Antimicrobial activity of Iberian macroalgae

Several marine organisms produce bioactive metabolites in response to ecological pressures such as competition for space, maintenance of unfouled surfaces, deterrence of predation and the ability to successfully reproduce (König et al., 1994). These bioactive compounds offer rich pharmacological potential (Muñoz, 1992). There are numerous reports of macroalgae derived compounds that have a broad range of biological activities, such as antibiotic, antiviral, antiSCIENTIA MARINA 71(1) March 2007, 101-113, Barcelona (Spain) ISSN: 0214-8358


INTRODUCTION
Several marine organisms produce bioactive metabolites in response to ecological pressures such as competition for space, maintenance of unfouled surfaces, deterrence of predation and the ability to successfully reproduce (König et al., 1994).These bioactive compounds offer rich pharmacological potential (Muñoz, 1992).
The purpose of this work was to evaluate the antibacterial and antifungal activity of Iberian marine macroalgae.To date, research on biologically active substances of Iberian seaweeds has been scarce (Serarols et al., 1982;Cabañes et al., 1984;Ballesteros et al., 1992).The relationships the geographical zone, sampling season and algal generation have with antimicrobial activity, as well as the influence of sample preparation methods on assay results, are of considerable interest and have scarcely been studied.This information could prove valuable for harvesting algae for industrial applications.In fact, the present study corresponds to the first experimental task of a European project aimed at evaluating using macroalgae as natural preservatives in the cosmetic industry.

MATERIAL AND METHODS
A total of 82 taxa (18 Chlorophyceae, 25 Phaeophyceae and 39 Rhodophyceae) were sampled at various sites along the northern Mediterranean (Llançà, Port de la Selva, Palamós, Begur, Lloret de Mar, Blanes and the Ebro Delta) and Atlantic (San Sebastián, Guetaria, Ondarreta, Zarauz, Ría de Vigo and Bayona) coasts of Spain.To evaluate the possi-ble influence of sampling season on antimicrobial activity, the maximum possible number of these taxa in each season (winter, spring, summer and autumn) was collected.Seaweeds were collected by scuba diving or snorkelling and preserved on ice until further processing.Seaweed samples were manually cleansed of sand, epiphytes and animals, then rinsed in distilled water to remove salt.Samples from each taxon were prepared using two different treatments: freezing at -40ºC (hereafter referred to as fresh) and lyophilisation.The bioactivities of the fresh and lyophilised samples were subsequently compared to determine any differences resulting from the respective preparation methods.
As the bioactivity of Bonnemaisoniales has previously been reported, we carried out complementary studies for some of the taxa present in the Iberian Peninsula (Asparagopsis armata, its tetrasporophyte Falkenbergia rufolanosa, Bonnemaisonia asparagoides and Bonnemaisonia hamifera).The Bonnemaisoniales present on both Atlantic and Mediterranean coasts were collected to assess the relationship between the geographical zone and bioactivity.To evaluate whether antimicrobial activity varies with life-cycle generations of algae, gametophytic and tetrasporophytic generations of A. armata were analysed.
The test microorganisms selected were spoiled microorganisms or common human pathogens, and were comprised of the three Gram-positive bacteria Bacillus subtilis (ATCC 6633), Bacillus cereus (identified strain by the CECT) and Staphylococcus aureus (ATCC 29213), the two Gram-negative bacteria Escherichia coli (ATCC 35218) and Pseudomonas aeruginosa (ATCC 9027), and the yeast Candida albicans (ATCC 48867).All cultures were kept on Brain Heart Infusion (BHI) agar plates and stored at 4ºC, except the initial stock cultures, which were stored at -40ºC in BHI broth containing 20% glycerol.
Solid extracts from fresh and lyophilised material were prepared for all taxa following a modified version of the extraction method of Burkholder et al. (1960).The extracts were obtained by milling algal samples without solvent using a Waring blender and/or manually with a mortar.Due to the high bioactivity observed for the solid extracts of Bonnemaisoniales, we sought to prove this high bioactivity by also obtaining methanolic extracts.The extracts were prepared using a modified version of the method used by Caccamese et al. (1981) from algal material (ca. 2 g of lyophilised or 6 g of fresh material) homogenised via a Polytron in 10 ml of methanol-toluene (3:1).The extracts were then centrifuged to remove insoluble material, the supernatants were evaporated at reduced pressure, and the solid residue was then dissolved in 1 ml of methanol.
Antimicrobial activity was evaluated by the agar diffusion method, which is the most widely used method for susceptibility testing and is simple, economical and reproducible (Álvarez Benito, 1990).Moreover, this standardised procedure is accepted for determining antimicrobial susceptibility by the National Committee for Clinical Laboratory Standards (NCCLS).
A liquid microorganism suspension corresponding to a 0.5 McFarland scale (standard suspension of barium sulfate which represents 1.5x10 8 bacterial/ml) was applied to Mueller-Hinton plates using a cotton swab.After a few minutes, to allow complete absorption of the inoculum, the crude extracts were placed on the agar plates.The solid extracts (0.2 g), obtained from fresh and from lyophilised material of each taxon, were placed in 9.3 mm diameter wells made on the plates with a sterilised cork borer.The methanolic extracts from Bonnemaisoniales were absorbed onto non-impregnated discs (bioMérieux, 6 mm diameter), air-dried to eliminate residual solvent, and then placed onto the inoculated plates.
During overnight incubation at 37ºC, the yeast or bacterial lawn grew over the agar surface (Hodgson, 1984), except where it was inhibited by the radial diffusion of antimicrobial compounds of the extracts.The diameter of the inhibition halo is considered to be indicative of the bioactivity of the seaweed extract, and was measured (including the well or disc diameter) with a caliper.Mean diameter values were calculated from triplicate runs of each assay.Standardised values for diameters of the inhibition halo, expressed in mm, produced by the microorganisms against known antibiotics are listed in the literature (Álvarez Benito, 1990).Our results were interpreted according to these values, whereby a diameter less than 1 mm was interpreted as representing a taxon with trace activity, a diameter between 1 and 20 mm was interpreted as representing an active taxon, and a diameter larger than 20 mm was interpreted as representing a taxon with a level of bioactivity sufficient for antibiotic use (hereafter referred to as high activity).
The influence of algal treatment and sampling season on the results from solid extracts was assessed using variance analysis (ANOVA, Statgraphics Plus 5.1, Statistical Graphics Corp., 1994-2001).Both analyses were applied to the most sensitive microorganism, Bacillus cereus, and did not include the taxa belonging to Bonnemaisoniales.
Three analyses of variance were carried out for the Bonnemaisoniales taxa, including in this case all test microorganisms.Bonnemaisoniales from Atlantic and Mediterranean coasts were analysed for their season of maximum activity by two-way ANOVA with geographical zone and test microorganisms as factors.The difference in bioactivity between Mediterranean specimens of A. armata and its tetrasporophyte F. rufolanosa was assessed by two-way ANOVA using generations and test microorganisms as factors.To compare the bioactivities of fresh and lyophilised material from Bonnemaisoniales, solid and methanolic extracts were analysed by two-way ANOVA.
This work includes the first descriptions ever published of the bioactivities of 15 of the taxa studied (Tables 1-4).Of these taxa, seven were active against at least one microorganism: the green alga Enteromorpha muscoides, the brown algae Fucus spiralis var.platycarpus and Spatoglossum solieri and the red algae Boergeseniella fruticulosa, Gracilaria dura, R. tinctoria, Schottera nicaeensis and Scinaia furcellata.G. dura and R. tinctoria were notably active, the former against yeast and Escherichia coli,

Antimicrobial activity of solid extracts according to taxonomic group
Chlorophyceae had the lowest percentage of active taxa (44%, Fig. 1), with low bioactivity and a narrow spectrum of action that was generally limited to Gram-positive bacteria.Nevertheless, two taxa  were active against yeast, and Palmophyllum crassum was the only Chlorophyceae active against Escherichia coli (Tables 3-4).The largest number of active Chlorophyceae taxa belonged to the genus Bryopsis, with Bryopsis muscosa and B. corymbosa showing the highest values and the broadest spectrum respectively.Phaeophyceae had the highest percentage of active taxa (84%, Fig. 1), although these did not exhibit the highest antimicrobial activity among the taxa tested.The action spectrum of Phaeophyceae was broader than that of Chlorophyceae, as some taxa were active against the yeast or the Gram-negative bacteria (Tables 1-4).The highest activities for Phaeophyceae were observed for taxa from the genera Cystoseira, Dictyota and Taonia.Although all of these genera were active against the Gram-positive bacteria, Cystoseira was also active against the Gram-negative bacteria whereas Dictyota and Taonia showed antifungal action.Finally, among all of the brown algae tested, Hapalospongidion macrocarpum showed the highest antimicrobial activity.
Rhodophyceae demonstrated the highest antimicrobial activity and the highest number of taxa active against Gram-negative bacteria and yeast, thus it was the group with the broadest spectrum of action.However, the percentage of active Rhodophyceae (67%) was lower than that of Phaeophyceae (Fig. 1).Within this group, Ceramiales and Gigartinales had noteworthy antimicrobial activity, and Bonnemaisoniales was the order that had the highest bioactivity (Fig. 2).

Antimicrobial activity of the Bonnemaisoniales
Both solid and methanolic extracts from B. hamifera and B. asparagoides of the genus Bonnemaisonia showed a broad spectrum and high bioactivity; although the values for inhibition obtained from methanolic extracts were lower than those from solid extracts (Fig. 3).The microorganisms most inhibited by these taxa were the Grampositive bacteria: Bacillus subtilis for B. hamifera, and Bacillus cereus and Staphylococcus aureus for B. asparagoides (Tables 1, 2).
The solid and methanolic extracts of Asparagopsis armata and its tetrasporophyte Falkenbergia rufolanosa also exhibited a broad spectrum and high activity, although as above, methanolic extracts were somewhat less active than solid extracts (Fig. 3).For both generations the most inhibited microorganism was Candida albicans (Tables 1-4).The variation in antimicrobial activity between the algal generations was assessed by means of two-way ANOVA, using the test microorganisms and the generations as factors.Significant differences were observed for the generation (F 1, 54 = 8.89, p<0.01), A. armata was more active than F. rufolanosa, and for the microorganisms (F 5, 54 = 10.29,p<0.001), C. albicans was the most sensitive.The interaction term was not significant (F 5, 54 = 0.79, p>0.05), which indicates that the action spectrum did not vary with the generation.Three average groups were obtained in relation to the Tukey comparisons: one group comprised of C. albicans (the most inhibited microorganism), another group with all the Gram-positive bacteria and E. coli and finally one made up of the most resistant microorganism P. aeruginosa.
Out of all tested taxa, the Bonnemaisoniales had the maximum activity against all microorganisms: A. armata showed the maximum bioactivity against Gram-negative bacteria whereas the Bonnemaisonia species presented the maximum bioactivity against Gram-positive bacteria and yeast (Fig. 3).
For the order Bonnemaisoniales, the two treatments (fresh and lyophilised) were compared across taxa by two-way ANOVA for solid and for methanolic extracts.The differences between treatments were significant for methanolic extracts (F 1, 173 = 55.91,p<0.001), but not for solid extracts (F 1, 255 = 0.1, p>0.05).Significant differences between species were found for both solid (F 3, 255 = 39.62,p<0.001) and methanolic extracts (F 3, 173 = 13.5, p<0.001).The interaction term was also significant for both solid (F 3, 255 = 9.03, p<0.001) and methanolic extracts (F 3, 173 = 18.7, p<0.001).For Asparagopsis armata, lyophilisation was the most effective treatment for both types of extracts, whereas for its tetrasporophyte Falkenbergia rufolanosa, similar results were obtained from fresh and lyophilised material.For Bonnemaisonia species, the results varied according to the extracts: solid extracts from fresh samples had higher bioactivities than those from lyophilised samples, whereas the results were the opposite for the methanolic extracts.
The bioactivity values of the taxa were evaluated by two-way ANOVA with taxonomic group and season as factors.No significant differences between seasons (F 3, 330 = 0.68, p>0.05) were found, whilst they were found among taxonomic groups (F 2, 330 = 9.63, p<0.05).The significant interaction term found (F 6, 330 = 3.32, p<0.05) reflected the different variations in bioactivity of the taxonomic groups according to season: Chlorophyceae and Rhodophyceae demonstrated maximum activity during spring and autumn-winter respectively, and Phaeophyceae had constant activity throughout the year.
The greatest antimicrobial activity observed for the order Bonnemaisoniales against all test microorganisms was in winter for Asparagopsis armata and in spring for Bonnemaisonia hamifera (Tables 1-4).However, it must be taken into account that these taxa were only available during the winter-spring period due to their own seasonal dynamic.Falkenbergia rufolanosa, which was present all year-round, showed the greatest activity in autumn-winter.
The aforementioned observations and the differences in bioactivity between Mediterranean and Atlantic specimens of the Bonnemaisoniales observed in this work suggest that the bioactivity of the same taxon can vary with the geographical sampling zone.Martí et al. (2004) pointed out that these differences could depend on ecological parameters such as irradiance and nutrients.
Our observations of the effects of the sample preparation method/algal treatment (i.e., fresh or lyophilised) on bioactivity revealed that lyophilisation generally allows greater compound extraction.However, as the differences with the fresh material were not significant, it was not possible to determine the most universally efficient treatment.Previous studies that compared different treatments are scarce and were carried out on only a few taxa.Campos-Takaki et al. (1988) and Padmini Sreenivasa Rao et al. (1986) compared fresh and dried algal material; their results also showed lower activity in extracts from fresh tissue than in extracts from dried material.This is probably due to a higher dilution of the bioactive metabolites in the fresh material because of the higher water content.Only Della Pietà et al. (1996) employed lyophilised material, among other materials, but they did not compare their results.Nevertheless, we can conclude from their results that, as in our study, the lyophilised material showed the highest values of antimicrobial activity.
As regards seasonal variation of bioactivity, for all of the taxa tested, autumn was the season with the highest percentage of active taxa against at least one test microorganism, followed by spring.These results agree with those obtained from Indian samples by Sreenivasa Rao and Parekh (1981), and Arun Kumar and Rengasamy (2000), and from Mediterranean samples by Martí et al. (2004).In contrast, in the study carried out by Hornsey and Hide (1974) using Atlantic samples, the most active season was spring.In relation to taxonomic groups, the season with the highest percentage of active taxa was autumn for Phaeophyceae and Rhodophyceae, and summer for Chlorophyceae.However, the results observed for Chlorophyceae did not concur with the constant production of active compounds by this group throughout the year reported by Padmakumar and Ayyakkannu (1997).Some authors have associated peak activity with physiological phenomena; however, the peaks observed in the present work could not be attributed to a single biological process.In some taxa, such as Bonnemaisonia hamifera and Falkenbergia rufolanosa, the peak of bioactivity observed in our study may be related to the reproductive or growth period, as reported by some authors (Hornsey and Hide, 1974;Moreau et al., 1984;Muñoz, 1992).However, in other cases, such as for Osmundea truncata, the peak of bioactivity (autumn-winter) occurred after the reproductive period (springsummer).This finding was in agreement with that of Martí et al. (2004), who stated that peak bioactivity may be related to processes of ageing and allocation of resources from growth or reproduction to production of toxic compounds.
Of all the Iberian taxa screened, the highest antimicrobial activity was observed for Rhodophyceae, among which the order Bonnemaisoniales was the most active.Among the taxa tested in the present work, Bonnemaisonia asparagoides and B. hamifera had the highest degree of antimicrobial action against Gram-positive bacteria and yeast.Likewise, A. armata, F. rufolanosa, and its tetrasporophyte, which have been highlighted by other authors previously (Serarols et al., 1982;Cabañes et al., 1984;Pesando and Caram, 1984;Ballesteros et al., 1992), presented the highest activity against Gramnegative bacteria out of all the taxa tested in the present article.Comparing both generations, the gametophyte exhibited a broader spectrum and higher degree of antimicrobial action than its tetrasporophyte.Literature data about differences in antimicrobial activity between generations of the same species are scarce.In contrast to our results, Hornsey and Hide (1985) reported higher activity for the tetrasporophyte (Trailliella intricata) than for the gametophyte (B.hamifera).
Preservatives are described as substances that guarantee microbiologically safe products.After perfumes, preservatives are the cosmetic ingredients that cause the most skin irritations, allergies and atopic reactions.Based on the results of this paper, we suggest that the taxa B. asparagoides, B. hamifera and F. rufolanosa may have potential as industrial preservatives, analogous to the currently used A. armata (Seguin et al., 1995;Algues et Mer 2002).Out of these, F. rufolanosa could be the most suitable taxon for use as a natural preservative due to its year-round presence and its easiness to culture.Nevertheless, due to the high bioactivity obtained against Gram-positive bacteria and yeast, for the two Bonnemaisonia species, and against Gram-negative bacteria for F. rufolanosa, we propose a mixture of their active extracts to obtain a preservative with a broad spectrum of action.Moreover, these taxa merit further studies both with the aim of isolating their active metabolites and for assaying culture methods for supplying algal biomass for industry.We suggest analysing other taxa of this order for which antimicrobial activity is unknown but probably notable, such as B. clavata, which has never been tested before because it is generally misidentified (Salvador et al., 2006) with B. asparagoides.