Biodiversity and characterization of marine mycota from Portuguese waters

Biodiversity and characterization of marine mycota from Portuguese waters.— The occurrence, diversity and similarity of marine fungi detected by the sum of direct and indirect observations in Fagus sylvatica and Pinus pinaster baits submerged at two Portuguese marinas are analyzed and discussed. In comparison with the data already published in 2010, the higher number of specimens considered in this study led to the higher number of very frequent taxa for these environments and substrata; the significant difference in substrata and also in fungal diversity detected at the two environments is also highlighted, in addition to the decrease in fungal similarity. Because the identification of Lulworthia spp., Fusarium sp., Graphium sp., Phoma sp. and Stachybotrys sp. down to species level was not possible, based only on the morphological characterization, a molecular approach based on the amplification of the LSU rDNA region was performed with isolates of these fungi. This was achieved for three isolates, identified as Fusarium solani, Graphium eumorphum and Stachybotrys chartarum. To achieve this with the other isolates which are more complex taxa, the sequencing of more regions will be considered.


Introduction
Fungi have been known to exist in marine environments since early times.Hyde et al. (2000) highlighted the first reports of marine fungi up until 1846; however, interest in marine mycology only increased worldwide with Barghoorn & Linder (1944).
Considerable progress has been made in inventorying endophytes from marine hosts including seagrass (Alva et al., 2002;Sakayaroj et al., 2010).The diversity found comprises mostly anamorphic fungi and sterile mycelia and some isolates revealed to be producers of cellulases and xylanases (Alva et al., 2002).
Among the marine fungi isolated by Azevedo et al. (2010), five taxa could not be identified to species level based only on morphology: Lulworthia spp., Fusarium sp., Graphium sp., Phoma sp. and Stachybtrys sp.
In temperate waters Lulworthia species are among the most frequently detected fungi in submerged woods (Byrne & Jones, 1974;Mouzouras et al., 1985;Grasso et al., 1985Grasso et al., , 1990;;Azevedo et al., 2010) and halophytes (e.g. S. maritima) (Barata, 1997).Several investigations reported Fusarium species in sediments, sand dunes and recovered submerged twigs of Tamarix aphylla (Jones et al., 2009), on coral reefs (Morrison-Gardiner, 2002) and in wood baits submerged in marine environment (Azevedo et al., 2010).Graphium species are also found in wood in marine environments (Vrijmoed et al., 1982(Vrijmoed et al., , 1986;;Gonzálvez et al., 1998;Maria & Sridhar, 2003;Azevedo et al., 2010).Phoma species are widespread, occurring in a variety of environments and ecological niches; they are less explored in marine environment in which Phoma species completely new to science are regularly found (Aveskamp et al., 2010).Finally, Stachybotrys species are also detected in marine environment (Landy & Jones, 2006;Jones et al., 2009;Azevedo et al., 2010), having been considered important by Jones et al. (2009) to document the occurrence of these taxa in the sea and to discover their ecological role.
One goal of this work was to present and analyze, in a comprehensive manner, the data from the survey of Azevedo et al. (2010) concerning the occurrence, diversity and similarity of the marine mycota detected in wood baits before and after incubation in moist chambers (direct and indirect observations respectively).We further proposed to compare the results from this and other surveys carried out in temperate waters.
A second goal was to present and discuss the results of a molecular approach performed to characterize the isolates that had not been possible to identify down to species level based only on morphological characters.

Sampling strategies
Two marinas located on the western coast of Portugal, Cascais (38º 40' N 09º 25' E) and Sesimbra (38º 26' N 09º 06' W), were selected for the submersion of wood baits from Pinus pinaster Aiton and Fagus sylvatica L., as described by Azevedo et al. (2010) and shown in figure 1.The experimental design of baits is presented in table 1 and figure 1.The wood baiting technique involved a previous overnight soaking of the baits in distilled sterilized water followed by 20-minute autoclave sterilization at 121ºC.
After submersion, collections were performed periodically each eight to 10 weeks, on a total of six collections, at each marina, The baits were examined as soon as possible after collection under the dissecting microscope to detect spores and fruit bodies.Microscopic characterizations were performed under the light microscope (Leitz Laborlux S with Normarski) in slides prepared with seawater as mounting media and microphotographs were taken (fig.2).Thereafter, identifications were made following the dichotomous keys of Kohmeyer & Kolhmeyer (1979), Kohlmeyer & Volkmann-Kolhmeyer (1991) and Hyde & Sarma (2000).
The baits were analyzed by direct observation and then incubated in moist chambers for 12 months.They were re-examined on a monthly basis, following the procedures described by Vrijmoed (2000).The isolates of marine fungi subjected to molecular analysis were obtained by the single spore method (Azevedo et al., 2010).

Analysis of fungal occurrence, diversity and similarity
Frequencies of occurrence, expressed as percentages, were calculated taking the results from direct and indirect observations together.Marine fungi were classified as 'very frequent', 'frequent' or 'infrequent' based on Tan et al. (1989).
The average numbers of fungi per bait, species richness (S), Shannon (H') and evenness (E) diversity indices, as well as the Sorenson similarity index (Cs), were calculated as described by Figueira & Barata (2007).The values of Shannon Index were compared applying a t-test as proposed by Hutcheson (Zar, 1999).The cultures selected for molecular analysis were grown in Malt extract broth prepared with sea water on a rotary shaker at 200 rpm for 6-15 days at 20ºC.Fungal biomass was harvested, washed three times with sterile distilled sea water and frozen in liquid nitrogen to be ground into a fine powder with a mortar and pestle.DNA was extracted following the instructions of Nucleospin Plant DNA extraction Kit (Machery-Nagel, Germany).
A partial LSU rDNA sequence was amplified with LROR and LR5 primers (Viglays & Sun, 1994) and PCR reactions were carried out in a total volume of 25 µl with Phire Hot Start DNA polymerase (Finnzymes Oy., now Thermo Scientific) and 1 µl DNA sample, following the manufacturer' instructions.
The amplification program consisted of an initial 3-minute denaturation step at 98ºC followed by 35 cycles of (i) denaturation (98ºC for 10''), (ii) annealing (58.5ºC for 10'') and (iii) elongation (72ºC for 30 '') and a final extension of 1' at 72ºC.After a sample being resolved on 0.7% agarose gel, PCR products were purified by Jet quick DNA Clean Up Kit (Genomed GmbH), according to the manufacturer's instructions, and sent to be sequenced by a commercial lab.
Direct sequencing was performed by STAB VIDA (Portugal), using the same set of primers and the the big dye terminator kit on ABI automated DNA sequencer.
BioEdit Sequence Alignment Editor v7.0.9.0 (Hall, 1999) and ClustalW (Thompson et al., 1997) with default parameter settings were used for alignment and to obtain the consensus sequences.The obtained consensus sequences were compared to data in GenBank (National Center for Biotechnology Information, Bethesda, USA) online (www.ncbi.nih.gov), with GenBank BLASTn search engine.

Marine fungi occurrence, diversity and similarity
Table 2 presents the marine fungi detected by direct and indirect observations.The taxa are listed by decreasing values of frequency of occurrence in the ensemble of the two marinas; only infrequent fungi for both marinas were not listed.Diversity and similarity indices per environment and per substratum are presented respectively in tables 3 and 4.  3), the difference being highly significant for both types of baits: F. sylvatica (t 322.2 = -3.73;P < 0.001) and P. pinaster (t 521.2 = -4.49;P < 0.001).
Comparing the two marinas for mycota similarity, the Sorenson index presented a mean value for all analyzed situations (tables 3, 4), except for the comparison between the two types of baits submerged at Sesimbra marina (table 4).

Lignicolous marine mycota occurrence in temperate locations
Table 5 lists the very frequent and frequent marine fungi recorded in this and in other surveys carried out with submerged woods in temperate waters.

Sequence analysis of the selected fungi
Comparisons were made between partial sequences of the LSU rDNA region from our isolates and sequences from Genbank.Our sequences ranged between 886 and 915 base pairs.Concerning Lulworthia spp., the results from alignments and comparisons of sequences from selected isolates are until now inconclusive to achieve species level (data not shown).

Discussion
Marine mycota collected from wood baits submerged in temperate regions This analysis includes the total mycota detected on the survey of Azevedo et al. (2010).
The data of frequency of occurrence highlight the increase of the very frequent fungi (four taxa) in relation to the results reported by Azevedo et al. (2010) because C. maritima and Z. maritima (very frequent fungi) and R. quadriremis (frequent fungus) were not detected by direct observation.
The average number of fungi per Fagus sylvatica and Pinus pinaster baits increased respectively from 1.70 to 2.88 and from 1.92 to 3.68.This shows how the incubation on moist chambers significantly contributed for the differentiation of reproductive structures from the marine fungi mycelia already present when direct observations were carried out (Azevedo et al., 2010; table 2).
The diversity was significantly higher at Sesimbra than at Cascais and in P. pinaster than in F. sylvatica baits; a highly significant value was obtained when comparisons were done only with baits from Cascais.It is to be stressed that no significant differences were found between the two types of baits from Sesimbra marina when comparisons were made only with results of direct observations (Azevedo et al., 2010).
The values of fungal similarity (Cs) decreased for all analyzed situations when compared with the results presented by Azevedo et al. (2010).This evidences the advantages of using different types of substrata and subjecting them to long incubation periods in order to achieve better inventories of marine fungal communities.
Evenness values indicate that individuals recorded for each species were more evenly abundant in Cascais marina and for P. pinaster baits.
Studies in temperate open coastal waters relative to wood inhabiting fungi are based both in submerged and in drift or intertidal wood.When comparing the results of the survey of Azevedo et al. (2010) with other surveys carried out in temperate waters, differences found in fungal richness (table 5) could be due to the different nature of the woods used, to duration and depth of submersions in sea water and also to different abiotic conditions (oxygen, temperature, salinity) to which the woods were subjected as well as to the number of analyzed samples.
Lulworthia species were the most common fungi (present in five surveys), followed by Remispora maritima (observed in four surveys) C. maritima, H. appendiculata and M. pelagica (observed in three surveys) (table 5).The most common species can be considered species that play an important role in wood degradation (Alias & Jones, 2000).Additionally, considering the results expressed in table 2, it is to be emphasized that, for some of these taxa, there are references to production of enzymes and biocompounds.Bucher et al. (2004) reported production of cellulase, xylanase and peroxidase for one isolate of Lulworthia sp., and laccase for T. achrasporum.In relation to C. maritima, Jensen & Fenical (2002) found that an isolate of this fungus was able to produce a new secondary metabolite (Corollosporine) and Bucher et al. (2004) referred the production of cellulase and xylanase.

Sequence analysis of the selected fungi
The sequence data obtained suggest that our isolate of Fusarium sp. is closely related to Fusarium solani and F. lichenicola.However, the morphological characters are only compatible with the descriptions of Domsch & Gams (1980) and Samson et al. (2002) for F. solani as well as with the dichotomous key presented by Samson et al. (2002) for Fusarium species.Taking together morphological and molecular data, our isolate was considered to be Fusarium solani.
Concerning Graphium sp., the result indicating identity with Scedosporium apiospermum was evaluated, although the morphological features of our isolate (figs.2I, 2J, 2K, 2L) did not correspond with the description of this fungus (www.mycobank.org).The molecular results also revealed a close relation to the teleomorph Pseudollescheria boydii.It is worth nothing that Graphium eumorphum (Sacc.) is described as anamorph of this fungus (www.mycobank.org).The morphological features of our isolate are in accordance with the original description of Saccardo (www.indexfungorum.org),with slight differences on the length of conidia.For this reason, our isolate was considered Graphium eumorphum.
For the isolate of Phoma sp., our molecular results pointed out members of two other genera (L.aestuarii and C. obiones) as well as 11 species of Phoma that have never been described for marine habitats (Jones et al., 2009) Eight of these species of Phoma are included in clade 7 (Leptosphaeriaceae and Pleosporaceae) in the study performed with 159 species of Phoma and its associated teleomorphs by Aveskamp et al. (2010).These authors recognize the complexity of this group, which is considered to be one of the largest fungal genera.This explains why a better identification of our isolate was not achieved, also because only one DNA region was accessed by sequence data.
Concerning our isolate of Stachybotrys sp., comparisons of sequence data support the coincidence found between our morphological characterization and the one made by Samson et al. (2002) for Stachybotrys chartarum (= S. atra corda).S. atra was referred by Jones et al. (2009) for marine environments, however indicating conidia dimensions slightly smaller.For this reason our molecular data were determinant in considering our isolate as Stachybotrys chartarum (= S. atra).
Finally, regarding Lulworthia spp., our results pointed out the necessity of further analysis to accomplish the objective of characterizing the Portuguese isolates.The phylogenetic trees recently proposed for Lulworthiales comprise many isolates to be identified down to species level (Campbell et al., 2005;Jones et al., 2009) as well.We intend to contribute for the establishment of phylogenetic relationships within this taxon with the molecular characterization (still currently underway) of our isolates.
In conclusion, this molecular approach pursuing a contribution for the identification of these Portuguese isolates down to species level showed to be valuable as this goal could be achieved for three of them (Fusarium solani, Graphium eumorphum and Stachybotrys chartarum).However, the sequencing of more regions always allows more accurate results.This procedure will be mandatory for more complex taxa such as Lulworthia spp., and Phoma sp., those that, in this study, remain to be better characterized.

Fig. 1 .
Fig. 1. A. Cascais marina; B. Sesimbra marina; C. Set of wood baits before submersion; D. Pinus pinaster bait colonized with marine organisms after six months of submersion; E. Box of wood baits at the moment of submersion; F. Fagus sylvatica bait colonized with basidiocarps of Nia vibrissa.

Table 1 .
Experimental design of the baits.