 2014).
To manufacture a quantity of these substances on an industrial scale, it is necessary to maintain the type, quantity, quality of the product and optimize the process conditions. This is required primarily because biosurfactant manufacturing is directly affected by different variables, including oxygen availability, dilution rate, temperature, pH, nature of carbon and nitrogen sources and concentration of metal ions (Desai and Banat 1997;GBM 2021). To reduce this problem, researchers have been incorporating industrial waste as low-cost substrates in biosurfactant production, such as corn steep liquor (Gudiña et al. 2015), residual soy oil (Camargo et al. 2018), molasses (Saimmai et al. 2011), amongst other substrates and cultivation conditions .
The advantage of controlled cultivation conditions is the significant yield increase in biosurfactant production. Concerning the yields obtained from the biosurfactant production on a laboratory scale, optimization experiments have been already carried out with sponge-associated bacteria and these examples are summarized in Table 1. Hence, their sustained adoption can render higher concentration of biosurfactants from the sponge microbiome in the forthcoming future. It is crucial to highlight that the optimization assays can be harder to implement in micro-/mesocosm conditions, as abovementioned. If the conditions evaluated in the mesocosm experiments are efficient in getting detectable yields of the surfactant molecules, it should be further tried to increase their production considering the complex nature of interactions of the producer strain with the tested microenvironment. This gains a whole new level of complexity if sponge microbial consortia (which can be mixed with strains deriving from other environments) is applied in the mesocosm trials and the interactions between these strains and their impact on the biosurfactant production and release should also be addressed. For instance, metabolic engineering strategies (Andreolli et al. 2021) could be regarded to gain greater insights in scaling-up biosurfactant production under the micro/mesocosms conditions depending on the aimed application.
The sponge microbiome comprehends a prominent biomolecular reservoir for biotechnological exploitation.
An ancient coevolution coupled with a remarkable adaptation to ever-changing habitat conditions has allowed these symbiotic microorganisms to produce substances with envisaged properties by several modern industries. Even though the biotechnological harnessing of this host-associated microbiome is booming, biosurfactants from the sponge microbiome have received little attention in comparison with other biologically active molecules. Here, we overviewed the current state-of-art biosurfactant from sponge-associated microorganisms, showing its promising potential in environmental, biomedical and other industrially relevant sectors. The most advocated uses of sponge microbial surfactants are in bioremediation processes and as alternative antimicrobial and antibiofilm agents. Novelty in the chemical nature of the produced surfactant along with higher yields in optimization assays exemplified how the poriferan microbiome is a well-gifted source for these surface-active molecules. Future research should fill the existent gaps in the screening and characterization of these biosurfactants molecules to be unravelled in this role model holobiont and will probably renew the efforts in their biodiscovery.

Disclosure statement
The authors declare that they have no conflict of interest.

Funding
This work was supported by Conselho Nacional de Desenvolvimento Cient ıfico e Tecnol ogico (CNPq) under SSC: solid-state culture; C: temperature; NI: no information. a Quantity contained in the basal artificial seawater (ASW) medium. b Media with distinct pH values were developed using appropriate phosphate buffer instead of water. c According to the authors, in the preliminary experiments, biosurfactant production was performed in submerged culture, but the production was not increased significantly, as soon the optimization was performed in SSC. d Purified surfactant concentration in the extract.