*Article* **Biofilm Inhibition and Antimicrobial Properties of Silver-Ion-Exchanged Zeolite A against** *Vibrio* **spp. Marine Pathogens**

**Zarina Amin 1, Nur Ariffah Waly <sup>2</sup> and Sazmal Effendi Arshad 2,\***


#### **Featured Application: This study highlights a potential application of ion exchanged zeolite A against marine microbial pathogens and their biofilms.**

**Abstract:** A challenging problem in the aquaculture industry is bacterial disease outbreaks, which result in the global reduction in fish supply and foodborne outbreaks. Biofilms in marine pathogens protect against antimicrobial treatment and host immune defense. Zeolites are minerals of volcanic origin made from crystalline aluminosilicates, which are useful in agriculture and in environmental management. In this study, silver-ion-exchanged zeolite A of four concentrations; 0.25 M (AgZ1), 0.50 M (AgZ2), 1.00 M (AgZ3) and 1.50 M (AgZ4) were investigated for biofilm inhibition and antimicrobial properties against two predominant marine pathogens, *V. campbelli* and *V. parahemolyticus*, by employing the minimum inhibitory concentration (MIC) and crystal violet biofilm quantification assays as well as scanning electron microscopy. In the first instance, all zeolite samples AgZ1–AgZ4 showed antimicrobial activity for both pathogens. For *V. campbellii*, AgZ4 exhibited the highest MIC at 125.00 μg/mL, while for *V. parahaemolyticus*, the highest MIC was observed for AgZ3 at 62.50 μg/mL. At sublethal concentration, biofilm inhibition of *V. campbelli* and *V. parahemolyticus* by AgZ4 was observed at 60.2 and 77.3% inhibition, respectively. Scanning electron microscopy exhibited profound structural alteration of the biofilm matrix by AgZ4. This is the first known study that highlights the potential application of ion-exchanged zeolite A against marine pathogens and their biofilms.

**Keywords:** biofilms; zeolite A; *Vibrio* spp.; antimicrobials

3

#### **1. Introduction**

Aquaculture farming has been the fastest growing food producing sector in the last few decades and an important industry in many developing countries. However, the industry currently faces a threatening challenge due to the bacterial disease outbreaks resulting in high mortality rates in the aquaculture population [1,2]. This is in part due to extensive use of antibiotics in fish farms leading to antimicrobial resistance in fish pathogens [3–5]. Vibriosis is an important bacterial disease in wild and farmed marine fishes, which results in severe economic loss of more than USD 1 billion [6]. In addition, bacteria from seafood sources have been associated with foodborne outbreaks [7].

The formation of biofilms by marine pathogens further amplifies problems faced in the aquatic systems and the aquaculture industry. Biofilms are self-assembled communities of bacteria embedded in a self-developed extracellular matrix (ECM) and are adherent to abiotic or abiotic surfaces. The ECM contains exopolysaccharides, proteins and extracellular DNA (eDNA) [8–10]. Biofilm formation is induced by genetic factors and influenced by environmental conditions including pH, temperature and the availability of nutrients [9].

**Citation:** Amin, Z.; Waly, N.A.; Arshad, S.E. Biofilm Inhibition and Antimicrobial Properties of Silver-Ion-Exchanged Zeolite A against *Vibrio* spp. Marine Pathogens. *Appl. Sci.* **2021**, *11*, 5496. https:// doi.org/10.3390/app11125496

Academic Editors: Anna Poli and Valeria Prigione

Received: 3 April 2021 Accepted: 21 May 2021 Published: 14 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

The formation of ECM in biofilms presents a physical barrier that allows the bacteria to protect themselves from external disturbances. This further leads to the improvement of their resilience to the external environmental conditions and enhance their virulence to cause disease. Examples of enhanced resilience include resistance against antimicrobial agents, disinfectants and host defense mechanisms, making them more difficult and expensive to treat [11,12].

The extensive use of antibiotics in marine systems coupled with the enhanced resilience of marine pathogens as biofilms leads to a "double whammy" scenario of antimicrobial resistance in the aquaculture industry, resulting in huge economic losses due to high mortality rates of fish and seafood as important commodities in countries with high economic dependency on seafood farming such as South East Asia and Japan. The prevalent challenge has necessitated efforts on new antibacterial materials that can effectively inhibit growth and resistance of aquatic pathogens while minimizing negative impacts on human and animal health and the environment.

*V. parahemolyticus* is a prevalent food-poisoning bacterium associated with seafood consumption, typically causing self-limiting gastroenteritis and commonly found in temperate and tropical marine and coastal waters globally [13,14]. Vibriosis is a systemic bacterial infection in farmed and wild marine fishes, which is considered to be a profoundly significant problem due to intensive economic losses in aquaculture industry worldwide [15]. *V. campbelli* is as an emerging marine pathogen recently associated with diseased farm shrimps [16].

Several studies have reported on the development of bacterial biofilms in aquatic environments, particularly challenges faced due to the persistence of these biofilms as important sources of infection and disease in the aquaculture industry. The diseases result in huge economic losses due to high mortality rates of seafood as important commodities [17–19]. Several strategies have been evaluated to combat biofilm development in aquaculture systems, and these include seawater ozonation [20], use of probiotics [21] and sanitizers [22]. In a recent minireview by Artunes et al., the strategy of quorum sensing (QS) inhibition was discussed. QS refers to chemical signaling molecules that control biofilm formation in bacteria [23]. Recently, several studies on antibacterial agents for the aquaculture systems have also been explored. These include non-chemotherapeutic methods such as natural therapeutics from plants and immunostimulants [24–26] and by alternative inorganic materials [27]. Additionally, studies on the application of zeolites as an antimicrobial agent have started to gain traction particularly in the field of aquaculture.

Zeolite is a microporous, crystalline aluminosilicate with a framework that is made up of [SiO4] <sup>4</sup><sup>−</sup> and [AlO4] <sup>5</sup><sup>−</sup> tetrahedra. The tetrahedra are joined together by sharing their corner oxygen atom [28]. In this framework, positively charged silicon ion was balanced by oxygen ion. While the negative charged alumina remains unbalanced resulting in a negative charge of the total structure of zeolite, and it is balanced by extra framework cations [29]. These cations can be exchanged with any other positive ions in order to modify their properties to match with the desired application.

Although there has been extensive research on the antibacterial activity of ion-loaded zeolite against common bacteria including *Escherichia coli, Bacillus subtilis, Staphylococcus aureus, Klebsiella pneumoniae, Pseudomonas aeruginosa* and wastewater biofilm bacteria [30–33], to our knowledge, however, there is still relatively a small number of studies of the effectivity of ion-loaded zeolite A against marine pathogens, in particular *Vibrio campbellii* and *Vibrio parahaemolyticus*.

This study aimed to investigate antimicrobial and antibiofilm applications of silver-ionexchanged zeolite A against marine bacterial pathogens *V. parahaemolyticus* and *V. campbelli*.

#### **2. Materials and Methods**

#### *2.1. Liquid Ion Exchange of Zeolite A*

The initial synthesis of zeolite A from bentonite clay comprised the activation of precursor bentonite clay by thermochemical treatments in HCl, addition of alkaline activators, ageing and crystallization processes as well as characterization by X-ray diffraction (XRD) and scanning electron microscopy (SEM) analysis (data published elsewhere).

For liquid ion exchange, the integration of silver (Ag) ion into zeolite A was performed by using the liquid ion exchange method as described by Demirci et al. [31], with slight modifications. Briefly, 1 g zeolite A was added into 10 mL of four AgNO3 concentrations: 0.25, 0.50, 1.00 and 1.50 M (denoted AgZ1, AgZ2, AgZ3 and AgZ4, respectively) and mixed at 150 rpm for 3 days. The obtained zeolite A were then vacuum filtered, washed with deionized water and dried at 80 ◦C overnight.

For the metal ion concentration analysis, a Perkin-Elmer Optima 5300DV (Waltham, MA, USA) axial viewing ICP-OES was used. The analytical wavelengths of elements included were: Na 589.592, Ag 328.068 and Cu 327.393 nm. The software used was WinLab32 for ICP. Briefly, 200 mg of AgZ1, AgZ2, AgZ3 and AgZ4 was added into separate tubes containing 14 mL of acid mixture, aqua regia (40% HNO3 + 60% HCl) and left overnight at room temperature. Four milliliters of aqua regia solution was then added into the tubes and mixed for 30 min at 80 ◦C before being filtered and diluted prior the ICP-OES analysis.

#### *2.2. Minimum Inhibition Concentration (MIC) Assay of Ag-Exchanged Zeolite A*

The minimal inhibitory concentration (MIC) assay of Ag-exchanged zeolite A against *V. parahaemolyticus* and *V. campbellii* was carried out as described [33]. Briefly, frozen stocks of *V. campbellii* and *V. parahaemolyticus* were grown on nutrient agar (NA) supplemented with 2% of NaCl and incubated for 16 h at 37 ◦C. Cultures were then inoculated into tryptone soy broth (TSB), and absorbances were adjusted accordingly to an initial starting OD600 of 0.05. As seen in Table 1, samples AgZ1, AgZ2, AgZ3 and AgZ4 of respective AgZ concentrations were then added. The assay for each sample was tested with two technical replicates.

**Table 1.** Silver ion content (mg/g) of zeolite of samples.


For each bacterial species, doubling dilutions up to 1/512 of AgZ1, AgZ2, AgZ3 and AgZ4 were firstly carried out. Five hundred microliters of liquid inoculum (OD600 adjusted to 0.05) was then added and incubated in a shaking 37 ◦C incubator for 16 h. One hundred microliters of the culture was then plated and spread on to nutrient agar (NA) plates supplemented with 2% NaCl and further incubated for 16 h at 37 ◦C with appropriate controls. The minimum inhibitory concentration (MIC) of AgZ1, AgZ2, AgZ3 and AgZ4 was determined by the lowest concentration of samples that did not exhibit bacterial growth.

#### *2.3. Quantitation of Bacterial Biofilm Growth by Crystal Violet Assay*

The biofilm growth of AgZ-treated biofilm were quantitated by the crystal violet assay using the methods described by O'Toole [34] with modifications. Briefly, cultures of *V. campbellii* and *V. parahaemolyticus* were incubated for 16 h at 37 ◦C in tryptone soy broth (TSB) supplemented with 2% NaCl.

For the biofilm assay, the sublethal concentration obtained from the MIC assay from was selected as the starting assay concentration. Briefly, serial two-fold dilutions of the sample were performed in a 96-well microtiter plate (corning) containing tryptone soy broth supplemented with 2% NaCl. Fifty microliters of the overnight cultures (with OD600 adjusted to 0.05) were then inoculated into each well, and it was then incubated at 37 ◦C for 72 h. Following incubation, wells were washed three times with distilled H2O, desiccated at 50 ◦C, stained with 1% crystal violet for 30 min and added with 95% ethanol. The absorbance of solubilized dye was then determined at 570 nm (Shimadzu, Spectramax, Kyoto, Japan).

#### *2.4. Scanning Electron Microscopy (SEM) of V. campbellii and V. parahaemolyticus Biofilms*

The cell morphologies of *V. campbellii* and *V. parahaemolyticus* biofilms grown in TSB media and in TSB with AgZ4 were analyzed by SEM. Briefly, 5 mL overnight liquid cultures of *V. campbellii* and *V. parahaemolyticus* were cell fixated according to protocol by Gomes and Mergulhão [35] with modifications. The biofilm samples were then fixed in 5% glutaraldehyde prepared in 0.1 M PBS pH 7.2 at 4 ◦C for 12 h. Following the fixation process, the samples were dehydrated by introduction into a series of ethanol solution of varying concentration gradients (35, 50, 75, 95 and 2 × 100%). The dehydrated samples were then immersed in HMDS for 10 min. Upon dehydration, the samples were dried overnight and then sputter-coated with platinum before being analyzed by SEM (Hitachi SEM, S 3400N, Tokyo, Japan).

#### **3. Results**

The incorporation of silver ion (Ag) into the zeolite A framework (AgZ) was determined by introducing the zeolite sample to four different concentrations of Ag solution AgZ1, AgZ2, AgZ3 and AgZ4. During this process, the potassium and sodium ions that exist in the zeolite framework were exchanged by metal ions. ICP-OES analysis of AgZ1, AgZ2, AgZ3 and AgZ4 is as shown in Table 1.

As seen in Table 1, the highest silver ion content was observed for AgZ4 at 190.511 ± 1.989 mg/g, followed by the lowest concentration for AgZ1 at 80.949 ± 0.389 mg/g.

#### *3.1. Minimum Inhibitory Concentration (MIC) of V. campbellii and V. parahaemolyticus in AgZ*

As seen in Table 2, both *V. campbellii* and *V. parahaemolyticus* showed susceptibility to all four AgZ1-AgZ4 concentrations. For *V. campbellii*, the highest MIC was observed in AgZ4 at 0.1250 mg/mL. For *V. parahaemolyticus*, the highest MIC was observed in AgZ3 at 0.0625 mg/mL. The lowest MIC for both *V. campbellii* and *V. parahaemolyticus* was observed in AgZ1 at 2.0 and 0.5 mg/mL, respectively.


**Table 2.** Minimum inhibitory concentration (MIC) of silver-ion-loaded zeolite A against *V. campbellii* and *V. parahaemolyticus*.

#### *3.2. Quantitation of Bacterial Biofilm Density by Crystal Violet Assay*

The AgZ4 MIC concentration for both pathogens were further selected for biofilm inhibition crystal violet assay. Figure 1 demonstrated biofilm inhibition percentages of ion-exchanged zeolite in comparison with the untreated control. As shown in Figure 1, at 570 nm absorbance, biofilm formation of AgZ4 against *V. campbellii* and *V. parahaemolyticus* exhibited up to 60.2 and 77.3% inhibition, respectively.

**Figure 1.** Biofilm inhibition of *V. campbellii* and *V. parahaemolyticus* by 1.50 M AgZ4 assessed by CV staining. Biofilm inhibition of *V. campbellii* and *V. parahaemolyticus* isolates after 24 h growth with 1.50 M AgZ4 was assayed by CV staining (A570).

> Scanning Electron Microscopy of *V. campbellii* and *V. parahaemolyticus* in AgZ4, Figures 2a and 3a represent the SEM images of untreated samples of *V. campbellii* and *V. parahaemolyticus*, respectively, while Figures 2b and 3b represent the SEM images of *V. campbellii* and *V. parahaemolyticus* after growth in media culture treated with 1.50 M AgZ4.

**Figure 2.** SEM analysis showing (**a**) untreated *V. campbellii* biofilm bacteria embedded in extracellular matrix biofilm and (**b**) *V. campbellii* treated with AgZ4 where profound loss of biofilm matrix is seen.

**Figure 3.** SEM analysis showing (**a**) untreated *V. parahaemolyticus* biofilm bacteria embedded in extracellular matrix biofilm and (**b**) *V. parahaemolyticus* treated with AgZ4 where profound loss of biofilm matrix is seen.

Figure 2a shows clustering and aggregation of *V. campbellii* covered with a network of extracellular matrix (ECM) as a typical representation of intact biofilms. In contrast, Figure 2b shows markedly reduced ECM and aggregation of bacteria.

Similarly, Figure 3a below shows clustering and aggregation of untreated *V. parahaemolyticus* covered with a fine network of extracellular matrix (ECM). In contrast, as seen

in Figure 3b, *V. parahaemolyticus* exposure to AgZ4 showed a markedly reduced ECM and cell clustering.

#### **4. Discussion**

This study attempted to evaluate the antimicrobial and antibiofilm activities of silverion-exchanged zeolite (AgZ) against two marine bacterial pathogens, *V. campbellii and V. parahaemolyticus*. The first stage of this study involved the ICP-OES evaluation of Ag ion incorporation of zeolite synthesized from local bentonite clay (data published elsewhere). The initial ICP-OES analysis of AgZ1, AgZ2, AgZ3 and AgZ4 revealed the successful incorporation of Ag ion into the zeolite samples to be typically in direct correlation with the increase in Ag ion concentration introduced. The highest Ag ion concentration was observed in sample AgZ4 (1.50 M) at 190.511 ± 1.989 mg/g. At 0.50 M, the Ag ion concentration of AgZ2 was double that of AgZ1 at 160.041 ± 1.328 mg/g compared to 80.949 ± 0.389 mg/g. Nonetheless, doubled concentration increases observed for AgZ2 were not indicated in AgZ3 and AgZ4. A possible explanation for this is saturation of available sites in AgZ3 and AgZ4 for the ion exchange to occur.

In general, minimum inhibitory concentration (MIC) evaluation against *V. campbellii* and *V. parahaemolyticus* firstly revealed that the incorporation of Ag ions into the zeolite samples AgZ1, AgZ2, AgZ3 and AgZ4 was successful in inhibiting bacterial growth. Silver (Ag) has long been established in various studies as an effective antibacterial agent. Several studies have also demonstrated the efficiency of Ag-exchanged zeolite against many microbial pathogens [31–33] and concur with the findings of this study. Additionally, a typically inverse relationship between sample concentration and MIC for AgZ1, AgZ2, AgZ3 and AgZ4 was also observed. As seen in Table 2 a lower MIC, which signifies stronger antibacterial activity, was indicated for each bacterial type as the concentration of metal ion in the zeolite increased. Therefore, the higher the Ag ion loading in the zeolite, the lower the concentration of ion-loaded zeolite A needed to inhibit the growth of the bacteria. There have been many studies that have investigated the mechanism of microbial killing by metal ion. A possible mechanism involves the ability of the metal ion to attach to the bacteria membrane through electrostatic interaction and drastically alter the integrity of the bacterial membrane. Consequently, it promotes the formation of reactive oxygen species, (ROS) which will induce the oxidative stress to the bacteria cell resulting in the oxidation of cellular component, DNA damage, mitochondria damage and disruption of the cell membrane, which lead to the death of the bacteria [36–39].

MIC studies also revealed that between the two bacterial species, *V. parahaemolyticus* indicated higher susceptibility against Ag ion compared to *V. campbellii*, as lower MICs were observed across AgZ1–AgZ4. Under antibiotic pressure, bacterial phenotypes such as susceptibility, resistance, tolerance and persistence differ from one bacteria to the other. In a review by Li et al. [40], the efficacy of antimicrobials are influenced by many factors including bacterial status, host factors and antimicrobial concentrations.

A typical feature of bacterial biofilms is the extracellular matrix, which provides protection and structure to the cell population within it. The CV assay is a useful tool for rapid and simple assessment of biofilm formation differences between bacteria, as it stains the extracellular matrix as well as the aggregated bacterial cells [41]. The CV biofilm assays of *V. campbelli* and *V. parahaemolyticus* isolates grown in AgZ4 overnight showed that biofilm growth was effectively inhibited in the presence of AgZ4. However, *V. parahaemolyticus* biofilms were indicated to be more susceptible against AgZ4 with a percentage inhibition of 77.3% compared to *V. campbelli* at 60.2%. This concurred with the MIC assays, which also showed a higher susceptibility of *V. parahaemolyticus* when compared against *V. campbelli* and demonstrated significant attenuation of biofilm formation against *V. campbellii* and *V. parahaemolyticus*.

The ability of Ag ion-exchanged zeolite AgZ4 to disrupt biofilm development for *V. campbellii* and *V. parahaemolyticus* was further supported by SEM analysis. As seen in Figures 2 and 3, the exposure of AgZ4 to both *V. campbellii* and *V. parahaemolyticus* displayed

significant structural alteration of biofilm phenotypes when compared to bacterial isolates grown in the absence of AgZ4, including profound loss of the biofilm extracellular matrix (ECM) as well as markedly reduced cell aggregation. While SEM analysis of untreated isolates showed tight aggregation of cells held together by ECM, isolates exposed to AgZ4 showed higher numbers of singular isolates with lessened clustering. Breakages on the extracellular matrices of biofilms will result in increased susceptibility of bacteria against antibacterial agents and chemicals [42]. Despite much literature on antibiofilm activities of bacteria by zeolite, the mechanism of toxicity of AgZ against biofilms of *V. campbellii* and *V. parahaemolyticus* is still poorly understood. Therefore, future studies on the probability of modification of gene expression in the *Vibrio* polysaccharide (VPS) and matrix protein biosynthesis are recommended to further inform on genes or protein that are significantly affected by metal-loaded zeolites.

#### **5. Conclusions**

The study findings strongly indicate antimicrobial and antibiofilm characteristics of the silver-ion-exchanged zeolite A against the bacterial pathogens, with the highest MIC levels observed for AgZ4 (1.50 M) for *V. campbelli* and AgZ3 (1.00 M) for *V. parahaemolyticus*. Scanning electron microscopy exhibited profound breakages in the biofilm structures of both marine pathogens when grown in media added with 1.50 M silver-ion-exchanged zeolite A (AgZ4). Taken together, the results of this study strongly indicate the strong antibacterial and antibiofilm potentials of Ag-ion-exchanged zeolite A, which can be applied in the aquaculture industry to combat against infectious pathogens, in particular *V. campbellii* and *V. parahaemolyticus*.

**Author Contributions:** Formal analysis, Z.A. and N.A.W.; Supervision, S.E.A.; Writing—original draft, Z.A.; Writing—review & editing, Z.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Universiti Malaysia Sabah, grant number GUG0111-1/2017. The APC was funded by Universiti Malaysia Sabah.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article Corollospora mediterranea***: A Novel Species Complex in the Mediterranean Sea**

**Anna Poli, Elena Bovio, Iolanda Perugini, Giovanna Cristina Varese \* and Valeria Prigione**

Mycotheca Universitatis Taurinensis, Department of Life Sciences and Systems Biology, University of Torino, Viale Mattioli 25, 10125 Turin, Italy; anna.poli@unito.it (A.P.); elena.bovio@inrae.fr (E.B.); jolanda.perugini@unito.it (I.P.); valeria.prigione@unito.it (V.P.)

**\*** Correspondence: cristina.varese@unito.com

**Abstract:** The genus *Corollospora*, typified by the arenicolous fungus *Corollospora maritima*, consists of twenty-five cosmopolitan species that live and reproduce exclusively in marine environments. Species of this genus are known to produce bioactive compounds and can be potentially exploited as bioremediators of oil spill contaminated beaches; hence their biotechnological importance. In this paper, nine fungal strains isolated in the Mediterranean Sea, from the seagrass *Posidonia oceanica* (L.) Delile, from driftwood and seawater contaminated by an oil spill, were investigated. The strains, previously identified as *Corollospora* sp., were examined by deep multi-loci phylogenetic and morphological analyses. Maximum-likelihood and Bayesian phylogeny based on seven genetic markers led to the introduction of a new species complex within the genus *Corollospora*: *Corollospora mediterranea* species complex (CMSC). The Mediterranean Sea, once again, proves an extraordinary reservoir of novel fungal species with a still undiscovered biotechnological potential.

**Keywords:** marine fungi; new taxa; phylogeny; lignicolous fungi

I.; Varese, G.C.; Prigione, V. *Corollospora mediterranea*: A Novel Species Complex in the Mediterranean Sea. *Appl. Sci.* **2021**, *11*, 5452. https://doi.org/10.3390/ app11125452

**Citation:** Poli, A.; Bovio, E.; Perugini,

Academic Editor: Natália Martins

Received: 12 May 2021 Accepted: 11 June 2021 Published: 11 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

#### **1. Introduction**

The last decades have seen an increasing interest in marine fungi due to the need to broaden our knowledge on aquatic biodiversity and to exploit these organisms as a source of novel bioactive molecules. Although more than 1800 species inhabiting the oceans have been described so far, most of the fungal diversity, estimated to exceed 10,000 taxa [1], is yet to be unveiled. Several marine habitats and substrates, both biotic and abiotic, are still being explored worldwide, leading to the discovery of new marine fungal lineages. Among Sordariomycetes, one of the ascomycetous classes mostly detected in the sea, the family Halosphaeriaceae (order Microascales) usually dominates this habitat, with 65 genera and 166 species occurring on driftwood, algae, and seagrasses worldwide [2]. While most of the genera of this family are represented by one or two species, the genus *Corollospora*, typified by *Corollospora maritima* Werderm, includes 25 arenicolous species typically found in beach sand, sea-foam, shell fragments, and algal thalli [3–6]. Besides their ability to rapidly degrade cellulose [7], species affiliated with this genus are known to produce bioactive metabolites [8,9] and can be potentially exploited as bioremediators of oil spill contaminated beaches [10]. For instance, the phthalide derivative corollosporine isolated from *C. maritima* demonstrated a strong antibacterial activity against *Staphylococcus aureus*, *Bacillus subtilis*, and *Escherichia coli* [11], whereas pulchellalactam produced by *C. pulchella*, along with the antimicrobial dioxopiperazines melinacidins II–IV and gancidin W [12], showed inhibitory activity against the tyrosine phosphatase CD45 in lymphocytes [13]. Weak antibacterial activity was also observed in fractions of *C. lacera* mycelial extract [14].

In general, this genus includes morphologically diverse species whose most distinctive features are the ascospore apical primary and equatorial secondary appendages, respectively formed from the epispore and by the fragmentation of the exospore layer [15–18]. Indeed, in reference to the ascospore appendage ontogeny, a revision of the genus was done

by scanning and transmission electron microscope investigations [16]. In almost twenty years, other species of *Corollospora* were morphologically described [3,19–24]. With the upcoming molecular techniques, Campbell and collaborators [15] performed phylogenetic analysis of *Corollospora* spp. (and related taxa) based on 28S rDNA sequences and confirmed the monophyletic nature of the genus. However, the authors concluded that more genetic markers, including protein-coding genes, were necessary to resolve relationships among the species of *Corollospora* [15]. New sequence data, including 18S rDNA, ITS, and RPB1, relative to *Corollospora* spp., were then generated in the framework of the Fungal Barcoding Consortium and AFTOL project (Assembly Fungal Tree Of Life) [25,26].

Recently, nine unidentified strains belonging to the genus *Corollospora* were isolated in surveys aimed to investigate the underwater fungal diversity of the Mediterranean Sea: 4 isolates were retrieved from the seagrass *Posidonia oceanica* [27], 4 from seawater contaminated by oil spill [28], and 1 from submerged wood [29]. It is not uncommon, as it is in this case, to come across marine fungi that neither sporulate nor develop reproductive structures in axenic culture, leaving traditional morphology-based identification impossible. Consequently, the identification of these sterile mycelia must rely on molecular data [30–33].

With this study, we wish to provide an accurate phylogenetic placement of the Mediterranean strains by applying a combined multi-locus molecular phylogeny. Besides, the paper gives morphological insights into the strains that turned out to represent new species.

#### **2. Materials and Methods**

#### *2.1. Fungal Isolates*

The isolates analyzed in this study were previously retrieved from the Mediterranean Sea. In detail, 4 isolates derived from a site chronically contaminated by an oil spill in Sicily (Gela, Italy) [28] 1 from driftwood sampled in the seawater off Porto Badisco (Apulia, Italy) [29] and 4 from leaves of *P. oceanica* collected in good health condition in Tuscany (Elba island, Italy) [27] (Table 1). Originally, the strains were isolated on Corn Meal Agar medium supplemented with sea salts (CMASS; 3.5% *w*/*v* sea salt mix, Sigma-Aldrich, SL, USA, in ddH2O) and are preserved at the Mycotheca Universitatis Taurinensis (MUT), Italy.

**Table 1.** Dataset used for phylogenetic analysis. Genbank sequences include newly generated nrITS, nrLSU, nrSSU, RPB1, RPB2, TEF-1α, and βTUB amplicons (in bold) relative to the novel species Corollospora mediterranea.



**Table 1.** *Cont.*

\* = newly generated sequences; n.d. = not determined; <sup>T</sup> Type Strain. AFTOL = Assembly Fungal Tree Of Life; BBH = BIOTEC Bangkok Herbarium; CBS = Centraalbureau voor Schimmelculture; MUT = Mycotheca Universitatis Taurinensis; NBRC = Nite Biological Resource Centre; NTOU = National Taiwan Ocean University.

#### *2.2. Morphological Analysis*

All isolates were pre-grown on Malt Extract Agar-sea water (MEASW; 20 g malt extract, 20 g glucose, 2 g peptone, 20 g agar in 1 L of seawater) for one month at 21 ◦C prior to inoculation in triplicate onto new MEASW Petri dishes (9 cm Ø). Petri dishes were incubated at 15 and/or 21 ◦C. The colony growth was monitored periodically for 28 days, while macroscopic and microscopic features were assessed at the end of the incubation period.

In an attempt to induce sporulation, sterile pieces of *Quercus ruber* cork and *Pinus pinaster* wood (species autochthonous to the Mediterranean area) were placed on 3-week old fungal colonies [34]. Petri dishes were further incubated for 4 weeks at 21 ◦C. Following, cork and wood specimens were transferred to 50 mL tubes containing 20 mL of sterile seawater. Samples were incubated at 21 ◦C for at least three months.

Morphological structures were observed, and images captured using an optical microscope (Leica DM4500B, Leica microsystems GmbH, Germany) equipped with a camera (Leica DFC320, Leica microsystems GmbH, Germany).

#### *2.3. DNA Extraction, PCR Amplification, and Data Assembling*

Approximately 100 mg of fresh mycelium were carefully scraped from MEASW plates, transferred to a 2 mL Eppendorf tube, and disrupted by the mean of an MM400 tissue lyzer (Retsch GmbH, Haan, Germany). Genomic DNA was extracted following the manufacturer's instructions of a NucleoSpin kit (Macherey Nagel GmbH, Duren, DE, USA). The quality and quantity of DNA were measured spectrophotometrically (Infinite 200 PRO NanoQuant; TECAN, Switzerland); DNA samples were stored at −20 ◦C.

The partial sequences of seven genetic markers were amplified by PCR. Primer pairs ITS1/ITS4 [35], LR0R/LR7 [36], NS1/NS4 [35] were used to amplify the internal transcribed spacers, including the 5.8S rDNA gene (nrITS), 28S large ribosomal subunit (nrLSU) and 18S small ribosomal subunit (nrSSU). The translation elongation factor (TEF-1α), the βtubulin (β-TUB) and the largest and second-largest subunits of RNA polymerase II (RPB1 and RPB2) were amplified by using the following primer pairs: EF-dF/EF-2218R [37], Bt2a/Bt2b [38], RPB1Af/RPB1Cr [39] and fRPB2-5F/fPB2-7R [40]. Amplifications were run in a T100 Thermal Cycler (Bio-Rad, Hercules, CA, USA) programmed as described in Table 2. Reaction mixtures consisted of 20–40 ng DNA template, 10× PCR Buffer (15 mM MgCl2, 500 mM KCl, 100 mM Tris-HCl, pH 8.3), 200 μM each dNTP, 1 μM each primer, 2.5 U Taq DNA Polymerase (Qiagen, Chatsworth, CA, USA), in 50 μL final volume. For problematic cases, additional MgCl2, BSA, and/or 2.5% DMSO facilitated the reaction.


**Table 2.** Genetic markers, primers, and thermocycler conditions were used in this study.

Amplicons, together with a GelPilot 1 kb plus DNA Ladder, were visualized on a 1.5% agarose gel stained with 5 mL 100 mL−<sup>1</sup> ethidium bromide; PCR products were purified and sequenced at the Macrogen Europe Laboratory (Madrid, Spain). The resulting Applied Biosystem (ABI) chromatograms were inspected, trimmed, and assembled to obtain consensus sequences using Sequencer 5.0 (GeneCodes Corporation, Ann Arbor, MI, USA, http://www.genecodes.com acessed on 12 May 2021). Newly generated sequences were deposited in GenBank (Table 1).

#### *2.4. Sequence Alignment and Phylogenetic Analysis*

A dataset consisting of nrSSU, nrITS, nrLSU, and RPB1 was assembled based on BLASTn results and of the most recent phylogenetic studies focused on Halosphaeriaceae and *Corollospora* [30,41]. Reference sequences were retrieved from GenBank (Table 1). Sequences were aligned using MUSCLE (default conditions for gap openings and gap extension penalties), implemented in MEGA X (Molecular Evolutionary Genetics Analysis), visually inspected, and manually trimmed to delimit and discard ambiguously aligned

regions. Since no incongruence was observed among single-loci phylogenetic trees, alignments were concatenated into a single data matrix with SequenceMatrix [42]. The best evolutionary model under the Akaike Information Criterion (AIC) was determined with jModelTest 2 [43]. Phylogenetic inference was estimated using Maximum Likelihood (ML) and Bayesian Inference (BI) criteria. The ML analysis was generated using RAxML v. 8.1.2 [44] under GTR + I + G evolutionary model and 1000 bootstrap replicates. Support values from bootstrapping runs (MLB) were mapped on the globally best tree using the "-f a" option of RAxML and "-x 12345" as a random seed to invoke the novel rapid bootstrapping algorithm. BI was performed with MrBayes 3.2.2 [45] with the same substitution model (GTR + I + G). The alignment was run for 10 million generations with two independent runs each containing four Markov Chains Monte Carlo (MCMC) and sampling every 100 iterations. The first 25% of generated trees were discarded as "burn-in". A consensus tree was generated using the "sumt" function of MrBayes and Bayesian posterior probabilities (BPP) were calculated. Consensus trees were visualized in FigTree v. 1.4.2 (http://tree.bio.ed.ac.uk/software/figtree acessed on 12 May 2021). Three species of Microascaceae, namely *Doratomyces stemonitis*, *Microascus trigonosporus*, and *Petriella setifera*, were used to root the tree. Due to the topological similarity of the two resulting trees, only Bayesian analysis with MLB and BPP values was reported (Figure 1).

**Figure 1.** Phylogenetic inference based on a combined nrITS, nrSSU, nrLSU, RPB1 dataset. The tree is rooted in three species of Miroascaceae. Branch numbers indicate BYPP and BS values; <sup>T</sup> Type Strain; Bar = expected changes per site (0.03).

Following, a new phylogenetic analysis was conducted only on the strains investigated, whose relationship was unclear. To this aim, TEF-1α, β-TUB, and RPB2 sequences were added to the restricted dataset. Alignments and multi-loci phylogeny were conducted as described above. Sequence alignments and phylogenetic trees were deposited in TreeBASE (http://www.treebase.org, submission number S27921 and S27923, acessed on 12 May 2021).

#### **3. Results**

#### *3.1. Phylogenetic Inference*

Preliminary analyses carried out individually with nrITS, nrSSU, nrLSU, and RPB1 revealed no incongruence in the topology of the single-locus trees. The combined fourmarkers dataset–built based on BLASTn results and of recent phylogenetic studies [30,41]— Consisted of 42 taxa (including MUT isolates) that represented 9 genera and 20 species (Table 1). A total of 82 sequences (9 nrITS, 13 nrSSU, 13 nrLSU, 13 TEF-1α, 13 β-TUB, 11 RPB1, and 10 RPB2) were newly generated while 91 were retrieved from GenBank.

The dataset combining nrSSU, nrITS, nrLSU and RPB1 had an aligned length of 2431 characters, of which 1530 were conserved, 263 were parsimony-uninformative and 638 parsimony informative (TL = 2192, CI = 0.550901, RI = 0.758483, RC = 0.417849, HI = 0.449099). The strains under investigation, MUT 1587, MUT 1938, MUT 1950, MUT 1954, MUT 1961, MUT 5040, MUT 5048, MUT 5049 and MUT 5082 formed a strongly supported monophyletic lineage (BYPP = 1.00; MLB = 100%) close but well set apart from Corollospora marittima (Figure 1). Within this new group, MUT 1954 and MUT 1961 seemed to form a separated clade, as well as MUT 5040 and MUT 5049. However, the relationships between the taxa were unclear.

The additional dataset, implemented with the addition of TEF-1α, β-TUB, and RPB2 sequence data relative to the novel lineage and to Corollospora marittima only, had an aligned length of 4086 characters, of which 3693 were constant, 236 were parsimonyuninformative and 157 parsimony informative (TL = 452, CI = 0.843537, RI = 0.900217, RC = 0.759367, HI = 0.156463). Two clusters could be observed: the first grouped MUT 1954 and MUT 1961 while the second one was further split into two subclades that separated MUT 1587, MUT 1938, and MUT 5049 from MUT 5040, MUT 5048, and MUT 5082. Finally, MUT 1950, although included second cluster, was not part of any of the two subclades (Figure 2).

**Figure 2.** Phylogenetic inference based on a combined nrITS, nrSSU, nrLSU, RPB1, RPB2, TEF-1α, and βTUB dataset. The tree is midrooted. Branch numbers indicate BYPP and BS values; <sup>T</sup> Type Strain; Bar = expected changes per site (0.004).

#### *3.2. Taxonomy*

*Corollospora mediterranea* sp. nov. A. Poli, E. Bovio, G.C. Varese and V. Prigione, MYCOBANK: MB 839640, Type. Italy, Sicily, Mediterranean Sea, Gela (CL), July 2013, from seawater contaminated by an oil spill, R. Denaro, MUT 1950 holotype, living culture permanently-preserved in metabolically inactively state by deep-freezing at MUT. Additional material examined. Italy, Sicily, Mediterranean Sea, Gela (CL), July 2013, from seawater contaminated by an oil spill, R. Denaro MUT 1938, MUT 1954 and MUT 1961. Italy, Apulia, Mediterranean Sea, Porto Badisco (LE), July 2011, from driftwood, L. Garzoli, MUT 1587. Italy, Tuscany, Mediterranean Sea, Elba Island (LI), from the seagrass *Posidonia oceanica*, March 2010, R. Mussat-Sartor and N. Nurra, MUT 5040, MUT 5048, MUT 5049 and MUT 5082. Etymology. In reference to the Mediterranean Sea. Description. Growing actively on *Pinus pinaster* and *Quercus ruber* cork. *Hyphae* two types observed, one thin (1 μm wide) and hyaline, one thicker (3 μm wide) and melanized. *Chlamydospores* numerous, in chain globose to subglobose, 4–13 × 4–11 μm diameter (Figure 3). Sexual morph not observed. Asexual morph with differentiated conidiogenesis not observed. Colony description. Colonies on MEASW attaining 38–54 mm diam after 28 days at 21 ◦C (Figure 4), mycelium white, grey, grey-green, sometimes with pinkish shades; dense and feltrose, occasionally umbonate in the middle, with radial grooves; reverse from light brown to dark green, occasionally with concentric rings, lighter to the edges. Neither soluble pigments nor exudates were observed (Figure 3).

**Figure 3.** *Corollospora mediterranea* sp. nov. MUT 1950 (holotype), the 28-days-old colony at 21 ◦C on MEASW (**A**) and reverse (**B**); chlamydospores in the chain (**C**); mycelial growth on *Pinus pinaster* wood (red arrow) and *Quercus ruber* cork (blue arrow) (**D**). Scale bars: 10 μm.

**Figure 4.** Growth rate on MEASW of the strains investigated over 28 days. The symbol indicates the isolation substrate (asterisk: driftwood; full circle: oil spill; vertical line: leaves of *Posidonia oceanica*); the color indicates the position in the phylogenetic tree (red: Clade 1, Subclade 1a; green: Clade 1, Subclade 1b; purple: Clade 1; blue: Clade 2).

#### **4. Discussion**

The description of the strains investigated in this study was particularly tricky and complicated since neither asexual nor sexual reproductive structures developed in axenic conditions. As a consequence, it was not possible to describe the range of diagnostic traits amongst these newly identified lineages.

To better characterize these fungi, we tried to mimic the saline environment by using a culture medium supplemented with seawater, since it is well known that only this method supports a measurable growth of vegetative mycelium [33]. The family Halospaheriaceae, to whom the genus *Corollospora* belongs, includes the largest number of marine lignicolous species. It was thus realistic to induce sporulation by placing wood and cork specimens on the colony surface prior to transfer them into seawater. Despite wood colonization, sporulation did not occur, whereas chlamydospores were abundantly produced. Likewise, strains of *Corollospora* sp. isolated from intertidal decayed wood of mangrove trees and driftwood in Saudi Arabia did not develop reproductive structures but only chlamydospores that were much bigger and not comparable to those observed in this study [30]. It must be considered that strictly vegetative growth with no sporulation is a common feature of marine fungi [46–48] that are likely to rely on their dispersion to hyphal fragments and/or chlamydospores. Most probably, a necessary requisite for fungi to develop reproductive structures is the occurrence of those environmental conditions these organisms are adapted to (e.g., high salinity, low temperature, high pressure, wet-dry cycles, etc.).

The phylogenetic analysis based on ribosomal genes (nrITS, nrLSU, nrSSU) and RPB1, shows a clear distance between the strains under investigation and the other species of *Corollospora*, *Corollospora maritima* being the closest one. This new and strongly supported clade may include one or more novel species. Hypothetically, the tree highlights the presence of three clusters that include: (i) MUT 1954 and MUT 1961; (ii) MUT 1938, MUT 5040, MUT 5049 and MUT 5082; (iii) MUT 1587, MUT 1950, and MUT 5048 (Figure 1). Bearing in mind the conclusion drawn by Campbell et al. [15] that indicated the need of sequencing more genetic markers to clarify the relations among the species of the genus *Corollospora*, a dataset focused on this new group, together with *C. maritima*, was built with the addition of three more protein-coding genes, namely RPB2, TEF-1α and β-TUB (Figure 2). From one side, this strongly supported tree points out once more the distance from *C. maritima*, strengthening the idea that we are dealing with a new species. It is

undeniably complicated to define a fungal species, and *Corollospora mediterranea* was here established following the recommendations outlined by Jeewon and Hyde [49], who, dealing with the issue of the species boundaries and identification of new taxa, pointed out a number of key elements to follow. Notably, all the ITS sequences (including 5.8S) analyzed are longer than the minimum 450 base pairs required and display a percentage of identity with the closest relative *C. maritima* < 95%. In addition, as recommended, the strongly supported phylogenetic tree includes the minimum number (4–5) of closely related taxa of the same genus (Figure 1). A thorough inspection of the tree shown in Figure 2 reveals the presence of two clusters: the former (Clade 1) consists of MUT 1950 and two subclusters (MUT 5040, MUT 5048 and MUT 5082; MUT 1587, MUT 1938, MUT 5049), the latter (Clade 2) includes MUT 1954 and MUT 1961. On the other hand, the lack of distinct micromorphological traits (neither sexual nor asexual morph with differentiated conidiogenesis detected) and the huge variability observed among the colonies (in terms of growth rate, texture, surface and reverse color) lead us to introduce the *C. mediterranea* species complex (CMSC), since we could not discern the species boundaries with certainty (Figures 5 and 6). Indeed, the term "species complex" comes to help taxonomists when: (i) a group of organisms may represent more than one species; (ii) morphological features are overlapping due to extreme variability; (iii) the species may be somehow related although no certain assumption can be assessed. Furthermore, no clear correlation between colony features, growth rates, source of isolation, and/or phylogenetic position was noticed. In general, researchers introduce a species complex when facing a problematic topic. This is the case for example of *Fusarium oxysporum* species complex (FOSC) [50,51], *Fusarium solani* species complex (FSSC) [52], *Wallemia sebi* species complex (WSSC) [53], *Colletotrichum acutatum* species complex (CASC) [54] and many others. Most of these cases were resolved with a revision of the species complex, where individual species were identified based on a multiloci phylogeny. However, in our study, this approach did not lead to a sharp resolution of the complex. The genome sequencing of all the strains investigated would be an option to sort out this intriguing issue. This approach would possibly reveal those genetic regions that may allow an easy distinction of the species within the complex [55]. On the other hand, the same goal could be achieved by the analysis of secondary metabolites [53]. An investigation of this sort is important also from another point of view: it is now recognized that marine fungi are a reservoir of novel active metabolites that can be harnessed for pharmaceutical, nutraceutical, cosmetic, and environmental purposes.

Kirk and Gordon [10] demonstrated that a number of strains of *C. maritima*, *C. lacera*, and *C. intermedia* were capable of using hexadecane as the sole carbon source, and assumed that the genus *Corollospora* could find utility in petroleum degradation. Supporting this idea is the isolation of 4 strains of *C. mediterranea* (MUT 1938, MUT 1950, MUT 1954, and MUT 1961) from a site chronically contaminated by an oil spill in the Mediterranean Sea [28]. The aforementioned findings indicate the great versatility and adaptability of these fungi to hydrocarbons-contaminated environments, with the consequent possibility of being used as bioremediators. In addition, knowing that the species of *Corollospora* belong to a family of lignicolous fungi and that are rich in cellulase and lignin-degrading enzymes [7], their retrieval from leaves of the seagrass *P. oceanica* (MUT 5040, MUT 5048, MUT 5049, and MUT 5082) and from driftwood (MUT 1587) does not come as a surprise. It is, therefore, reasonable to assume a role of *C. mediterranea* in degrading and recycling organic matter, making them available for other organisms.

**Figure 5.** *Corollospora mediterranea* sp. nov. 28-days-old colonies at 21 ◦C on MEASW: (**A**) MUT 1587, (**B**) MUT 1938, (**C**) MUT 5049 (strains belonging to the Clade 1, Subclade 1a); (**D**) MUT 5048, (**E**) MUT 5082, (**F**) MUT 5040 (strains belonging to the Clade 1, Subclade 1b); (**G**) MUT 1954, (**H**) MUT 1961 (strains belonging to the Clade 2).

**Figure 6.** *Corollospora mediterranea* sp. nov. reverse of 28-days-old colonies at 21 ◦C on MEASW: (**A**) MUT 1587, (**B**) MUT 1938, (**C**) MUT 5049 (strains belonging to the Clade 1, Subclade 1a); (**D**) MUT 5048, (**E**) MUT 5082, (**F**) MUT 5040 (strains belonging to the Clade 1, Subclade 1b); (**G**) MUT 1954, (**H**) MUT 1961 (strains belonging to the Clade 2).

#### **5. Conclusions**

In conclusion, the retrieval of *C. mediterranea* from different substrates, localities and in different sampling campaigns indicates its constant presence in (at least) the Mediterranean Sea and points out how the marine environments is still largely uninvestigated from a mycological point of view. Therefore, it is more and more glaring that the Oceans are a huge reservoir of unidentified microorganisms with a valuable biotechnological potential not yet disclosed. In the next future, we would aim at resolving the species complex, by applying more approaches to induce sexual and/or asexual sporulation. In parallel, all the strains studied in this work, will be investigated for the production of new and powerful bioactive molecules.

**Author Contributions:** Conceptualization, A.P., E.B., and V.P.; methodology, A.P., E.B., and V.P.; software, A.P.: validation, A.P., I.P., and V.P.; formal analysis, A.P., E.B., and V.P.; investigation, A.P., E.B., and V.P.; resources, V.P. and G.C.V.; data curation, A.P., I.P., and V.P.; writing—original draft preparation, A.P. and V.P.; writing—review and editing, A.P., V.P., and G.C.V.; visualization, A.P. and V.P.; supervision, V.P. and G.C.V.; project administration, A.P., V.P., and G.C.V.; funding acquisition, G.C.V. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the University of Torino (ex 60%).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Sequence data are available at Genbank (NCBI) under the accession numbers reported in the manuscript.

**Acknowledgments:** The authors would like to thank JRU MIRRI-IT for technical and scientific support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Extra-Heavy Crude Oil Degradation by** *Alternaria* **sp. Isolated from Deep-Sea Sediments of the Gulf of Mexico**

**Lucia Romero-Hernández 1, Patricia Velez 2, Itandehui Betanzo-Gutiérrez 3, María Dolores Camacho-López 1, Rafael Vázquez-Duhalt <sup>3</sup> and Meritxell Riquelme 1,\***


**Abstract:** The Gulf of Mexico (GoM) is an important source of oil for the United States and Mexico. There has been growing interest, particularly after the Deepwater Horizon oil spill, in characterizing the fungal diversity of the GoM and identifying isolates for use in the bioremediation of petroleum in the event of another spill. Most studies have focused on light crude oil bioremediation processes, while heavy crude oil (HCO) and extra-heavy crude oil (EHCO) have been largely ignored. In this work, we evaluated the ability of fungal isolates obtained from deep-sea sediments of the Mexican economic exclusive zone (EEZ) of the GoM to degrade HCO (16–20◦ API) and EHCO (7–10◦ API). *Alternaria* sp., *Penicillium* spp., and *Stemphylium* sp. grew with HCO as the sole carbon source. Remarkably, *Alternaria* sp. was the only isolate able to grow with EHCO as the sole carbon source, degrading up to 25.6% of the total EHCO and 91.3% of the aromatic fraction, as demonstrated by gas chromatography analysis of the saturate, aromatic, and polar fractions. These findings proved to be significant, identifying *Alternaria* sp. as one of the few fungi reported so far capable of degrading untreated EHCO and as a suitable candidate for bioremediation of EHCO in future studies.

**Keywords:** heavy crude oil; mycodegradation; deep-sea fungi; bioremediation; fungal isolation

#### **1. Introduction**

Crude oil has become the most important source of energy for humankind. It has been estimated that, by 2035, it will supply more than one-third of the total global energy demands [1,2]. Light crude oil has been the primary option to produce refined products for everyday use, such as gasoline, diesel, and kerosene, among many others [3,4]. However, reserves of conventional light crude oil are rapidly being depleted all around the world, while, at the same time, the demand for fossil fuel is continually increasing as a result of human population growth and economic development [5]. Alternative green energy sources with lower carbon footprints (e.g., solar and eolic) have improved to become economically viable, but they are nonetheless insufficient to supply the increasing energy demand [6]. Therefore, the exploitation of unconventional petroleum fuel resources (tar sands and extra-heavy oil deposits) might emerge as an alternative source of fossil fuel in the next decades [7]. Heavy and extra-heavy crude oil (HCO and EHCO, respectively) make up 70% of the world's oil reserves [7]. However, their high content in resins and asphaltenes represents a significant challenge for its extraction, production, and refinement, in addition to the associated risk of accidental spills. Additionally, polycyclic aromatic hydrocarbons (PHAs) are also present in a significant ratio in HCO and EHCO. These fractions are one of the major concerns in oil spills, given their resistance to microbial degradation [8], long-term persistence in the environment [9], and ecotoxicity [10–12].

**Citation:** Romero-Hernández, L.; Velez, P.; Betanzo-Gutiérrez, I.; Camacho-López, M.D.; Vázquez-Duhalt, R.; Riquelme, M. Extra-Heavy Crude Oil Degradation by *Alternaria* sp. Isolated from Deep-Sea Sediments of the Gulf of Mexico. *Appl. Sci.* **2021**, *11*, 6090. https://doi.org/10.3390/ app11136090

Academic Editors: Anna Poli and Valeria Prigione

Received: 20 May 2021 Accepted: 23 June 2021 Published: 30 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Bioremediation is a well-documented option that involves applying microorganisms to neutralize petroleum in the environment [13]. It targets those microbes capable of synthesizing enzymes that cleave C:H bounds from petroleum hydrocarbons and transforms them into harmless and less persistent molecules [14]. However, while the microbial biodegradation of hydrocarbons and crude oils has been widely studied for decades, most studies have focused on light crude oils and their low molecular weight fractions or purified components [15]. In contrast, the microbial biodegradation of HCO and EHCO, particularly by fungi, has been scarcely studied.

Only a few fungal strains isolated from extreme environments are able to degrade asphaltenes and high molecular weight PAHs [16]. *Aspergillus fischeri* (previously *Neosartorya fischeri*), isolated from an asphalt lake in Venezuela, was the first microorganism able to grow using purified asphaltenes as the sole carbon source [17]. This fungus set a unique precedent demonstrating its remarkable ability to degrade extremely complex hydrocarbon molecules, especially high molecular weight PHAs [18]. The extremophilic fungus *Pestalotiopsis palmarum* BM-04 isolated from an asphalt lake in Venezuela synthesizes oxidative exoenzymes that catalyze the biotransformation of maltenes, asphaltenes, and the petroporphyrins-rich fraction of biotreated EHCO [19]. *Paecilomyces variotii*, *Fusarium decemcellulare*, *Candida palmioleophila*, and *Pichia guilliermondii*, isolated from a petroleum activity site in Indonesia, were able to degrade 10–15% of resins and asphaltenes using BAL150 light crude oil as the sole carbon source [20]. A chloroperoxidase obtained from *Caldariomyces fumago* transformed the porphyrin-free asphaltene fraction recovered from Maya HCO, reducing it by 24% [21]. Furthermore, *Daedaleopsis* sp., isolated from a forest in Iran, showed biodegradation of 38% of the asphaltene and aromatic fractions of HCO [22].

The Gulf of Mexico (GoM) is rich in petroleum reservoirs, and since 2017, their ultradeep waters have been exploited to obtain crude oil [23]. Therefore, it represents an ideal prospecting source to isolate indigenous fungal strains adapted to deep-sea conditions and capable of degrading crude oil. In a previous study, the fungal taxa *Aureobasidium* sp., *Penicillium brevicompactum*, *Penicillium* sp., *Phialocephala* sp., and *Cladosporium* sp. isolated from sediments of the Mexican economic exclusive zone (EEZ) of the GoM were identified as able to metabolize alkane and alkene, long-chain hydrocarbons poorly degraded by bacteria, as the sole carbon sources [24]. Furthermore, the RNA-seq expression analysis of *Penicillium* sp. cultured with hexadecane or 1-hexadecene showed the upregulation of genes involved in transmembrane transport, metabolism of six-carbon carbohydrates, and nitric oxide pathways [24].

In this study, we set out to identify culturable strains of marine fungi native to the seafloor of the Mexican EEZ of the GoM with the ability to degrade untreated EHCO. *Alternaria* sp. was the only isolate identified as capable of growing using EHCO as the sole carbon source and degrading a significant portion of the whole EHCO.

#### **2. Materials and Methods**

#### *2.1. Sampling*

The deep-sea sediment samples were collected from twenty-four stations located in Coatzacoalcos (also known as the Salina basin) and in the Perdido fold belt in the Mexican EEZ of the GoM (Figure 1). Samples were collected employing a Reineck box corer at different depths (Table S1) during the Metagenomics (MET-II) campaign that took place in 2017 onboard the research vessel Justo Sierra (UNAM). Subsamples were taken from the first 20 cm of the sediment collected with the box corer using pre-cut 50-mL sterile syringes. Syringes were sealed with plastic film and stored at 4 ◦C and transported to the laboratory for processing.

**Figure 1.** Localization of twenty-four sampled stations in the Gulf of Mexico, nine in Coatzacoalcos, also known as the Salina basin (orange dots), and fifteen in the Perdido fold belt region (purple dots).

#### *2.2. Fungal Isolation and Identification*

Sediment samples were processed as described previously [24]. Briefly, fungi were initially isolated in a five-fold diluted medium. For maintenance, fungal isolates were grown on a potato dextrose agar (PDA) medium. Colony morphology and pigmentation were registered using a Nikon D3100 Digital Camera (Nikon Inc., Tokyo, Japan). In addition, to analyze the microscopic characteristics, including the hyphal shape, morphology, size of spores, and conidiogenous cells, fungi were grown on water agar (1.5% agar). Samples were processed by using the inverted block technique [25]. Imaging was carried out by DIC (Differential Interference Contrast) on a Nikon eclipse Ti-E microscope with an oil 60× objective. Macroscopic and microscopic characteristics, as well as Taxonomic Keys [26] and the data available in MycoBank (http://www.mycobank.org/) (accessed on 9 December 2020), were considered to reach genus-level identification. Total genomic DNA was extracted from axenic isolates with the DNeasy Plant Mini Kit (Qiagen). Sequencebased taxonomic verification was achieved through phylogenetic analyses [27] using: (1) the internal transcribed spacer (ITS) region of the nuclear ribosomal DNA ITS1-5.8 rDNA using the primer set ITS1F (5-CTTGGTCATTTAGAGGAAGTAA-3) [28] and ITS4 (5-TCCTCCGCTTATTGATATGC-3) [29], (2) the 18S rRNA gene using primer set NS1 (5- GTAGTCATATGCTTGTCTC-3) and NS4 (5-CTTCCGTCAATTCCTTTAAG-3) [30], and (3) the β-tubulin gene (benA) with primer set Bt2A (5-GGTAACCAAATCGGTGCTGCTTTC-3) and Bt2b (5-ACCCTCAGTGTAGTGACCCTTGGC-3) [31] and with degenerate primers LR1 (5-RRCRACAARTTCGTGCCCCG-3) and LR2 (5-AGTGAACTGGTCACCYACAC-3), newly designed to amplify another region of the β-tubulin gene downstream of the region amplified with primer set Bt2A and Bt2b for cases in which that primer set did not work. The PCR products were purified using a QIAquick Gel Extraction Kit and commercially sequenced by Eton Bioscience, Inc., San Diego, CA, USA.

Consensus sequences were generated using Consed v29.0 [32–34] and compared to the GenBank Database through a BLAST search (http://www.ncbi.nlm.nih.gov/BLAST) (accessed on 9 December 2020) using the MegaBLAST algorithm to retrieve the reference sequences for the phylogenetic analyses. Only hit sequences with a minimum coverage of 98% were considered for the analyses, pondering accessions associated with vouchers and type material (Table S2). Uncultured/environmental sample sequences were not considered for the analyses. Sequences were aligned with the software UGENE v36.1 [35] using MUSCLE [36]. Phylogenetic trees were inferred by MEGA X [37,38] with the Maximum Likelihood (ML) algorithm, a bootstrap test of 1000 replications, and the K2+I (ITS and 18S) and K2+G and HKY+G (β-tubulin benA and LR1-LR2, respectively) models of nucleotide substitution [39]. Initial trees for the heuristic search were obtained automatically by applying Neighbor-Joining and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach and then selecting the topology with a superior log likelihood value. The sequences are available in GenBank under the accession numbers MW412481-MW412485 for the 18S, MW412490-MW412494 for the ITS, and MZ392424-MZ392428 for the β-tubulin sequences (Table S3).

#### *2.3. Growth with Crude Oil as a Sole Source of Carbon*

Fungal conidia were cultured using crude oil as a sole carbon source to evaluate if oil metabolization occurred. Fungal strains were grown on PDA plates and incubated for five days at 30 ◦C to obtain conidia. Even though the strains could grow at room temperature, at 30 ◦C, an accelerated growth was observed. Conidia were recovered from the PDA plates using distilled sterile water, separated from mycelium through filtration, and washed twice with sterile distilled water. Conidia were counted using a Neubauer Chamber. Flasks containing 50 mL of modified Czapek minimal medium without a carbon source (4-g NaNO3, 2-g K2HPO4, 1-g MgSO4•7H2O, 1-g KCl, 0.02-g FeSO4•7H2O, 32-g sea salt, and 0.5 mL of trace mineral solution) [40] were inoculated with 2 × 107 conidia. The flasks were supplemented with 0.5% of three different types of petroleum according to the American Petroleum Institute (API) classification: light crude oil (40◦ API from the well Xux in Tabasco, Mexico), heavy (16–20◦ API from the field Bacal in Tabasco), or extra-heavy (7–10◦ API from the well Ayatsil-Telek in Campeche, Mexico) crude oil as a sole source of carbon. The cultures were incubated at 30 ◦C under 150 rpm up to a month with weekly revisions.

#### *2.4. Microscopic Analysis*

To assess the ability of conidia to germinate using crude oil as a unique carbon source, hyphal development was observed by Differential Interference Contrast (DIC) microscopy using a Nikon Inverted Microscope Eclipse Ti-E with a 40 × 0.95 objective. Mycelial samples from each culture were taken from the petroleum–culture medium interface with a sterile bacteriological loop. The sample was spread onto a microscope slide with a drop of water.

Since *Alternaria* sp. was the only strain that displayed visible growth on the EHCO added to the modified Czapek medium, the samples were analyzed by Scanning Electron Microscopy (SEM) to image the mycelium–petroleum interface further. After three weeks, when mycelium was visible, the medium was poured off. A small portion of the mix containing petroleum and fungal biomass was fixed for two hours with 50 mL of a solution containing paraformaldehyde 1%, glutaraldehyde 2%, and potassium buffer 0.1 M (K2HPO4 1M and KH2PO4 1M, pH 6.8). The solution was removed by decantation, and the sample was covered with 2% osmium tetroxide and incubated for one hour. Finally, alcohol at different concentrations (25%, 50%, 75%, and 100%) was employed to fix and dehydrate the sample, which was imaged on a Hitachi SU3500 SEM.

#### *2.5. Fungal Growth Quantification*

To corroborate the mycelial growth of *Alternaria* sp. with EHCO as a unique carbon source, the protein concentration was quantified. The fungal cultures were incubated for a month at 30 ◦C under 150 rpm. Next, 20 mL of toluene were added to separate the mycelium from the EHCO. The culture was shaken vigorously and transferred to a centrifuge tube. The mixture was centrifuged at 15,000 rpm for 10 min, and the organic upper layer containing the remaining oil was removed. The pellet containing the fungal biomass was recovered by filtration with a vacuum pump through a Whatman Grade 1 Qualitative Filter Paper 1 0.25 mm. The mycelium was ground in a mortar with liquid nitrogen; mixed with 1 mL of lysis buffer (1-M Tris pH 7.4, 2-M KCl, 1-M MgCl2, Triton X100, and one tablet of Complete Mini Protease Inhibitor Cocktail, Roche, CDMX, Mexico); and maintained for 30 min at 4 ◦C. After centrifugation at 12,000× *g* for 10 min, the supernatant

was collected to measure the total protein content of the lysate. Abiotic cultures (EHCO and modified Czapek minimal medium) and cultures without EHCO were used as controls, and all experiments were performed in independent triplicates. An independent samples *t*-test was conducted to compare the mg/L of protein in control and treatment cultures. A Bio-Rad Protein Assay was used to quantify the protein content, and a BSA standard curve was included.

When all the strains were tested in the early stages of this project, *Penicillium* spp. were grown under different concentrations of light crude oil (0.25, 0.5%, and 0.75% *v/v*) as the sole carbon source to determine at which concentration they produced more biomass. It was unnecessary to add toluene to separate the mycelium from the oil, because there was no strong attachment between them, and the mycelium was easily recovered through filtration. Abiotic cultures (light crude oil and modified Czapek minimal medium) and cultures without light crude oil were used as controls, and all experiments were performed in independent triplicates. A one-way between treatments ANOVA was conducted to compare the effect of petroleum at different concentrations. Post hoc comparisons were conducted using Turkey's honestly significant difference (HSD).

#### *2.6. Liquid–Liquid Extraction*

Flasks containing 50 mL of modified Czapek medium supplemented with 0.5% *w/v* of EHCO (7–10◦ API) as unique carbon sources and inoculated with 2 × 107 conidia of *Alternaria* sp. were incubated at 30 ◦C and stirred at 150 rpm for a month. The remaining oil was extracted to perform the gas chromatography analysis. The medium was acidified with HCl and extracted with 10 mL of dichloromethane three times. The organic extract was recovered, dehydrated on an anhydrous Na2SO4 column, and dried on a rotary evaporator under 0.79 atm.

#### *2.7. SAP Analysis*

To analyze the percentage of degradation of the different fractions of EHCO by *Alternaria* sp., the crude oil was fractionated by a modified SARA (saturates, aromatics, resins, and asphaltenes) standard method (ASTM D2007). The modification consisted in obtaining the resins and asphaltenes components together in one fraction only. As a result, the saturates, aromatics, and polar (resins and asphaltenes) fractions were obtained (SAP). All the fractions were analyzed by gas chromatography. The solvents used *were* n-hexane to separate the saturates, toluene for the aromatics, and a mixture of methanol and dichloromethane (50:50) for resins and asphaltenes. The separation was made through a silica column, and three fractions were obtained according to the increasing polarity of the solvents. The fractions were evaporated on a rotary evaporator and analyzed on an Agilent 7820A Gas Chromatographer with a Zebron Inferno 20 m × 0.18 mm × 0.18 μm column (Phenomenex, Los Angeles, CA, USA), according to the EPA-8015 method (US EPA). The Zebron Inferno column, with a high temperature range, allowed to analyze the volatile asphaltenes. The total petroleum hydrocarbons (TPH) were quantified as the response of the flame ionization detector (FID). The GC program started at 50 ◦C (2 min), followed by a temperature ramp of 10 ◦C/min at 375 ◦C (5 min). The spitless injector was set up at 360 ◦C and the detector at 400 ◦C. Helium was used as carrier gas at a constant flow of 0.9 mL/min.

#### **3. Results**

#### *3.1. All Fungal Isolates Recovered belonged to the Ascomycota*

Twenty-seven fungal isolates were obtained from the twenty-four sites of the GoM. These were clustered into five morphotypes based on macroscopic and microscopic morphological features (Figure 2). Based on the phylogenetic analysis of the sequenced ITS region, 18S rRNA gene, and β-tubulin gene, we unequivocally identified *Cladosporium halotolerans* to the species level (≥99% beta-tubulin sequence similarity and supported by 97% bootstrap value) and four isolates to the genus level (Table S3 and Figure 3). All the

isolates belonged to the phylum Ascomycota within the Aspergillaceae, Cladosporiaceae, and Pleosporaceae families (Figure 3). *Alternaria* sp. was isolated from the Perdido region (station N2; ID:4), whereas *C. halotolerans* (station C11; ID:1), *Penicillium* sp. 1 (station D16; ID:5), *Penicillium* sp. 2 (station C11; ID:6), and *Stemphylium* sp. (station C12; ID:3) were isolated from the Coatzacoalcos region (Table S3).

**Figure 2.** Macromorphology and micromorphology of the fungal isolates recovered from deep-sea sediment samples evaluated for their capacity to degrade petroleum. Top row: colony morphology of strains grown on Petri dishes containing potato dextrose agar medium. Bottom row: conidiophores and conidia of the fungal strains imaged by Differential Interference Contrast microscopy. Scale bar = 10 μm. (**A**) *Cladosporium halotolerans*, (**B**) *Stemphylium* sp., (**C**) *Alternaria* sp., (**D**) *Penicillium* sp. 1, and (**E**) *Penicillium* sp. 2.

**Figure 3.** Evolutionary analysis by Maximum Likelihood method for (**A**) 18S, (**B**) ITS, and (**C**) βtubulin. Bootstrap values are indicated above the branches. The phylogenetic position of deep-sea derived fungal sequences is denoted by blue color. The evolutionary history was inferred by using K2+I (**A**,**B**), K2+G (**C**), and HKY+G (**D**) substitution models. The tree with the highest log likelihood is shown. The initial tree for the heuristic search was obtained automatically by applying the Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach and then selecting the topology with the superior log likelihood value. The tree is drawn to scale, with the branch lengths measured in the number of substitutions per site.

#### *3.2. Alternaria sp. Was Able to Grow with Extra-Heavy Crude Oil as the Sole Carbon Source*

The five fungal strains were assayed for growth at 0.5% of three different crude oils: light, heavy, or extra-heavy. All five strains were able to grow in light crude oil as the sole source of carbon. However, the heavier the crude oil was, the fewer strains were able to metabolize it. While four strains (*Alternaria* sp., *Penicillium* sp. 1, *Penicillium* sp. 2, and *Stemphylium* sp.) were able to grow in HCO, only one, *Alternaria* sp., was able to grow in EHCO (Table 1).

**Table 1.** Growth of selected fungi in crude oils of different densities. The culture was carried out in 50 mL of modified Czapek minimal medium with 0.5% *w/v* of crude oil as the sole carbon source.


+ growth, − no growth.

Under light microscopy, *Alternaria* sp., *Penicillium* sp. 1, and *Penicillium* sp. 2 hyphae were observed strongly adhered to the surface of the HCO (Figure 4A–C). In contrast, *Stemphylium* sp. presented hyphal growth that did not adhere to the oil (Figure 4D). Regardless of how attached the hyphae were to the crude oil, the main purpose of this experiment was to confirm the presence of mycelium, and this was successfully achieved, as shown in the images obtained. When grown in EHCO, hyphae and conidia of *Alternaria* sp. also adhered to the sticky oil drops. By DIC microscopy, what appeared to be newly formed conidia at the tips of the hyphae were observed at the petroleum surface (Figure 5). By SEM, a dense mass of mycelium and conidia were observed as well growing at the surface of the petroleum (Figure 6).

**Figure 4.** Differential interference contrast microscopy of fungal mycelia grown in Czapek minimal medium containing 0.5% *w/v* of heavy crude oil (16–20◦ API). (**A**) *Alternaria* sp. conidia (arrowheads) and mycelium (arrows) adhered to the oil after six days of incubation. (**B**) *Penicillium* sp. 1 hyphae (arrows) and conidia growth (arrowheads) after 24 days of incubation; salts from the Czapek minimal medium (small arrows) can be observed. (**C**) *Penicillium* sp. 2 mycelium (arrows) and conidia (arrowheads) embedded in salt (small arrows) and petroleum after 24 days of incubation. (**D**) *Stemphylium* sp. hyphae (arrows) with no attachment to petroleum after 27 days of incubation. Scale bar = 25 μm.

**Figure 5.** Differential interference contrast microscopy of *Alternaria* sp. grown in Czapek minimal medium containing 0.5% *w/v* of extra-heavy crude oil (7–10◦ API). (**A**) Hyphae (arrow) and septate conidia (arrowhead) emerging from a dense mass of petroleum. (**B**) Mycelium (arrow) and conidia (arrowhead) with Czapek minimal medium salts and petroleum. Scale bar = 25 μm.

**Figure 6.** Scanning electron microscopy of *Alternaria* sp. grown in Czapek minimal medium containing 0.5% *w/v* of extraheavy crude oil (7–10◦ API). (**A**) Hyphae (small arrows), conidia (arrowheads), and smooth petroleum surface (arrows). Scale bar = 100 μm. (**B**) Mycelium (small arrows) and conidia (arrowheads). Scale bar = 200 μm. (**C**) Hyphae (small arrows) and conidia (arrowheads). Scale bar = 300 μm. (**D**) Mycelium (small arrows), conidia (arrowheads), and petroleum (arrows). Scale bar = 500 μm. In all images, the mycelium is in tight contact with the petroleum surface.

For both *Penicillium* spp., the maximum average amount of protein (40.5 ± 1.2 mg/L for *Penicillium* 1 and 46.0 ± 13.7 mg/L for *Penicillium* 2) was obtained for the cultures amended with 0.75% *v/v* light crude oil (Figure S1). The protein concentration in *Penicillium* sp. 1 was significantly (*p* = 0.012) higher at the 0.75% (*v/v*) concentration than 0.25% (*v/v*) of the light crude oil. Whereas, for *Penicillium* sp. 2, no significant differences could be observed at the different concentrations of light crude oil (*p* = 5.14) (Figure S1).

To quantify protein concentration in the cultures of *Alternaria* sp. grown in Czapek minimal medium supplemented with 0.5% *v/v* EHCO, it was necessary to add toluene to separate the mycelium from the oil, because there was a strong attachment between them, and the mycelium could not be easily recovered through filtration. An average total protein increase of 7.04 ± 0.20 mg/L was obtained for *Alternaria* spp. cultures that were grown for a month in Czapek minimal medium supplemented with 0.5% *w/v* EHC. That value was slightly higher, although not more significantly than the one obtained for the control cultures without EHCO (5.9 ± 0.32 mg/L) (*p* = 0.066).

#### *3.3. Alternaria sp. Mainly Metabolizes the Aromatic Components of the EHCO*

After 30 days of growth, EHCO degradation by *Alternaria* sp. was determined by gas chromatography in both whole EHCO and their fractions (Figure 7). The main components of the original EHCO corresponded to the aromatic fraction with 43.5% (±10.8), followed by saturates with 35.5% (±1.5) and the polar (resins-asphaltenes) fractions 21% (±1.9).

**Figure 7.** Hydrocarbon degradation after 30 days of growth by *Alternaria* sp. with extra-heavy crude oil (7–10◦ API) as the sole source of carbon initial and residual oil compositions. The number on the top of the columns represents the residual oil after biodegradation as a percentage of initial crude oil. The numbers on each column are the percentage of each fraction. The oils were fractionated by the modified SARA procedure to obtain the saturates, aromatics, and polar (resins and asphaltenes) fractions. The total petroleum hydrocarbons were determined by gas chromatography equipped with a flame ionization detector.

According to the chromatograms, *Alternaria* sp. degraded 25.6% (±7.6) of the total crude oil (Figure S2). Noteworthy, *Alternaria* sp. had a higher metabolization of the aromatic fraction, with 91.3% (±15.7) of degradation of these compounds. It also degraded the saturates fraction by 24.7% (±1.0). The resin–asphaltene fractions showed an increase of 108% (±5.8), mainly due to the production of polar molecules. In other words, the composition of the crude oil after biodegradation was 26.7% (±1.4) of the saturates fraction, 3.7% (±0.9) of the aromatic fraction, and 43.8% (±3.2) of polar (resin–asphaltene) fractions (Figure 7).

#### **4. Discussion**

Following the 2010 Deepwater Horizon oil spill, fungal strains from oil-soaked sand patties collected from beaches [41] and bacteria from oiled beach sands at Pensacola Beach, FL, USA [42], both sites located in the American EEZ, were identified as potential bioremediation microorganisms. Fungi synthesize a battery of enzymes, including laccases, peroxidases, lytic polysaccharide monooxygenases, and monooxygenase cytochrome P450 that degrade different and complex substrates, including PAHs [43,44]. In addition, isolates from extreme environments have been proven to produce extracellular extremophilic ligninolytic enzymes, which can perform catalytic activities under adverse environments without being denatured [45]. These metabolic capacities and rapid mycelial propagation pose fungi as attractive prospects for bioremediation applications [46]. Notably, marinederived fungi have been recognized as promising bioremediation agents due to their metabolic adaptations to naturally occurring oil sources [47,48]. Yet, they have been scarcely studied for their ability to use recalcitrant elements of HCO and EHCO.

In this study, we showed the ability of *Alternaria* sp., *Penicillium* spp., and *Stemphylium* sp., isolates recovered from deep-sea sediments of the Mexican EEZ of the Gulf of Mexico to grow using HCO and, most importantly, the ability of *Alternaria* sp. to grow using EHCO as sole carbon source. Unfortunately, given the nature of the EHCO and the strong attachment of *Alternaria* sp. mycelium to the EHCO in the glass rod, the quantification of the biomass was not successfully attained. However, at the visible level, one could observe the mycelium of *Alternaria* sp. only in the cultures supplemented with EHCO (Figure S3). We concluded that the toluene added for protein quantification to get an indirect measurement of the fungal biomass probably interfered with the complete recovery of the mycelium, and the protein content obtained was not an accurate estimate of the actual amount of protein. Although the fungal taxa identified in this study have been previously reported as light and medium crude oil degraders, there are no previous reports of HCO or EHCO degradation by any of them. *Penicillium* spp. isolated from diverse polluted environments have been extensively reported as capable of removing a wide array of contaminants, including heavy metals and petroleum hydrocarbons [49,50]. *Penicillium* spp. can efficiently degrade polycyclic aromatic hydrocarbons such as fluorene [51], pyrene [52], and phenanthrene [53]. The *Stemphylium* sp. has been isolated from petroleum-contaminated soils in Iran [54] and from a Mediterranean marine site after an oil spill, where *C. halotolerans* was recovered as well [55]. *P. corylophilum* was isolated from ship-breaking yards of Bangladesh and was able to degrade petroleum hydrocarbon [56].

*Alternaria* spp. have been isolated from terrestrial contaminated zones with evidence of PAH degradation [57], from tar balls with a very limited ability to degrade crude oil [49], and from noncontaminated gardens showing the ability to degrade octane and decane [58]. *A. alternata* has the capacity to synthesize enzymes related to hydrocarbon degradation, such as manganese peroxidase and lignin peroxidase [59,60]. There is a comparative analysis of the secretome profile of Manganese (II)-Oxidizing [61], as well as a report of lignin peroxidase production [62], and this sets a precedent for the potential degradation capacity of *Alternaria* sp.

Heavy crude oils contain a high proportion of high molecular weight hydrocarbons and elevated levels of hetero compounds, including sulfur, nitrogen, oxygen, and heavy metals [63]. The typical SARA composition of heavy oils is 4–23% saturates, 14–32% aromatics, 27–46% resins, and 15–43% asphaltenes [64]. The high concentration of resins and asphaltenes, together with the high concentration of high molecular weight PAHs in EHCO, poses a challenge for biodegradation and bioremediation processes. The chemical nature of the products from hydrocarbon degradation was not analyzed in this study. The composition of crude oil includes several thousands of different hydrocarbons that could potentially represent a carbon source for *Alternaria* sp. The products resulting from microbial degradation by *Alternaria* are also a complex mixture of polar compounds hard to resolve by conventional analytical chemistry tools. However, hydrocarbon degradation products' chemical natures have been extensively studied [65,66]. Different enzymes

transform saturated and aromatic hydrocarbons, increasing the relative abundance of polar fractions. Various pathways, such as terminal oxidation, subterminal oxidation, ω-oxidation, and β-oxidation, are involved in the degradation of aliphatic hydrocarbons. The initial alkane oxidation results in the formation of an alcohol, which can then be metabolized by the β-oxidation pathway of fatty acids. On the other hand, fungi's initial oxidation of aromatic hydrocarbons is mediated by monooxygenases forming a trans-diol. The benzene ring is then cleaved in different ways by appropriate enzymes leading to the formation of central intermediates, which are further converted to tricarboxylic acid (TCA) cycle intermediates. Finally, cyclic alkanes are converted to cyclic alcohols and dehydrogenated to ketones by an oxidase system. Then, a monooxygenase system forms lactonates, and a lactone hydrolase finally opens the ring.

Our study proved that *Alternaria* sp. has a great affinity for the aromatic fraction of EHCO, showing an extensive transformation of 91.3% of this fraction. The aromatic fraction that was transformed by *Alternaria* sp. to more polar compounds migrated to the resin–asphaltene fraction in the degraded hydrocarbons, as suggested by the percentage increase of this fraction. Despite this increase, it has been previously well-established that the polar compounds appearing in the resin–asphaltene fraction as a result of the biodegradation of saturates and, especially, of the aromatic fraction, are more biodegradable than the original hydrocarbons, and they can be more accessible for degradation by other microorganisms [67,68].

The station N2 (Perdido region), from where *Alternaria* sp. was isolated, did not display any remarkable difference in salinity, temperature, or dissolved oxygen concentration compared with the rest of the stations (Supplementary Table S2). The Gulf of Mexico has a two-layer system; the upper layer (above 800–1000 m) is highly regulated by the Loop Current and its associated eddies (LCEs) entering the Gulf of Mexico through the Yucatan Channel [69]. The river discharges, the temperature variation, and the freshwater input do not impact the salinity or temperature beyond the upper 100 m of the water column [70]. Consequently, the main driver in the hydrographic conditions in the GoM where the sediments samples were taken was the local presence of LCEs [70]. Regardless of the wide distribution of the sites sampled in the Gulf of Mexico, it is not possible to recognize a difference between the Perdido and Coatzacoalcos Stations. Temperature and dissolved oxygen variation respond mainly to the depth of the sediments, and there is an insignificant variation in the salinity in all the stations.

Recent estimations report that unconventional oil reserves, which include HCO, EHCO, and bitumen, could exceed six trillion barrels [71]. The ineludible extraction of these reservoirs will inevitably produce oil spills. It is essential to explore native fungal microorganisms capable of degrading HCO and, especially, EHCO to mitigate the environmental impact of accidental petroleum spills. Offshore oil spills are of tremendous concern due to their potential impact on the economy and especially on ecological systems [72]. The successful bioremediation of marine oil spills has been reported [73]. Among the bioremediation strategies, biostimulation is an *in situ* technology that involves the introduction of indigenous (obtained from the contaminated site) or exogenous hydrocarbonoclastic microorganisms to a polluted site to enhance microbial degradation at the site. Many degradation studies are focused on water column degradation, and no attention has been given to sediment oil degradation. Oil biodegradation in sediments is significant for the heavy fractions of oil, which tend to settle after the oil spill. In addition, the sediment environmental conditions are different such as oxygen and nutrient concentrations. There are nevertheless some challenges associated with bioremediation of spilled petroleum [74], including: (i) the resistance of heavy fractions, especially asphaltenes and high molar mass polycyclic aromatic hydrocarbons, to degradation. Thus, it seems crucial to have microbial strains able to perform the biodegradation of heavy and extra-heavy oils such as *Alternaria* sp. isolated and characterized in this work; (ii) eutrophication caused by biostimulation. Seawater is characterized by the low nitrogen and phosphorous concentrations that are limiting factors for oil degradation. The use of oleophilic fertilizers seems to be an effec-

tive nutrient type to be applied in marine environments to prevent eutrophication; (iii) unsustainability of bioaugmentation in the field. This can be reduced using well-adapted microorganisms, like, for instance, those adapted to sediment conditions. In addition, the encapsulation of the microorganism can protect it in the new environmental conditions; (iv) poor bioavailability of spilled petroleum. Oil biodegradation takes place at the oil– water interphase. Therefore, the widely used oil dispersion or solubilization by adding chemical or biological surfactants enhances biodegradation. However, the chemical agents could be potentially toxic to the environment; and (v) oil biodegradation inefficiency in semi-anoxic environments. As mentioned before, in deep sediments, there is a low oxygen concentration. It is also important to have microorganisms isolated from similar sediment conditions, such as the *Alternaria* sp. isolated and characterized in this work.

Considering all the above, microorganisms that, like *Alternaria* sp., are capable of degrading some of the heavier and more recalcitrant components of these crude oils can be good candidates for further studies testing different environmental conditions in order to assess their use for bioremediation in the event of an oil spill.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/app11136090/s1: Table S1: Geolocalization and data of stations sampled in the Gulf of Mexico. Table S2: Sampling stations and locations from where the fungal strains were isolated. Table S3: Reference DNA sequences of the ITS rDNA, 18S and ß-tubulin markers used in the phylogenetic analyses. Figure S1: Protein content (average of three replicates) of *Penicillium* sp. 1 (A) and *Penicillium* sp. 2 (B) grown under different light crude oil concentrations. Figure S2: Chromatograms of whole extra-heavy crude oil and its saturates, aromatics and polars (SAP) fractions. Figure S3: *Alternaria* sp. grown with (B, D) or without (A, C) 0.5% w/v of extra-heavy crude oil as sole carbon source after a month of incubation.

**Author Contributions:** Conceptualization, M.R.; methodology, L.R.-H., M.D.C.-L., P.V. and I.B.-G.; investigation, L.R.-H.; writing—original draft preparation, L.R.-H.; writing—review and editing, M.R., P.V. and R.V.-D.; formal analysis, L.R.-H.; supervision, M.R. and R.V.-D.; project administration, M.R.; and funding acquisition, M.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Council of Science and Technology of Mexico-Mexican Ministry of Energy-Hydrocarbon Trust, project 201441. This is a contribution of the Gulf of Mexico Research Consortium (CIGoM). LRH was supported by CONACYT doctoral fellowship 259695.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data is contained within the article and supplementary material.

**Acknowledgments:** We thank Karla Sidon for collecting the samples and Lluvia Vargas-Gastélum for carefully reading the manuscript. Microscopy analyses were carried out at the National Laboratory of Advanced Microscopy of CICESE (LNMA) and the Department of Optics of CICESE. We are grateful to Diego Delgado, Fabian Cordero, Pilar Sánchez-Saavedra, and Juan Manuel Martínez for their technical support.

**Conflicts of Interest:** The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

#### **Abbreviations**


#### **References**


## *Brief Report* **The Presence of Marine Filamentous Fungi on a Copper-Based Antifouling Paint**

**Sergey Dobretsov 1,2,\*, Hanaa Al-Shibli 3, Sajeewa S. N. Maharachchikumbura <sup>4</sup> and Abdullah M. Al-Sadi <sup>3</sup>**


**Abstract:** Marine biofouling is undesirable growth on submerged substances, which causes a major problem for maritime industries. Antifouling paints containing toxic compounds such as copper are used to prevent marine biofouling. However, bacteria and diatoms are usually found in biofilms developed on such paints. In this study, plastic panels painted with a copper-based self-polishing antifouling paint were exposed to biofouling for 6 months in the Marina Bandar Rowdha, Sea of Oman. Clean panels were used as a control substratum. Marine filamentous fungi from protected and unprotected substrate were isolated on a potato dextrose agar. Pure isolates were identified using sequences of the ITS region of rDNA. Six fungal isolates (*Alternaria* sp., *Aspergillus niger*, *A. terreus*, *A. tubingensis*, *Cladosporium halotolerans*, and *C. omanense*) were obtained from the antifouling paint. Four isolates (*Aspergillus pseudodeflectus, C. omanense,* and *Parengyodontium album*) were isolated from clean panels and nylon ropes. This is the first evidence of the presence of marine fungi on antifouling paints. In comparison with isolates from the unprotected substrate, fungi from the antifouling paint were highly resistant to copper, which suggests that filamentous fungi can grow on marine antifouling paints.

**Keywords:** marine fungi; copper; antifouling; coating; biofilm; Indian Ocean; Oman

#### **1. Introduction**

Marine biofouling is defined as the "undesirable accumulation and growth of organisms on submerged surfaces" [1]. Usually, biofouling organisms are divided by their size onto microfouling and macrofouling. Microfouling is composed of microscopic (<0.5 mm) organisms, mainly bacteria and diatoms [2,3]. Macrofouling, on the other hand, is composed of macroscopic organisms (>0.5 mm) visible by the naked eye, such as barnacles, mussels, bryozoans, macroalgae and others [4–6]. Microfouling has a significant impact on the recruitment of spores and larvae of algae and invertebrates (reviewed by [4,7]).

Marine biofouling causes significant problems for maritime industries [8,9]. It can increase the fuel consumption of ships, clog membranes and pipes, increase corrosion, decrease buoyancy, and destroy nets and cages [10,11]. Countries worldwide spend more than USD 7 billion per year in order to protect from biofouling and deal with is consequences [9].

In order to prevent submerged structures like boats and ships from biofouling companies are using antifouling coatings [9,12]. These antifouling coatings usually contain biocides that kill biofouling organisms. Currently, the most effective biocide is copper or cuprous oxide [12].

**Citation:** Dobretsov, S.; Al-Shibli, H.; Maharachchikumbura, S.S.N.; Al-Sadi, A.M. The Presence of Marine Filamentous Fungi on a Copper-Based Antifouling Paint. *Appl. Sci.* **2021**, *11*, 8277. https://doi.org/ 10.3390/app11188277

Academic Editors: Anna Poli and Valeria Prigione

Received: 24 June 2021 Accepted: 20 August 2021 Published: 7 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

While antifouling paints are supposed to prevent biofouling, most of them have biofilms on their surfaces [3,13,14]. There is limited information about the biofilm composition of antifouling paints (see [15,16]). It has been shown that biofilms on antifouling paints consist of diverse species of bacteria and diatoms [17–19]. Up to now, filamentous fungi on antifouling paints have not been observed. Previous studies showed that biocides of antifouling paints and environmental conditions shaped the structure of the microbial communities [20,21].

Marine fungi are an important component of the marine environment [22,23]. While marine filamentous fungi are not very well studied, they are widely distributed and associated with sediments, sand grains, seaweeds, submerged wood, sea animals and plants [24]. Chytidiomycota and Ascomycota are fungal divisions dominated in samples from six European near-shore sites [22]. Forty-six fungal isolates belonging to the genera *Cladosporium, Paraphaeosphaeria, Trichoderma, Alternaria, Phoma*, and *Arthrinium* were isolated from marine biofilms developed on different submerged substrata [25]. Chytidiomycetes fungi dominated in marine biofilms developed on glass and plastic substrates [26]. To our knowledge, marine fungi have never been recorded on antifouling paints, especially copper-based ones. However, it was observed using metabarcoding that most of the fungal species in marine periphyton biofilms were not affected by 10 μM of copper [27]. Some species of marine filamentous fungi are highly resistant to high concentrations of copper. For example, *Penicillium chrysogenum* was able to tolerate concentrations of 500 mg L−<sup>1</sup> of copper [28].

The main aim of this study was to identify the fungal isolates from biofilms developed on the surface of a cooper-based antifouling paint and demonstrate their copper resistance.

#### **2. Materials and Methods**

#### *2.1. Antifouling Paint and Other Substrata*

A commercial copper self-polishing paint Interspeed® BRA640 (International Paint, Gateshead, UK) was used in this study. The antifouling paint contains about 25–50% of cuprous oxide by weight [29]. An average release rate of copper from the paint was 3.8 μg cm−<sup>2</sup> day−<sup>1</sup> [30]. The paint was manually applied (thickness 125 μm) onto plastic fiberglass panels (15 cm × 28 cm) cleaned with ethanol (96%, Sigma, Ronkonkoma, NY, USA) in the laboratory. Fiberglass was obtained from a local Omani manufacturer (Al Kaboura, Muscat, Oman). This material was selected because it is used to make boats and it has a high biofouling potential. All coated panels were air-dried for several days at ambient temperature prior to deployment. All substrates (panels and ropes) were cleaned with 96% ethanol before the experiment to eliminate bacteria and fungi. No fungi were found on these substrates prior to the experiment.

#### *2.2. Experiment and Testing Site*

Three panels covered with the antifouling paint were exposed vertically to biofouling at the depth of 1 m for 6 months in Marina Bandar Rowdha (23,035 07 N 58,036 48 E), Muscat, Oman. As a control, uncoated fiberglass panels were used. Panels were fixed at the desired depth using a nylon rope (RopeNet, Taishan, China) attached to a pontoon. At the end of the rope, a weight was attached to keep the panels in a vertical position.

Marina Bandar Rowdha is a semi-enclosed bay for private recreational boats and yachts. It has a relatively high hydrocarbon and heavy metal pollution with one of the highest concentrations of TBT in Oman's waters [31]. This marina was selected due to its (very high) biofouling rates and a history of biofouling and antifouling investigations [21,32]. The experiment in the marina was conducted in 2018 between the months of February and September. During the study the seawater temperature varied from 24 to 30 ◦C, pH was about 8.2 and the salinity varied from 37 to 38 ppt. During the experiment, the seawater turbidity was 2–3 NTU (Nephelometric Turbidity Units).

#### *2.3. Isolation of Fungi*

Biofouled panels (painted and not) and ropes holding panels were collected from the marina in September 2018. At the marina, the ropes and the panels were individually packed into sterile bags and brought on ice to the laboratory. In the laboratory, the panels and ropes were washed several times with sterile distilled water (SDI). Using sterile scissors, the ropes were cut into 1.0 cm pieces. Surfaces of the panels and the ropes were disinfected using 1% sodium hypochlorite solution to eliminate bacteria (NaClO, Zhengzhou Sino Chemical Ltd., Beijing, China). Then, the panels and the ropes were washed three times with SDI. After that, the biofilms were removed from the panels using sterile cotton swabs. Finally, one piece from each rope or an individual swab was placed into a Petri dish containing a 2.5% potato dextrose agar (PDA, Merck, Kenilworth, NJ, USA) prepared using filtered (0.45 μm cellulose nitrate filter, Sartorius, Germany) and autoclaved seawater from the marina. As a control, PDA Petri dishes containing autoclaved seawater were used. Visible growth of fungi was checked after incubation at 25 ◦C for up to three weeks. Each individual fungal colony was transferred into a new fresh PDA plate. Pure fungal colonies were stored on PDA slants with 10% glycerol for further genetic identification (see below).

#### *2.4. Identification of Fungi*

Before the identification, the isolate was grown on PDA. The identification of filamentous fungal isolates was done based on sequences of the internal transcribed spacer region (ITS) of the ribosomal DNA [33]. Firstly, 80 g of fungal mycelia were harvested and freeze-dried. Then, its DNA was further extracted [34]. Secondly, the ITS rDNA region was amplified using the primer pairs of ITS4 (TCCTCCGCTTATTGATATGC) and ITS5 (GGAAGTAAAAGT CGTAACAAGG). The PCR program followed the conditions of [35]. MACROGEN, Korea sequenced the PCR products. In order to obtain the ITS rRNA sequence, two complementary sequences for each fungal isolate were aligned using MEGA v.6 [36]. Fungal isolates were identified based on a comparison of the ITS rRNA sequences against the National Center for Biotechnology Information (NCBI) database. Sequences of fungal isolates were deposited in the NCBI GenBank database with accession numbers MN947598–MN947607. For phylogenetic trees, maximum likelihood analysis with 1000 bootstrap replicates based on ITS sequence data was done using RaxmlGUI v. 1.3 [37]. The final phylogenetic tree was selected by comparing the likelihood scores using the GTR+GAMMA substitution model.

#### *2.5. Copper Resistance of Fungal Isolates*

In order to prove that fungal isolates are able to grow on antifouling paints, we tested their sensitivity to copper by an agar diffusion technique. Because copper oxide is not soluble in water, copper sulfate (CuSO4, Sigma Aldrich, Ronkonkoma, NY, USA) was used. Firstly, different concentrations (500–0.01 g L−1) of copper sulfate in autoclaved seawater were prepared. Secondly, fungal isolates were grown onto 2.5% potato dextrose agar (PDA, Merck, Kenilworth, NJ, USA) for 3 days. PDA was made using autoclaved seawater from the marina. A four mm disc of each fungal isolate was cut and individually placed onto the PDA Petri dish. Thirdly, 10 μL of copper sulfate solution was added to a sterile paper disk (diameter 6 mm). As a control, disks with 10 μL of seawater were used. The disks were air-dried at room temperature and placed in the middle of a PDA Petri dish two cm away from the isolate. The dishes were incubated at 25 ◦C for 5 days. The experiment was made in triplicate. The presence or absence of an inhibition zone was detected. Finally, the minimal inhibitory concentration of copper (II) sulfate (μg cm−2) for each fungal isolate was calculated.

#### **3. Results and Discussion**

#### *3.1. Biofouling on Different Substrata*

Biofouling on the antifouling paint was minimal and only biofilms were observed. In opposite, the ropes and unprotected panels were completely covered with macrofouling organisms, dominated by Tunicata and Bryozoa. This supports our previous data about performance of different antifouling paints in Oman waters [21,32]. The 1-year field experiment showed that copper-based antifouling paints have only diatom and bacterial biofouling [21]. A previous experiment with unprotected fiberglass and acrylic panels demonstrated dominance of Bryozoa, barnacles and sponges [38]. The absence of sponges and barnacles in the current study could be due to differences in the substratum chemistry (nylon versus acrylic) and shape (flat plates versus cylindrical ropes).

#### *3.2. Species of Fungi Isolated from Different Substrata*

In total, six fungal isolates were obtained from the antifouling paint and four were isolated from unprotected substrata (Table 1). Based on the phylogenetic analysis, the majority of isolates belonged to *Aspergillus* and *Cladosporium* genera (Supplement Figures S1–S4). The genera *Aspergillus* and *Cladosporium* are commonly found in the marine environment [39–41]. Additionally, *Aspergillus* and *Cladosporium* are associated with marine sponges [42,43]. There is limited information about marine derived filamentous fungi in Oman, but we have been able to isolate *Aspergillus terreus* from mangrove areas [44]. *Cladosporium omanense* found in this study (Table 1) was previously isolated from living leaves of *Zygophyllum coccineum* in Oman [45]. The presence of *C. omanense* on all investigated substrates could be due to several reasons. It could suggest that this species is very common in Omani waters and can colonize protected and unprotected substrata. Alternatively, it could be due to contamination of our culture by spores of this fungus. This is highly unlikely, as there were no fungi recovered from the control plates with autoclaved seawater.

Filamentous fungi belonging to the genera *Parengyodontium* were isolated from biofouled ropes only (Table 1). *Parengyodontium album* is an environmental saprobic mold and an opportunistic pathogen [46]. This species has been observed on buildings composed of limestone and plaster [47]. Additionally, *P. album* was found in sediments of polar-boreal White Sea [48]. The presence of this fungus in Oman waters suggests that this species can be found in tropical waters as well.

The genera *Aspergillus, Cladosporium* and *Alternaria* were found on the copper-based antifouling paint (Table 1). Moreover, *A. tubingensis*, *A. terreus*, *A. niger* and *C. halotolerans* were found only on the antifouling paint. *Alternaria* isolates were obtained exclusively from the paint. Previously, the fungi *Alternaria* were isolated from soft corals [49], sponges [50] and algae [51]. While 18S RNA of fungi belonging to the class Agaricomycetes was detected on an antifouling paint using Illumina amplicon sequencing [52], fungal isolates were obtained from antifouling paints for the first time in this study. Previously, only bacteria and diatoms were detected in biofilms on antifouling paints [3,21].

#### *3.3. Copper Resistance of Fungal Isolates*

In order to prove that fungal isolates are able to grow on antifouling paints, their sensitivity to different copper concentrations is tested in laboratory experiments (Table 2). Due to low solubility of CuO, CuSO4 was used in this experiment. Previous studies suggest that CuSO4 is more toxic compare to CuO [53]. Thus, the isolates are more resistant to CuO than is reported in Table 2. Generally, isolates from antifouling paint can tolerate higher concentrations of copper. Five out of six isolates from the antifouling paint can tolerate an average daily release rate of copper 3.8 μg cm−<sup>2</sup> day−<sup>1</sup> [30] from the tested paint (Table 2). The highest copper resistance was observed for *Aspergillus terreus*. This fungus can tolerate 2% of CuCl2 in a polyvinyl chloride coating in a laboratory experiment [54] and can be used to remove heavy metals from water [54]. In opposite, fungal isolates from unprotected substrata had low tolerance to copper (Table 2). This suggests that isolates

from the copper-based antifouling paint are adapted to high copper concentrations and could grow and play an important role in biofilms.

**Table 1.** The list of fungal isolates from panels painted with the antifouling paint and not painted (control) panels and ropes.



**Table 2.** The minimal inhibitory concentration of copper (II) sulfate (μg cm−2) for fungal isolates from panels painted with the antifouling paint, not painted (control), and ropes. Highlighted values exceed an average release rate of copper from the paint (3.8 μg cm−<sup>2</sup> day−<sup>1</sup> [30]).

#### *3.4. Importance of This Study*

Our finding has very important implications for antifouling industries. Firstly, it demonstrates that some fungal species can live on antifouling paints and tolerate relatively high copper concentrations. Compared to the isolates from unprotected substrata (ropes and panels), fungi from the antifouling paint were highly resistant to copper. Previous studies suggested copper resistance of some fungal species that bind copper to cell walls [55,56]. Additionally, fungi can produce copper-binding proteins and chelating compounds in response to elevated concentrations of copper [55]. Secondly, the role of fungal species on antifouling paints requires further investigations. It is possible to propose that filamentous fungi can degrade organic matrix of the paint, composed of vinyl or acrylic resin or silicone polymers, which in turn can affect release of the biocide and the life span of the antifouling paint. It has been shown that filamentous fungi can deteriorate synthetic paints [57,58]. Additionally, filamentous fungi can degrade biocides of antifouling paints, such as Irgarol 1051 [59] and TBT [60]. In opposite, some marine fungal species produce antimicrobial and antifouling compounds (see review [61]). Presence of these strains on paints could be beneficial and enhance their antifouling properties. Finally, more research is needed to understand if marine fungi can be found on other antifouling paints exposed to biofouling in different seas and investigate possible mechanisms whereby fungi transform these paints. Additionally, it is important to investigate the role of marine fungi on antifouling paints and possible mechanisms of their resistance.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/app11188277/s1, Figure S1: Phylogram generated from maximum likelihood analysis based on ITS sequence data of analyzed *Alternaria* species. Isolates derived from this study are in red. The tree is rooted to *A. alternantherae* (CBS124392), Figure S2: Phylogram generated from maximum likelihood analysis based on ITS sequence data of analyzed *Aspergillus* species. Isolates derived from this study are in red. The tree is rooted to *Penicillium herquei* (CBS 336.48), Figure S3: Phylogram generated from maximum likelihood analysis based on ITS sequence data of analyzed *Cladosporium* species. Isolates derived from this study are in red, Figure S4: The tree is rooted to *Cercospora beticola* (CBS 116456), Phylogram generated from maximum likelihood analysis based on ITS sequence data of analyzed *Parengyodontium* species. Isolates derived from this study are in red. The tree is rooted to *Purpureocillium lilacinum* (CBS 284.36).

**Author Contributions:** Conceptualization, S.D. and A.M.A.-S.; methodology, S.D. and A.M.A.-S.; formal analysis, H.A.-S., S.D. and S.S.N.M.; investigation, H.A.-S. and S.S.N.M.; writing—original draft preparation, S.D. and A.M.A.-S.; writing—review and editing, S.D., A.M.A.-S., H.A.-S. and S.S.N.M.; supervision, S.D. and A.M.A.-S.; project administration, S.D. and A.M.A.-S.; funding acquisition, S.D. and A.M.A.-S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the TRC grant RC/AGR/FISH/16/01, Omantel grant EG/SQU-OT/20/01, SQU internal grant IG/AGR/FISH/18/01 and the grant EG/AGR/CROP/16/01.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Sequences of fungal isolates can be found in the NCBI GenBank database with accession numbers MN947598–MN947607.

**Acknowledgments:** S.D. acknowledged financial support by the TRC grant RC/AGR/FISH/16/01 and SQU internal grant IG/AGR/FISH/18/01. A.M.A. research was supported by the grant EG/AGR/CROP/16/01 and RC/AGR/FISH/16/01.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Review* **Facing Phototrophic Microorganisms That Colonize Artistic Fountains and Other Wet Stone Surfaces: Identification Keys**

**Fernando Bolivar-Galiano 1, Oana Adriana Cuzman 2,3,\*, Clara Abad-Ruiz <sup>1</sup> and Pedro Sánchez-Castillo <sup>4</sup>**


**Featured Application: This work presents a helpful tool when dealing with the biodiversity of monumental fountains and other wet lithotype surfaces in the sector of stone conservation.**

**Abstract:** All fountains are inhabited by phototrophic microorganisms, especially if they are functional and located outdoors. This fact, along with the regular presence of water and the intrinsic bioreceptivity of stone material, easily favors the biological development. Many of these organisms are responsible for the biodeterioration phenomena and recognizing them could help to define the best strategies for the conservation and maintenance of monumental fountains. The presence of biological growth involves different activities for the conservation of artistic fountains. This paper is a review of the phototrophic biodiversity reported in 46 fountains and gives a whole vision on coping with biodeteriogens of fountains, being an elementary guide for professionals in the field of stone conservation. It is focused on recognizing the main phototrophs by using simplified dichotomous keys for cyanobacteria, green algae and diatoms. Some basic issues related to the handling of the samples and with the control of these types of microalgae are also briefly described, in order to assist interested professionals when dealing with the biodiversity of monumental fountains.

**Keywords:** green algae; diatoms; cyanobacteria; microalgae; stone conservation; diagnosis tool; preservation strategies; biodeterioration

#### **1. Introduction**

Fountains are structures with a functional, decorative or recreation purpose that have been built since ancient times [1]. Monumental and artistic fountains are often impressive constructions built all over the world, belonging to the public and garden art, such as 'La Joute' Fountain from Montreal (Canada), Fountain of Giant Wild Goose Pagoda from Xi'an (China), Flora Fountain from Mumbai (India), Keller Fountain from Portland, Oregon (USA), etc. They are mainly made of stone (structure, basin and decorations) and metallic alloys (hydraulic system and decorations). Stones with a local origin are preferred, but valuable stones are also used for the significant decorative elements (e.g., marble). However, other types of materials have been employed in the recent century, such as concrete material (e.g., Villancourt Fountain, San Francisco, CA, USA) or other modern materials such resins (Stravinsky Fountain, Paris, France, or Charybdis' Vortex Fountain, Sunderland, UK).

The roughness of the stone substrate is one of the main factors that influences the phototrophic colonization by epilithic, chasmolithic or endolithic microorganisms [2]. The last ones include the criptoendolithic microorganisms (which develop in layers, parallel with respect to the surface) and euendolithic microorganisms (which are the most aggressive, as they actively penetrate inside the stone). The epilithic communities are the most

**Citation:** Bolivar-Galiano, F.; Cuzman, O.A.; Abad-Ruiz, C.; Sánchez-Castillo, P. Facing Phototrophic Microorganisms That Colonize Artistic Fountains and Other Wet Stone Surfaces: Identification Keys. *Appl. Sci.* **2021**, *11*, 8787. https://doi.org/10.3390/app11188787

Academic Editors: Cédric Delattre, Anna Poli and Valeria Prigione

Received: 9 July 2021 Accepted: 13 September 2021 Published: 21 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

frequent. They mainly induce esthetical damage, but often corrosion phenomena can be observed (like little holes in the stones, corresponding to the phototrophic settlement). The chasmolithic development induces damage such as detachments or stone flakes formation and lifting. The biodeterioration phenomena were investigated in different studies [3–7].

The main types of phototrophic communities form pellicles, pustules, mats or layered sedimentary formations [6]. Often, the biological presence of phototrophic microorganisms can be considered a patina [8] or a phototrophic biofilm (due to its mucilaginous aspect).

Artistic fountains are not only heritage items but sociocultural artefacts as well, with a significant importance for the identity of certain places or periods [9,10]. They are part of the community and human treasures and raise preservation demands for transmitting their values to the further generations. Applying monitoring activity with a professional maintenance plan is essential for good and effective preservation of these structures. The preservation strategies should include different activities such as cleaning, repairs and consolidation, application or replacement of protective coatings, controlling of the water quality; all these interventions are related with the degradation problems induced by physical, chemical or biological agents. Coping with the phototrophic colonizers of artistic fountains and other wet surfaces means mainly cleaning and monitoring activity. The morphological identification of the phototrophic presence is usually enough for the documentation before and after the restoration. A detailed mapping of all types of deterioration phenomena present on a certain case study helps to decide the best strategies for both treatment and monitoring plans. Although the control treatments are often common for all the phototrophic groups, it worth mentioning that the principle of minimal intervention should always be kept in mind, being therefore ready to undertake, when necessary, selective treatments on the same stone artefact. This is due especially to two main reasons: (i) the phototrophic growth on the fountains is rarely formed by a singular group of phototrophs, and (ii) the cleaning treatments (chemical, physical or mechanical) are not so selective. However, in particular cases (e.g., the presence of phototrophic endolithic growth) it is possible to improve the cleaning procedures by making changes to the applications time, the concentration used, or to combine different types of cleaning treatments (e.g., mechanical and chemical; mechanical and physical). Moreover, the molecular identification could be necessary despite increasing procedure costs. These types of analysis need a small amount of sample and are definitely more precise than the morphological analysis. They are very useful to control the regrowth of dangerous species after a certain treatment.

An easy to use guide containing identification keys for the main phototrophs growing on monumental fountains and other wet stone surfaces is presented in this work. These new keys resulted after revising and simplifying other different identification keys [11–21] and they were tailored as a new tool for professionals in the field of stone conservation, that are interested in making their first steps in phototrophs identification. It is based on both macroscopic and microscopic characteristics for the main genera belonging to cyanobacteria, green algae and diatoms, the main three phototrophic groups found on the ornamental fountains. Moreover, some basic issues related to the handling of the samples and the control methodologies were also briefly described.

#### **2. Materials and Methods**

#### *2.1. Morphological Identification*

Phototrophic samples (about 265) were collected from various ornamental fountains and identified by the authors of this paper in precedent works [21–26] as indicated in Tables 1–4. The most frequent genera found by other authors [3,5,7,24,25,27–32] are also inserted in these identification keys.


**Table 1.** Artistic fountains investigated for the main phototrophic microorganisms.



**Table 3.** Green algae observed in different artistic fountains (see Table 3 for details of the location), where CO = Columbia, IT = Italy, IN = India, SK = Slovakia,

ES =

x

x

x x


**Table 4.** Diatoms observed in different artistic fountains (see Table 1 for details of the location), where CO = Columbia, IT = Italy, IN = India, ES = Spain and USA = United States.

> The collection of the phototrophic biomass is usually made by using sterile scalpels and tubes of different sizes. When possible, in the case of chasmolithic or endolithic developments, small pieces of stone substrata containing microorganisms can be sampled as well. A fixation with formalin (2%, final concentration) is recommended in order to preserve the morphology of the phototrophs [23]. The Lugol's iodine solution (1% *v*/*v*) is also used for preventing contamination [5], being also a useful test under the microscope, as it gives a dark purple/brown/blue black stain to the cells containing starch (green algae and diatoms), favoring the sedimentation of the cells and their easier observation as well [33].

> The microscopic observations of the samples can be made in transmission light, and good quality images can be obtained by using DIC (differential interference contrast) microscopy. The fresh samples will often contain more than one genus. The sample preparation for the microscopic analysis can be summarized as following: (i) first of all, place a drop of water on the center of a clean slide; (ii) take a very small part of the sample (e.g., with the help of a needle or tweezers) and place it inside the drop of water; if the sample is solid or too thick (e.g., a dense mat or a crust) it must be crushed and scattered in the water drop in order to be able distinguish individual cells by the light passing through the sample; it is important to not mechanically break the clusters before the slide preparation, as confusion may occur with the non-colonial genera; (iii) once the sample is uniformly distributed, place slowly the cover on top by firstly putting in contact its outer edge with the water drop. If the water overflows, gently remove the excess with a paper towel. When diatoms are present, a cleaning procedure of the frustules can be used before preparing the slides [34,35]. This procedure involves a careful use of some acids (nitric acid, hydrogen peroxide), able to remove the organic matter and carbonates from the sample, and therefore to highlight the morphological characteristics of the diatoms. Once prepared, the slides can be observed under different magnification levels (10×, 20×, 40×, 50× or 100×) in order to distinguish many of the morphological features of the cells. Some of the main features that would be observed are related to the shape of the cells, the presence or absence of the sheaths, the division type, the end cells morphology in the case of the filamentous cyanobacteria, the presence of motility phenomena, the plastids position and their shape in the case of green algae. It is possible to make deeper investigations by evaluating the ratio within the main phototrophic groups, as Popovi´c et al. [36] describe a detailed procedure for the mixed phototrophic biofilms.

#### *2.2. Identification Keys*

The keys derive from the specific literature survey on the taxonomy of cyanobacteria [11–19], green algae [11,13] and diatoms [11,13,21] and from the authors' collected material. The keys are dichotomous. Each number describes a couple of contrasting and distinctive characteristics. The reader makes a choice between the two options and follows the indications for the next couplet and so on. The dichotomous chain finishes with finding the genus. It should be specified that many species of cyanobacteria exhibit variable morphology, and their identification till the species level may need specific ultrastructural and molecular features, a fact that it was not considered necessary for the aim of this work and therefore they were not used in this key.

Five plates containing the illustrations of the main common genera were realized in watercolor technique (Arches cotton paper 300 g/m3, Sennelier and Mijello water colors) by F. Bolívar-Galiano, according mainly to our microscopic observations.

#### **3. Results**

#### *3.1. First Screening*

The first screening can be undertaken by naked eye or by using a portable microscope with about 20× magnification which can help to choose the sampling area and to make a rough morphological discrimination, through the inspection of the color and morphological aspects of the biological colonization (mat aspect with visible filaments, patina aspect with uniform color, spots like growth, etc.). A good camera with a macro objective can also be very useful.

In this step, it is difficult to assign the sample to a single phototrophic group, as on monumental fountains and other wet stone surfaces there are developing communities of microorganisms, which are a mix of different cyanobacteria and/or green algae and/or diatoms, as can be seen in some fresh samples observed under the microscope (Supplementary Materials S1–S5). Diatoms prefer high levels of humidity and are more often associated with green algae, and less only with cyanobacteria, but they can be found beside algae and cyanobacteria as well. Often, the color gives an idea about the possible phototrophic presence, but it is never conclusive [37].


#### *3.2. Cyanobacteria (Blue-Green Algae)*

They are prokaryotic phototrophic microorganisms, without chloroplasts (plastids), with an immense morphological variability, from free and isolated cells to very hard and consistent assemblages. Moreover, the Lugol solution used for sample preservation, does not stain the cells blue/black. Cyanobacteria have a characteristic blue-green color, hence their common name of blue-green algae. They can be divided in two main groups: I. simple cyanobacteria and II. filamentous cyanobacteria.

I. simple cyanobacteria (unicellular, mucilaginous groups, strong adherent group of cells) (Figures 1 and S6 and [23,38])

**Figure 1.** Some simple cyanobacteria (blue-green algae): (**A**) *Gloeothece* (4a), (**B**) *Gloeocapsa* (5a), (**C**) *Chroococcus* (5b), (**D**) *Microcystis* (7a), (**E**) *Aphanocapsa* (7b), (**F**) *Aphanothece* (8a), (**G**) *Chlorogloea* (9a), (**H**) *Chroococcidiopsis* (9b), (**I**) *Cyanosarcina* (12a), (**J**) *Myxosarcina* (12b). Scale bar 10 μm. The number in the brackets corresponds to the one indicated in the identification key.


II. Filamentous cyanobacteria (linear groups of cells called trichomes, with one or more rows, surrounded or not by sheaths) (Figures 2, S7 and S8 and [23,38])

**Figure 2.** Some filamentous cyanobacteria (blue-green algae): (**A**) *Chamaesiphon* (2a), (**B**) *Pleurocapsa* (2b), (**C**) *Oscillatoria* (5a), (**D**) *Nostoc* (6a), (**E**) *Calothrix* (9a), (**F**) *Tolypothrix* (10b), (**G**) *Symploca* (13b), (**H**) *Schizothrix* (14a), (**I**) *Phormidium* (16a), (**J**) *Microcoleus* (17a). Scale bar 10 μm. The number in the brackets corresponds to the one indicated in the identification key.


#### *3.3. Green Algae (Chlorophyta)*

This division contains very different organisms regarding the morphological, dimensional and reproductive aspects. They have a typical vivid light green color, but the cells with starch will stain dark purple/brown/blue black in Lugol's iodine solution (if used for preserving the sample). They contain chloroplasts (plastids) and can be divided into two main groups: I. simple green algae (and green flagellate) and II. filamentous green algae.

I. Simple green algae (and green flagellate Euglena) (Figures 3 and S9 and [21,23,38])

**Figure 3.** Some simple green algae: (**A**) *Euglena* (3b), (**B**) *Chlamydomonas* (4a), (**C**) *Haematococcus* (4b), (**D**) *Oocystis* (8a), (**E**) *Cosmarium* (12a), (**F**) *Chlorococcum* (13b), (**G**) *Scenedesmus* (17b), (**H**) *Chlorosarcinopsis* (19a), (**I**) *Tetracystis* (19b), (**J**) *Palmella* (20a). Scale bar 10 μm. The number in the brackets corresponds to the one indicated in the identification key.


**Figure 4.** Some filamentous green algae from the C2 group: (**A**) *Spirogyra* (2a), (**B**) *Oedogonium* (3a), (**C**) *Klebsormidium* (4b), (**D**) *Ulothrix* (4a), (**E**) *Apatococcus* (5a), (**F**) *Cladophora* (6a), (**G**) *Gongrosira* (7a), (**H**) *Leptosira* (8a), (**I**) *Dilabifilum* (9b), (**J**) *Pleurastrum* (9b). Scale bar 10 μm. The number in the brackets corresponds to the one indicated in the identification key.


#### *3.4. Diatoms (Bacillariophyta)*

They are unicellular reddish brown or golden algae (Figures 5 and S10 and [23,38]). The specific characteristic of this group is the presence of an external layer (the cell wall) made of silica that is called frustule. They are common on permanent wetted surfaces, and on monumental fountains the most encountered diatoms are those with bisymmetrical or monosymmetrical valves, one overlapping the other. The diatoms with radial symmetry are less common.

**Figure 5.** Some diatoms: (**A**) *Diatoma* (2a), (**B**) *Melosira* (2b), (**C**) *Navicula* (8a), (**D**) *Gomphonema* (9b), (**E**) *Cymbella* (10a), (**F**) *Amphora* (10b), (**G**) *Surirella* (11b), (**H**) *Nitzschia* (12a), (**I**) *Epithemia* (12b), (**J**) *Achnanthes* (13a). Scale bar 10 μm. The number in the brackets corresponds to the one indicated in the identification key.


#### **4. Discussion**

*4.1. Fountains' Phototrophic Biodiversity*

Fountains are ideal environments for the rapid development of microorganisms, which induce the occurrence of biodeterioration processes and algal blooms that should be controlled in the case of artistic structures. Besides the bioreceptivity of the material and the characteristics of the environmental factors, the water origin is another element that highly influences the biodiversity of the colonizers on monumental fountains. The water source used for their alimentation can be driven from a public water line (e.g., mainly in the city centers) or can be directly supplied from a nearby river (e.g., in some gardens) [4,7,8,21,23,24,27]. The latter type of water source induces a greater biodeterioration risk for the monumental fountains.

A review of the main phototrophic microorganisms detected by various authors on this kind of artefact (Table 1) is presented in the Tables 2–4. Table 1 contains the

legend of Tables 1–3. The microorganisms are divided in the following three groups: cyanobacteria (Table 2), green algae (Table 3) and diatoms (Table 4) and most of them are included in the identification keys. Very few studies have been undertaken on the molecular identification of the phototrophs dwelling on cultural fountain structures [23,27]. The identification to the species level needs axenic cultures of the isolated organisms, increasing the complexity of these types of analyses. Moreover, many of these phototrophs (mainly cyanobacteria) have a slow growth and their isolation is very time consuming. Thanks to the recent developments in the molecular biology domain, new techniques such as NGS (next-generation sequencing, called also high throughput sequencing—HTS) and metagenomics approaches give now valid and cost-effective solutions in the cultural heritage field [39,40]. The latter technique, called also community genomics, contributes with new insights for a better understanding of both the role and the structure of the microbial communities colonizing stone artefacts. However, these studies [23,27] confirmed the presence of various cosmopolitan phototrophic microorganisms with wide tolerance ranges concerning, for example, pH and temperatures. Many of these common genera (such as *Calothrix*, *Phormidium*, *Oscillatoria*, *Cosmarium*, *Scenedesmus*, *Ulothrix*, *Navicula*, *Nitzchia*, etc.) can be easily identified by their morphological characteristics by using fresh samples, a transmission microscope and an easy to use identification key. It should be noted that some genera can be easily confounded in between [41,42] and therefore it is important to pay attention to the minimal morphological differences (e.g., *Gloeocapsa* with *Aphanocapsa*, when the *Gloeocapsa* individual sheath cannot be distinguished in a colony; *Aphanocapsa* with *Microcystis*, as *Aphanocapsa* can be round-shaped as well; *Cyanosarcina* with *Myxossarcina* as the shapes of the single colonies in the agglomeration of cells are not always very clear; some small sized diatoms can be confused with *Navicula* or *Nitzschia*).

The phototrophic microorganisms form different types of biocenosis (biotic communities) and it can be stated that, almost always, the phototrophic growth will be a mix of different genera as can be usually observed when analyzing fresh samples (Figures S1–S5 Supplementary Materials), containing even other types of organisms such as bacteria or fungi. The stone support induces a certain bioreceptivity for the phototrophic colonization. Stone materials with higher porosity (travertine—fountain no. 7; sandstones—fountain no.3; cementitious materials—fountains no.4 and no.6; volcanic rocks—fountain no.1, see Table 1) are easily colonized by the phototrophic organisms with respect to the more compact stone materials (granite—fountain no. 5; marble—fountains no.5, no. 6, no.7 and no.15). The limestones and sandstones are more sensitive to biodeterioration with respect to granitic stones [43]. The common genera observed in fountains worldwide can have an epilithic growth forming mucous (*Apatococcus*, *Phormidium*) or fibrous mats (*Cladophora*, *Gongrosira*, *Melosira, Pleurastrum*). The existing fissures and cracks are colonized by chasmo-endoliths (mainly simple cyanobacteria and green algae) while the internal pores are colonized by cryptoendoliths (*Chroococcidiopsis*). Some genera (*Chroococcidiopsis*, *Chroococcus*, *Gloeocapsa*, *Klebsormidium*, *Schizothrix*, *Synechococcus*) can have an endolithic growth, actively penetrating inside the material, being therefore more harmful for the stone support [44,45].

Green algae and cyanobacteria present a greater diversity compared to diatoms when the number of most frequent genera in the three groups is analyzed (Tables 2–4). On the studied fountains, *Navicula* spp. and *Nitzschia* spp. are the most prevalent. Among cyanobacteria, the filamentous type, such as *Calothrix* spp., *Leptolyngbya* spp. and *Phormidium* spp., is the most common. These blue-green algae are able to form mats on their own or to be part of more diverse communities. The most common genera of green algae seem to vary more between countries than other groups of algae. Nevertheless, they are also cosmopolitan. *Apatococcus*, *Chlorella*, *Chlorosarcinopsis*, *Stichococcus* and *Scenedesmus* are the most frequent genera among all the countries. However, putting some differences in the distribution of the genera aside, it is apparent that the same types of microalgae colonize fountains all around the world. Therefore, a key that allows to identify them could be used internationally by restorers from any country.

Knowing the main types of phototrophic groups present on the monumental fountains helps to better understand the risk related to the biologically induced biodeterioration phenomena.

#### *4.2. Phototrophic Control on the Fountains*

Fountains need regular maintenance and restoration work [46] due both to material alterations and to the rapid colonization by phototrophic microorganisms that affects their esthetic value and induces biodeterioration processes.

Parameters such as the pH (an alkaline pH accelerates the algal growth, while an acidic one dissolves carbonatic stones and induces copper staining on stone), the calcium hardness (induces crust formations), the dissolved solids (organic and inorganic, favor biological growth and the clogging of pipes) and the temperature (air and water) are key elements in promoting or not the biodeterioration phenomena. Often, the water is recycled and chemically treated to avoid biological growth [47], but this only helps to postpone the biological development.

Different other strategies [48] can be used in controlling the biological development in the monumental fountains. One is related with the use of antifouling agents aiming to deter the biofilm formation [49–51]. They should be applied on cleaned surfaces in order to postpone biological colonization. These methods are still under investigations for defining practical effectiveness. Another strategy is related to the elimination of the biological growth with the help of biocides [52,53] and other chemical cleaners, such as chlorine. A new trend in controlling the unwanted phototrophic growth is the use of viruses [54]. However, the implementation of a management/maintenance plan (regular checks, periodic tests of water quality, protection from freeze–thaw damage, use of inert control treatments) is preferred, while knowing the phototrophic biodiversity (identified to the genera level by using these keys) helps in monitoring changes in the quality of water and in programming the interventions.

#### **5. Conclusions**

These identification keys could be a useful instrument for a better understanding of the phototrophic biodeteriogens present on the artistic fountains and other wet stone surfaces. These keys are indicative, and it is highly recommended to not assign a genera name to the observed microorganisms if the specific features are not observed, keeping the identification to the upper level (e.g., filamentous cyanobacteria with no clearly developed trichomes).

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/app11188787/s1, Figure S1: Microscopic observations of different samples taken from Sultana Fountain and North "Guitar" Fountain of the Court of the Myrtles, both in the Alhambra complex, Spain, Figure S2: Microscopic observations of different samples taken from Sultana Fountain and the Lions Fountain in the Alhambra complex, Spain, and from Villa la Pietra Fountain from Florence, Italy, Figure S3: Microscopic observations of different samples taken from Villa la Pietra Fountain from Florence and from different fountains from the Alhambra complex, Spain, Figure S4: Microscopic observations of different samples taken from Tacca Fountain from Florence, Italy, and from two fountains from the Alhambra complex in Spain, Figure S5: Microscopic observations of different samples taken from different fountains from the Alhambra complex, Spain, Figure S6: Microscopic observations of some isolated cyanobacteria from monumental fountains: *Aphanocapsa* (**a**), *Aphanothece* (**b**), *Synechococcus* (**c**), *Gloeobacter* (**d**), *Chroococcus* (**e**), and *Gloeocapsa* (**f**). Figure S7: Microscopic observations of some isolated cyanobacteria from monumental fountains: *Leptolyngbya*, that is thinner than *Oscillatoria*, presents sheath and lacks motility (**a**,**b**), *Pseudoanabaena* (**c**), *Pseudophormidium* (**d**), *Calothrix* (**e**,**f**). Figure S8: Microscopic observations of some isolated cyanobacteria from monumental fountains: *Nostoc* (**a**,**b**), *Pleurocapsa* (**c**), *Rivularia* (**d**). Figure S9: Microscopic observations of some isolated cyanobacteria from monumental fountains: *Chlorella* (**a**,**b**), *Scenedesmus* (**c**), *Bracteococcus* (**d**), *Monoraphidium* (**e**), *Dilabifilum* (**f**). Figure S10: Microscopic observations of some isolated cyanobacteria from monumental fountains: *Diatoma* (**a**), *Nitzchia* (**b**), *Achnantes* (**c**), *Surirella* (**d**).

**Author Contributions:** Conceptualization, F.B.-G., P.S.-C. and O.A.C.; methodology, F.B.-G., P.S.-C., O.A.C.; investigation, F.B.-G., P.S.-C., O.A.C., C.A.-R.; writing—review and editing, O.A.C., F.B.-G., P.S.-C. and C.A.-R.; supervision, O.A.C., F.B.-G. and P.S.-C.; project administration, F.B.-G.; funding acquisition, F.B.-G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the GOVERNMENT OF ANDALUSIA AND EUROPEAN RE-GIONAL DEVELOPMENT FUND, grant number A-HUM-279-UGR18 and P18-FR-4477 (FICOARTE regional project), and by the SPANISH MINISTRY OF ECONOMY AND COMPETITIVENESS, grant number PID2019.109713RB.100 (VIRARTE national project).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Acknowledgments:** The authors would like to thank the Alhambra and Generalife Patronate (Granada, Spain), the Real Alcázar de Sevilla (Seville, Spain), the Smithsonian Institution (Washington DC, USA) and the Florence Municipality and Villa la Pietra (Florence, Italy) for their assistance in collecting samples, the Ligalismo Foundation for its ongoing dissemination activity, and the Vice-Rectorate for Research and Knowledge Transfer of the University of Granada for their support and the creation of the unit of excellence Ciencia en la Alhambra.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


## *Article* **In Vitro Antibacterial Activity of Marine Microalgae Extract against** *Vibrio harveyi*

**Mohd Ridzwan Jusidin 1, Rafidah Othman 1,\*, Sitti Raehanah Muhamad Shaleh 1, Fui Fui Ching 1, Shigeharu Senoo <sup>2</sup> and Siti Nur Hazwani Oslan <sup>3</sup>**


#### **Featured Application: Potential application of** *N. oceanica***,** *Isochrysis* **sp. and** *T. weissflogii* **as inhibitory bacteria and probiotics in controlling bacterial diseases in the aquaculture industry.**

**Abstract:** Marine microalgae may produce antibacterial substances. At the exponential phase of growth, four species of marine microalgae were examined for their potential to create secondary metabolites that limit the growth of *Vibrio harveyi*: *Nannochloropsis oceanica*, *Chaetoceros gracilis*, *Isochrysis* sp. and *Thalassiosira weissflogii*. *V. harveyi* is a pathogenic bacteria that can cause severe mortality and loss in aquaculture. Disc diffusion assay and co-culture assay were used to determine antibacterial activity. On TSA % NaCl media, the disc impregnated with microalgae and extracted with ethanol, methanol, saline water, and dimethyl sulfoxide (DMSO) was tested against *V. harveyi* at concentrations of 1.0 <sup>×</sup> 105, 106 and 10<sup>7</sup> CFU mL<sup>−</sup>1. The disc diffusion assay revealed that *N. oceanica* extracted with ethanol had the largest inhibitory zone against *V. harveyi*. Meanwhile, only *N. oceanica*, *Isochrysis* sp. and *T. weissflogii* reduced the growth of *V. harveyi* (10<sup>5</sup> CFU mL−1) in the co-culture assay (*p* < 0.05). The current findings reveal that the hydrophilic chemicals in microalgae extract have antibiotic activity against the highly virulent *V. harveyi*, which causes vibriosis, a serious disease in farmed fish and aquaculture cultivation around the world.

**Keywords:** microalgae; antibacterial; *Vibrio harveyi*; *Nannochloropsis oceanica*; *Chaetoceros gracilis*; *Isochrysis* sp.

#### **1. Introduction**

In recent years, the aquaculture business has grown quickly and is predicted to supply approximately 62% of fish for human demand and consumption by 2030 [1,2]. However, diseases in aquaculture could impact the aquaculture production function by destroying basic resources, lowering the physical output or unit value of a production process, lowering the efficiency of a production process, and ultimately leading to economic losses in the aquaculture sector [2]. Vibriosis is one of the most prevalent bacterial diseases and is claimed to have damaged farmed marine fish in Malaysia, resulting in a USD 7.4 million loss in 1990 [3] and severe economic loss to Asian seabass farmers in 2017 [4].

Vibriosis is associated with infections in fish, such as skin necrosis, ulceration, and scale drops on the abdomen caused by a variety of *Vibrio* spp., including *Vibrio harveyi*, *V*. *vulnificus*, *V. parahaemolyticus*, *V*. *alginolyticus*, *V*. *penaeicida*, and *V*. *splendidus* [5]. *V. harveyi*, the primary bacterium responsible for catastrophic death in fish farming, has led to massive economic loss [6]. Antibiotics and chemicals are frequently used by farmers to combat harmful organisms. However, they have been used sparingly since they are expensive,

**Citation:** Jusidin, M.R.; Othman, R.; Shaleh, S.R.M.; Ching, F.F.; Senoo, S.; Oslan, S.N.H. In Vitro Antibacterial Activity of Marine Microalgae Extract against *Vibrio harveyi*. *Appl. Sci.* **2022**, *12*, 1148. https://doi.org/10.3390/ app12031148

Academic Editors: Anna Poli and Valeria Prigione

Received: 20 December 2021 Accepted: 18 January 2022 Published: 22 January 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

non-biodegradable, highly biomagnified, and antibiotic resistance has grown [7]. Antibiotic overuse can result in various environmental problems, including contamination of the culture environment, organism harm, and the development of bacterial resistance that can extend to the food chain [8]. Alternatively, numerous techniques for controlling pathogenic vibriosis have been proposed, including phage therapy, short-chain fatty acid inhibition of bacterial growth, quorum-sensing disruption, probiotics, immunostimulants, vaccinations, and green water [9].

Pathogen management in aquaculture, particularly disease prevention employing herbs and phytochemicals, has garnered a considerable amount of attention in the previous decade [10]. Furthermore, microalgae can produce a variety of bioactive compounds such as carotenoids, polysaccharides, vitamins, and lipids [11]. Recently, microalgae have been frequently utilised in aquaculture as nutritional supplements, immunostimulants to strengthen immune systems, and to increase disease resistance against harmful bacteria [2]. Furthermore, they contain antibacterial and antiviral properties that could be exploited for disease prevention and management in the aquaculture industry [11]. Thus, because the habitat is in a trapped and limited area, such as in fish-rearing tanks or cage culture, the transmission of numerous viral, fungal, parasitic, and bacterial diseases easily occurs among cultured fish, which can further cause co-infections and mortality in cultured fish [12,13]. It has been found that adding microalgae to farmed fish reduces the bacterial load of larval rearing systems by reducing the number of opportunistic bacteria [14].

Furthermore, the growing desire for more ecologically friendly disease control strategies has prompted academics to investigate alternate ways with little negative effects. Thus, the primary goal of this work was to investigate the antibacterial activity effects of several marine microalgae cultures, including *Nannochloropsis oceanica*, *Chaetoceros gracilis*, *Isochrysis* sp. and *Thalassiosira weissflogii*, against *V. harveyi* at different cell viability levels.

#### **2. Materials and Methods**

#### *2.1. Cultures and Preparation of Microalgae Extract*

The species of unicellular marine microalgae were tested. *N. oceanica*, *C. gracilis*, *Isochrysis* sp. and *T. weissflogii* microalgae strains were collected from the Culture Collection of the Live Feed Culture Laboratory, Borneo Marine Research Institute, University Malaysia Sabah. Each microalgae culture was produced and maintained in 250 mL Erlenmeyer flasks with 100 mL Guillard's F/2 media [15]. The microalgae culture was reactivated as an inoculum for four days and the starting culture concentration was 1 × 105 cells mL−1. The cultivation conditions were set at 25 ± 1 ◦C with 24 h of continuous illumination at 1000 μmol photon m−<sup>2</sup> s−<sup>1</sup> given by cool white fluorescent lights. The microalgae inoculum culture was put into 500 mL Erlenmeyer flasks with 200 mL of Walney's media [16]. After 10 days in culture, the cultivation reached the exponential phase of growth and cell density (1 × 107 cells mL−1). The cell viability of cell microalgae was evaluated using a Malassez haemocytometer on a daily basis. By plating 50 μL samples to tryptic soy agar (TSA) with 2% (*w*/*v*) NaCl and Thiosulfate Citrate Bile Salt Sucrose (TCBS) agar, the viability of cell microalgae culture was tested for contamination. The microalgae culture was afterwards recovered and harvested by centrifugation at low speed (2000× *g*) for 10 min at 15 ◦C, and the algal pellet was filtered through Whatman no. 1 filter paper [17]. The extraction of microalgae biomass was carried out with four different solvents: methanol, ethanol, Dimethyl sulfoxide (DMSO), and saline water, with a ratio of 0.1 g of algal biomass for 1.0 mL of solvent. Every microalgae extract was fixed into 106 cells mL−1. Before further usage, the crude extracts were promptly refrigerated at 4 ◦C.

#### *2.2. Preparation of Vibrio harveyi Inoculums*

Vibrio strains were collected from the Live Feed Culture Laboratory Culture Collection at the Borneo Marine Research Institute, University Malaysia Sabah. *V. harveyi* strains were obtained from an epidemic of vibriosis in Asian seabass, *Lates calcarifer* (Bloch). *V. harveyi* stock culture was subcultured into TCBS agar plates and incubated overnight at 28 ◦C. After 24 h of incubation, pure single colonies were selected and subcultured in 1.5 percent NaCl Tryptic Soy Broth (TSB). The bacterium was then cultured for 24 h at an incubator shaker. Then, 1 mL of inoculum was transferred to a 1.5 mL microcentrifuge tube and centrifuged for 5 min at 2000 rpm (10,000× *g*). After discarding the supernatant, 700 μL of TSB containing 1.5% NaCl was added to the tube and vortexed to mix the bacterial pellet with the TSB. To obtain varying amounts of inoculum, appropriate dilutions were made, and the viable cell count was validated using a spread plate test using a serial dilution of the bacterial suspension.

#### *2.3. Disc Diffusion Antibacterial Assay*

The various microalgae extractions were tested for their antibacterial activity against the pathogenic bacteria strains of *V. harveyi*. A modified agar disc diffusion assay method was used to investigate the antibacterial activity of algal extracts [18]. Petri dishes containing TSA and 2% (*w*/*v*) NaCl were seeded with *V. harveyi* inoculum at three different concentrations: 10<sup>5</sup> CFU mL−1, 106 CFU mL−1, and 107 CFU mL−1. To test the extracts' activity, sterile filter paper discs (6 mm) were impregnated with 20 μL of the various algal extracts. The discs were allowed to dry at room temperature before being placed on test plates inoculated with *V. harveyi*. As a control, discs with the same volume of extractants (20 μL) were made. The plates were incubated for 48 h at 35 ◦C. Extracts containing antibacterial components produced distinct, clear, and circular zones of inhibition around the filter discs, and this positive activity was quantified by measuring the growth inhibition zone (mm) surrounding the discs after 24 h and scoring them based on the diameter of the inhibition zone [19].

#### *2.4. Co-Culture Antibacterial Assay*

Cultures of four microalgae species, *N. oceanica*, *C. gracilis*, *Isochrysis* sp. and *T. weissflogii*, were utilised at 10<sup>6</sup> CFU mL−<sup>1</sup> concentration to evaluate their potential to inhibit the various viability of cell growth of *V. harveyi* by incubating this bacteria in these microalgae cultures (co-culture). The concentration of *V. harveyi* used in this co-culture experiment was determined by the disc diffusion assay result that indicated the broadest inhibitory zone. Co-culture assay samples were collected at 0, 12, 24, 48, 72, 96, and 120 h after the experiment began, and 50 μL of 10-fold dilutions were spread on TCBS agar plates to assess the bacterial content in the samples. All studies were performed in triplicate, and each microalga studied, as well as *V. harveyi*, received a control treatment.

#### *2.5. Statistical Analysis*

All data for the concentration of *V. harveyi* at each point of time for co-culture assay were statistically analyzed by one-way ANOVA using software SPSS (version 22), and graphs have been plotted using Microsoft Excel.

#### **3. Results**

#### *3.1. Disc Diffusion Antibacterial Assay*

According to the findings of this investigation, four microalgae extracts showed varying degrees of inhibitory impact against *V. harveyi*, as shown in Table 1, and the extraction had broad-spectrum activity against pathogenic bacteria. On the other hand, ethanolic extracts inhibited all microalgae with the broadest (+++) range of inhibition against the lowest concentration of *V. harveyi* at 105 CFU mL<sup>−</sup>1. However, at the maximum concentration of *V. harveyi* (107 CFU mL<sup>−</sup>1), *N. oceanica*, *Isochrysis* sp. and *T. weissflogii* extracts displayed a moderate (++) spectrum of inhibition, whilst *C. gracilis* showed low (+) inhibition. Interestingly, the ethanolic extract of *N. oceanica* demonstrated the broadest (+++) inhibitory zone against all *V. harveyi* concentrations tested; 105, 106, and 107 CFU mL<sup>−</sup>1.


**Table 1.** Antibacterial susceptibility test of the microbial extract against *V. harveyi*.

Note: (−): No activity, (+): D < 6 mm, (++): 6 < D < 8.5 mm, (+++): D > 8.5 mm. D: Diameter of the inhibition zone in milimeters.

#### *3.2. Co-Culture Antibacterial Assay*

In the co-culture assay, three (3) microalgae species—co-culture assays of *N. oceanica* (Figure 1a), *T. weissflogii* (Figure 1b), and *Isochrysis* sp. (Figure 1d)—demonstrated minimal antibacterial activity by preventing significant (*p* < 0.05) proliferation of *V. harveyi* at different time intervals when compared to the control group culture of *V. harveyi* with no microalgae.

In the co-culture *V. harveyi* with *N. oceanica* (Figure 1a), the concentration of *V. harveyi* was significantly lower (*p* < 0.05) at 24 to 48 h incubation as compared to control. However, there was no significant difference (*p* > 0.05) in *V. harveyi* concentration at 72, 96, and 120 h incubation.

For the co-culture *V. harveyi* with *T. weissflogii* (Figure 1b), the concentration of *V. harveyi* was significantly lower (*p* < 0.05) at 24, 48, and 72 h incubation, followed by no significant difference (*p* > 0.05) at 96 and 120 h incubation as compared to control.

Meanwhile, the co-culture *V. harveyi* with *Isochrysis* sp in Figure 1d demonstrated a significantly lower (*p* < 0.05) concentration of *V. harveyi* at 72, 96, and 120 h, as compared to the control. In comparison to the control, co-culture *V. harveyi* with *C. gracilis* (Figure 1c) exhibited no significant difference (*p* > 0.05) in *V. harveyi* concentration. In fact, no bibliographic data on the antibacterial activity of *C. gracilis* described in the prior study is available. The growth of *V. harveyi* strains increased steadily with time up to 48 h until reaching a plateaued condition and subsequently decreased insignificantly (*p* > 0.05) at the end of the experiment in the control group culture with no microalgae added.

**Figure 1.** *Cont*.

**Figure 1.** (**a**) *V. harveyi* concentration was significantly lower (*p* < 0.05) at 24–48 h incubation in co-culture assay with *N. oceanica* as compared to control *V. harveyi* with no microalgae added. (**b**) *V. harveyi* concentration was significantly lower (*p* < 0.05) at 24 to 72 h incubation in co-culture assay with *T. weissflogii* as compared to control *V. harveyi* with no microalgae added. (**c**) *V. harveyi* concentration was significantly lower (*p* < 0.05) at 72 to 120 h incubation in co-culture assay with *Isochrysis* sp as compared to control *V. harveyi* with no microalgae added. (**d**) *V. harveyi* concentration showed no significant difference (*p* > 0.05) in co-culture assay with *C. gracilis* as compared to control *V. harveyi* with no microalgae added. Note: Values are presented as mean ±SD. Mean values of *V. harveyi* within the same time (hours) with different lowercase letters are significantly different (*p* < 0.05).

#### **4. Discussion**

Many bioactive and pharmacologically active chemicals, particularly antibacterial compounds, are potentially produced by marine microalgae [6,20,21]. Furthermore, microalgae are gaining popularity due to their ability to create bioactive metabolites with anticancer, anti-inflammatory, antibacterial, and antioxidant effects [11,22]. Previous studies have suggested that fatty acids [23], terpenoids, carbohydrates [24], peptides, polysaccharides, and alkaloids are responsible for antibacterial activity in microalgae [25]. Novel antibacterial substances were also discovered in microalgae, with *Coccomyxa onubensis* fatty acid extracts inhibiting *E.coli* and *P. mirabilis* [26]. Furthermore, microalgae have been increasingly treated with other compounds such as cyanovirin, oleic acid, linoleic acid, palmitoleic acid, -carotene, or phycocyanin, which have antioxidant or anti-inflammatory properties, as well as antimicrobial activity, such as against *Staphylococcus aureus* and methicillin-resistant *Staphylococcus aureus* (MRSA) [22,23].

Table 1 shows variable degrees of inhibition on several species of microalgae extract against *V. harveyi* from this investigation. As a result, the generation of antibiotics is heavily reliant on microalgal species [27]. Furthermore, the availability of antibiotic agents can vary greatly between various species within the same class or within a single species, depending on which ecotypes are adapted to certain conditions [27,28]. The green microalga *Dunaliella* sp. isolated from severely polluted waters, for example, was found to be more active against bacteria than ecotypes obtained from less polluted waters [29]. Recently, microalgae of the genus *Nannochloropsis* have been shown to be rich in polyunsaturated fatty acid compounds (PUFAs), carotenoids, polyphenols, and vitamins, and have previously been used in aquaculture [30], while the microalgae strain of *Isochrysis* sp. was shown to be high in fucoxanthin and DHA [31]. *T. weissflogii*, on the other hand, has antibacterial properties due to its high presence of substances such as betain, lipids, phospolipid, polyunsaturated fatty acid, fucoxanthin (FX), and eicosapentaenoic acid (EPA) [32]. *C. gracilis* showed a limited inhibited zone as compared to other microalgae extracts, particularly at higher concentrations of *V. harveyi*. This could be because the compound composition of *C. gracilis* is minimal and contains only non-essential amino acids [33].

Apart from microalgae species, the presence of antibacterial chemicals in microalgae extracts is also strongly reliant on the extraction process and solvent utilised. According to the findings of this investigation, ethanol is the best extractant since it demonstrated the broadest inhibition against *V. harveyi* in the disc diffusion method. Based on prior research, ethanol was also used to extract the thalli of *Gracilaria fisheri* from India, the extract of which exhibited a high inhibitory effect against *V. harveyi* [34]. An ethanol extract of *Gracilaria corticata* from India was found to be highly active against *V. cholerae* and *V. parahaemolyticus*, but less active against *Pseudomonas aeruginosa* and *Shigella flexneri* [35]. Another study on ethanolic extract of *Spiriluna* also demonstrated more effective against two species of *Vibrio* [36]. These findings suggest that molecules with antibacterial action in microalgae are generally hydrophobic and might even be extracted more easily with organic solvents [27]. This could indicate that ethanol is a competent extraction method capable of extracting lipids with a specific set of chemicals present during extraction.

The methanolic extract is a well-established and reported method to isolate active antimicrobial components from microalgae [37,38]. In another study, methanol was found to be the best organic solvent to obtain bioactive compounds from *Dunaliella* sp. [39]. The methanolic extract of freshwater microalgae exhibited antibacterial activity against important human bacterial pathogens [40]. Methanolic extract in the current study inhibited *V. harveyi* moderately, whereas the antibiotic synergism with the methanolic extract of *C. vulgaris* exerted enhanced anti-bacterial activity in *E. coli*. [41]. Therefore, the selection of a solvent is also important for the further utilisation of post-extracted microalgal biomass. According to Navarro et al., 2017 [26], it was suggested that fatty acids could be involved in the antibacterial activity due to increased activity from green microalga extracts obtained with non-polar solvents compared to polar solvents.

In the present study, microalgae extraction by saline water and DMSO demonstrated a comparable less effective antibacterial activity against *V. harveyi* based on the concentration of *V. harveyi*. Nevertheless, it was reported that the extraction of antimicrobial compounds from *Scenedesmus subspicatus* using DMSO inhibited the growth of *Klebsiella pneumoniae* and *Escherichia coli* [42]. As observed in previous studies, water extracts of *Dunaliella* sp. had lower bioactive properties than the ones obtained from organic solvents [43]. Hence, the selection of a solvent is important for optimisation of the extraction process and generation of its bioactive compounds, which have diverse applications not only in aquaculture but also against human pathogens. According to Pradhan et al., 2012 [36] no antimicrobial activity was detected in the aqueous extracts, which was probably because of the low polar nature of the active components.

The number of *V. harveyi* increased exponentially, especially in the first 24 h during the experimental period, without the presence of microalgae cells (control) demonstrating that the bacteria cells were able to utilise the growth medium of the microalgae cultures. The inhibition of *V. harveyi* in co-culture assay with *N. oceanica*, *Isochrysis* sp. and *T. weissflogii* indicated the production of antibacterial compounds by these microalgae cells. Previous research has shown that *Isochrysis* sp. and *Nannochloropsis* exhibit antibacterial action against the majority of vibriosis pathogens, including *V. alginolyticus*, *V. lentus*, *V. splendidus*, *V. scophthalmi*, *V. parahaemolyticus*, and *V. anguillarum* [44]. The microalgae species *Nannochloropsis* sp. and *Isochrysis* sp. have been demonstrated to synthesise short-chain fatty acids, which are implicated in antibacterial activity [45,46]. Thus, microalgae biomass is regarded as a source of valuable chemical elements that can be used in animals, particularly farmed fish, as pharmaceuticals in aquaculture industries.

Another element that could explain these discrepancies is that antibacterial activity is linked to the culture conditions and growth phase of the microalgae cultures. Although all microalgae species were at the same concentration in all trials in this study, the growth phase for each microalgae culture could have been different and may have influenced the results. Previous studies suggested that modifying the culture conditions of green microalgae exhibited differences in antibacterial activity [47–49], whereas modifications on microalgae culture conditions could be stimulated to produce secondary metabolites with antibacterial activity, as well as potentially larger quantities of these secondary metabolites [50]. This may include modifications on media composition, pH, light, and temperature [51,52]. As an example, the highest light intensity (4800 lux) exposure effectively induced the production of antibacterial compounds by *Dunaliella* sp. and suggested that the higher antimicrobial effect was related to the content of bioactive compound produced in stress conditions [39]. Other than the culture environment, the period of microalgae cultivation also plays an important parameter in producing bioactive compounds [39,53]. Therefore, the culture environment could be artificially manipulated in order to allow for increased production of antibacterial compounds in microalgae.

#### **5. Conclusions**

Out of four species of microalgae, only *N. oceanica*, *Isochrysis* sp. and *T. weissflogii* demonstrated a significant ((*p* < 0.05) inhibitory effect against *V. harveyi* in an in vitro co-culture assay. This indicated a positive effect of the addition of these microalgae in the rearing of fish larvae and implicated the production of antibacterial compounds. Ethanolic extract from *N. oceanica*, *Isochrysis* sp. and *T. weissflogii* demonstrated promising antibacterial activity against *V. harveyi*. Meanwhile, *C. gracilis* inhibited the development of *V. harveyi* with minimal inhibition but no significant difference (*p* > 0.05) in the co-culture assay and disc diffusion method, indicating that it had less potential to limit the growth of *V. harveyi*. Therefore, in the aquaculture industry, *N. oceanica*, *Isochrysis* sp. and *T. weissflogii* act as potential inhibitory bacteria and probiotics in controlling the disease. More comprehensive studies are recommended to optimize its cultivation conditions, the extraction process, and the purification of its antibacterial compounds to unleash their potential.

**Author Contributions:** Conceptualization, R.O., S.R.M.S. and M.R.J.; methodology, R.O. and M.R.J.; validation, S.R.M.S. and R.O.; formal analysis, M.R.J.; investigation, M.R.J.; resources, F.F.C. and S.R.M.S.; data curation, M.R.J. and R.O.; writing—original draft preparation, M.R.J., R.O. and S.N.H.O.; writing—review and editing, R.O., S.N.H.O. and S.S.; visualization, M.R.J. and F.F.C.; supervision, R.O. All authors have read and agreed to the published version of the manuscript.

**Funding:** This project was funded by the Minister of Higher Education Malaysia under the Fundamental Research Grant Scheme FRGS0469-2017.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

**Acknowledgments:** This project was funded by the Minister of Higher Education Malaysia under the Fundamental Research Grant Scheme FRGS0469-2017. The publication fee was provided by the Research Management Centre, Universiti Malaysia Sabah.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Applied Sciences* Editorial Office E-mail: applsci@mdpi.com www.mdpi.com/journal/applsci

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com

ISBN 978-3-0365-3798-6