Composition and Diversity of Fungal Decomposers of Submerged Wood in Two Lakes in the Brazilian Amazon State of Pará

Programa de Pós-Graduação da Rede de Biodiversidade e Biotecnologia da Amazônia Legal-Bionorte, Manaus, Amazonas, Brazil Universidade Federal do Oeste do Pará, UFOPA, Santarém, Pará, Brazil Instituto Nacional de Pesquisas da Amazônia, INPA, Laboratório de Micologia, Manaus, Amazonas, Brazil Museu Paraense Emilio Goeldi-MPEG, Belém, Pará, Brazil Universidade Federal de Mato Grosso, UFMT, Sinop, Mato Grosso, Brazil Miami University, Department of Biological Sciences, Middletown, OH, USA


Introduction
Freshwater fungi include species that inhabit water for all or part of their life cycle or any species adapted to colonize predominantly aquatic or semiaquatic substrates in nature [1]. ese organisms serve as key agents in the decomposition of submerged dead plant material. Fungal biomass is a source of nutrition for other organisms in aquatic ecosystems [2,3]. Fungal enzymatic activity also modifies plant substrates to make them more palatable to invertebrate shredders and scrapers [4].
Freshwater ascomycetes are a group of fungi known to actively participate in the decomposition of submerged woody debris [5,6]. is group is polyphyletic, with the majority of species belonging to the subphylum Pezizomycotina in the classes Leotiomycetes, Sordariomycetes, and Dothideomycetes [2,7]. Members of the Eurotiomycetes [8] and Orbiliomycetes [9] are less common. Shearer and Raja [10] reported that freshwater ascomycetes are distributed in relatively few orders: Helotiales, Pleosporales, Sordariales, Savoryellales, Microascales, Eurotiales, and Jahnulales. e number of known taxa has increased in recent years due to broader sampling and the use of molecular methods for identification [11,12].
e Amazon basin is arguably the most complex network of aquatic habitats on the planet [13]. is vast region contains a variety of aquatic environments that include large rivers, lakes, swamps, and seasonally inundated floodplains [13,14]. Aquatic ecosystems can be roughly classified into three main types: clear waters [15], white waters [16], and black waters [14,17]. Water characteristics such as temperature, dissolved oxygen, and amount of organic matter may influence the diversity of freshwater fungal species [18]. e first published studies of freshwater fungi in Amazonian waters (Peru) were conducted by Matsushima in white water rivers [19]. More recent studies have described new species of ascomycetes in the Peruvian Amazon [20][21][22][23][24] and in the Brazilian states of Pará [25,26] and Amazonas [27,28]. In general, the microbial community of the Amazon region is still underexplored, and studies describing the diversity and ecology of fungi in the Amazon region are urgently needed. erefore, we sought to investigate new environments such as the Tapajós River, one of the largest clear water tributaries of the Amazon River [17]. e present work aimed to describe the composition and diversity of fungi present on submerged decomposing wood in two lakes in the Tapajós River basin near Santarém, in the Brazilian state of Pará.

Sample Collection.
A single collection was made at each location in Lake Juá in October 2017 and in Lake Maicá in November 2017. At each site, 30 submerged wood fragments that showed signs of active decomposition were collected. e lengths of the samples ranged from 6 to 22 cm and diameters ranged from 6 to 15 cm (Figure 2(c)). ese substrates were transported in sealable plastic bags (20 cm × 10 cm), with a small amount of local water, to the Laboratory of Multidisciplinary Teaching in Applied Biology-Labio, Universidade Federal do Oeste do Pará (UFOPA). In the laboratory, the samples were gently rinsed with running water, placed in moist chambers (plastic boxes with moist paper towels), and incubated at room temperature with 12/12 h light/dark conditions.

Isolation and Identification of Fungi Present on Submerged
Wood. Every ten days, for six months, a Stemi DV4 stereomicroscope (Zeiss) was used to search for fungal structures on the collected wood samples. e structures found were transferred, with dissection needles, to microscope slides containing distilled water. e identification of sexual ascomycetes was performed based on the micromorphology of ascomata, hamathecium, asci, and ascospores using an Axioskop 40 microscope (Zeiss) as described previously [22,29,30]. Aqueous lactophenol cotton blue solution was used to determine the staining reactions of the apical ascus apparatus. Asexual ascomycetes were identified by investigating the types of conidiogenesis, conidiophore, and conidia, using an Axioskop 40 microscope (Zeiss). Isolation of fungal was carried out by transferring the fungal structures to the surface of Antibiotic Water Agar (20 g/L agar and chloramphenicol 250 mg/L), Potato Dextrose Agar (PDA) (Kasvi), and Malt Extract Agar (MEA) (malt extract 30 g/L, mycological peptone 5g/L and 15g/L agar, pH 5.4 ± 0.2). e developed colonies were transferred to PYG agar (1.25 g/L peptone, 1.25 g/L yeast extract, 5 g/L glucose, 250 mg/L chloramphenicol, and 18 g/L agar) [22] (Figures 2(d)-2(g)).

DNA Extraction, Amplification, and Sequencing.
DNA extraction was performed according to the protocol of Ferrer et al. [31]. Fungi were grown in test tubes containing 10 mL of potato dextrose broth (120 g potato, 10 g dextrose, and 1000 mL distilled water) for 10 days and transferred to Eppendorf tubes (2 mL) containing 500 μL of buffer (2.5 mg SDS, 7.0 mg NaCl, 3.65 mg EDTA, 20 mL of Tris-HCl and 100 mL of DI water, and 5 μL of ß-mercaptoethanol). e microtubes were subjected to freezing (−20°C) and heating (65°C for 1 h) in order to rupture the cellular structures. Next, 500 μL of a phenol/chloroform/isoamyl alcohol solution (v/v/v: 25/24/1) was added to the microtubes and vortexed until a homogeneous suspension was obtained. e suspension was then centrifuged (14,000 rpm, 15 min) and the supernatant was transferred to a new microtube (1.5 mL). Isopropanol was added to the supernatant in equal volume, and the mixture was homogenized and incubated at −20°C overnight to precipitate the DNA. e precipitated DNA was centrifuged (14,000 rpm for 15 min), washed twice with 70% ethanol, and resuspended in 100 μL of sterile water distilled. Molecular data were gathered by sequencing the ITS region of rDNA according to the procedure of White et al. [32]. e reaction yielded a final volume of 50 μL, consisting of PCR buffer (10 mM Tris-HCl, pH 8.3, 50 mM KCl), 1.5 mM MgCl 2, 0.5 μM of ITS-1 (5'-TCCGTAGGT-GAACCTGCGG-3') and ITS-4 (5'-TCCTCCGCTTATT-GATATGC-3'), 200 μM dNTPs, 1.5 U Taq DNA polymerase, and 100 ng of fungal DNA. ermocycling was performed on an Applied Biosystems ProFlex PCR System with initial denaturation at 94°C for 5 min, followed by 35 cycles of DNA denaturation at 94°C for 1 min, annealing at 55°C for 1 min, extension at 72°C for 2 min, and a final extension at 72°C for 10 min. Two positive controls and one negative control were also tested. Electrophoresis was performed on 1.5% agarose gel in 1x TBE buffer (100 mM Tris base, 100 mM boric acid, and 2 mM EDTA pH 8.0), plus 10 μL/100 mL of Siber Green SYBR ® Safe. Amplified fragments were visualized under UV 2 International Journal of Microbiology light. e amplification products were purified using a solution of polyethylene glycol-PEG (10 g polyethylene glycol 800, 7.3 g NaCl, and 35 mL of water) as previously described [33]. Similar volumes of the PEG and amplicon solutions were transferred to microtubes (1.5 mL), homogenized, and incubated (37°C for 15 min). e mixture was then centrifuged (15 min at 6,000 rpm), the supernatant was discarded, and the precipitate was washed twice with 70% ethanol and resuspended in water. e sequencing reaction was performed using the BigDye ® kit (Applied Biosystems). Sequencing was performed on the Seq 3130 Genetic Analyzer (Applied Biosystems). e sequences used in this study were compared and deposited in the GenBank database (Table 1).

Richness, Diversity, and Equitability of the Fungal
Community.
e number of individuals was determined for each piece of wood by the presence or absence of a species on it. In this study, an individual was quantified on a persubstrate basis. One individual was counted if it was found on a single piece of wood. However, multiple taxa were usually found to colonize individual substrates. Frequency, richness (species numbers), alpha diversity (α), and similarity of fungi were calculated for each lake. To characterize the diversity of the fungal community, the Shannon-Weaver diversity index (H′) was employed [34]. is index describes diversity with a higher value signifying greater heterogeneity and greater diversity. e following formula was used to perform these calculations: where p in � ni/N, ni � is the individual number of i th taxa, and N is the individual number of all taxa.
In addition, to interpret the Shannon-Weaver index, the evenness index (E) was calculated. is index is the uniformity of copy numbers among taxa. e evenness tends toward 0 when one taxon dominates the community and approaches 1 when the taxa have the same abundance. e index is expressed by the following formula: where H′ is the Shannon-Weaver Index based on the number of individuals and S is the number of taxa present in the sample. To evaluate differences in the fungal communities, the similarity among the samples was calculated using the Sørensen index (S′) with values ranging from 0 (no similarity) to one (absolute similarity). is was calculated using the following formula: where a is the total number of taxa collected in Lake Juá, b is the total number of taxa collected in Lake Maicá, and c is the number of common taxa in both lakes. A taxa-area curve was plotted for the two collections to assess the sampling effort. Statistical analyses were performed using the PAST program, version 3.2 [35].    In order to determine whether the sample size was sufficient to describe the diversity of the collection sites, the accumulation curve of the taxa found in the samples of submerged wood decomposed in the two lakes under study was analyzed by the ratio between the number of new taxa and the number of wood samples.

Physicochemical Characteristics.
In order to characterize the aquatic environments in this study, physicochemical analyses of the waters of the Lakes Juá and Maicá were performed at the time of sampling of wood samples. Table 2 lists the values of the parameters.

Sample Size.
e accumulation curves obtained for the samples from Lakes Juá and Maicá show that the number of taxa does not asymptote, suggesting that we have not adequately captured the freshwater fungal species (Figure 3) and additional collections may be necessary.

Identification of Fungi.
Morphologically, 23 different taxa were identified in the Lake Juá, with 43.5% (n � 10) being sexual (5 Dothideomycetes and 5 Sordariomycetes) and 56.5% (n � 13) being asexual. In the Lake Maicá, 26 taxa were identified morphologically, of which 65.4% (n � 17) were sexual (4 Dothideomycetes and 13 Sordariomycetes), and 34.6% (n � 9) were asexual. We carried out DNA analysis studies with only fourteen taxa (Table 1). It was not possible to investigate all taxa because most of them were uncultivated fungi that did not allow DNA extraction. ITS sequence data were successfully obtained and used to corroborate morphological identification (Table 3).
e Sørensen index of the fungal communities between the two lakes studied was 0.3673. Eight taxa occurred in both environments. Information on the richness, diversity, and equitability of the taxa observed in the two lakes is described in Table 4.

Discussion
In the present study, we described the fungal diversity associated with the decomposition of wood in lakes belonging to the Tapajós River (PA, Brazil). is is an important study because it is the first to describe the diversity of these organisms in the lakes of the Tapajós River region. In the experimental conditions, both lakes presented high diversity and few similarities among the taxa.
e Lake Juá had a low pH (4.9 ± 0.2) and low electrical conductivity (9.3 ± 0.2 μS). is was expected because Lake Juá is influenced by the Tapajós River [15,36]. Its turbidity was 30 ± 0.1 NTU, perhaps due to the direct interference of anthropic actions such as wastewater disposal due to the proximity to the urban area of Santarém. e Lake Maicá had neutral pH (6.7 ± 0.2) and high electrical conductivity (42 ± 0.2 μS).
is may be due to the influence of the Amazon River during the flood season [14]. is river has an alkaline pH, high electrical conductivity, and high turbidity due to the large amount of sediment coming from the Andean regions [15,16]. It is important to know the characteristics of the water in aquatic fungal diversity studies because previous studies have shown that factors such as pH, nutrients, substrate type, river geomorphology, and seasonality affect the fungal communities in aquatic environments [37][38][39].
e accumulation curve showed approximate stability after the collection of 20 to 25 samples from each lake studied (Juá and Maicá, respectively). Luo et al. [40] collected 100 samples of grass and bamboo from the Lake Dianchi in ailand during four sampling events over a oneyear period and reached asymptote in 50 and 70 samples. Hu et al. [41] collected 100 samples from a dammed lake and 90 samples in a stream, obtaining the stabilization of the curve in 60 and 80 samples, respectively. Cortez [28] collected 60 fragments of submerged wood in a lake in the region of Iranduba (Amazonas, Brazil) in two periods of the seasonal cycle, obtaining a plateau of species in 15 samples collected in the nonrainy period. In this study, 30 samples per lake were collected and appeared sufficient to near a plateau for local diversity.
At the Lakes Juá and Maicá, we observed 23 and 26 taxa, respectively. e number of unique taxa was 15 and 18 in the Lakes Juá and Maicá, respectively (Table 3). ese results are similar to those observed by Cortez [28], who identified 25 taxa in a black water lake in Amazonas State. When compared to other works [41][42][43], the total number of taxa in this study was lower. is may be related to differences in the number of samples per period, the quantity of collections made, and the types of environments studied. In addition, previous studies have stated that physicochemical factors of water play an important role in the richness of ascomycetes present in freshwater and may be a factor in these lakes [44].
e Lake Juá presented greater richness of asexual fungi (56.5%) and Maicá of sexual fungi (65.4%). In the Lake Juá,    this may be related to the influence of a stream that drains into this environment, supporting the increase of this taxon in the lake [45,46]. In the Lake Maicá, the sexual species richness was greater. ese results resemble those reported by Hu et al. [41], who found greater sexual richness in a lentic environment. Cortez [28] also found similar results and attributed these results to pluviometric action because rainwater can carry fungi from soil or vegetation to the lake. e five most frequent taxa in the Lake Juá were Pseudoxylomyces elegans, Xylomyces giganteus, Pleosporaceae MT017518, Fluviatispora sp., and Teratosphaeriaceae MT017515. In the Lake Maicá, the five most frequent taxa were ielavia sp., Longicollum biappendiculatum, Submersisphaeria sp., Aquaticola sp., and Morosphaeriaceae MT017525. Aquaticola sp. and Longicollum biappendiculatum were the most frequent taxa described by Cortez [28]. Abdel-Aziz [43] in a study on submerged stems of Phragmites australis (Cav.) Trin. ex Steud listed seven species that were common in this study (Aniptodera sp., Annulatascuslike, Aquaticola sp., Ophioceras sp., Pseudohalonectria sp., ielavia sp., and Monodictys-like). Raja et al. [42], in freshwater environments of the Florida Peninsula, shared five taxa (Aniptodera sp., Ophioceras sp., Submersisphaeria aquatica, Fluviatispora reticulata, and Longicollum biappendiculatum). It was observed that these taxa occurred in both lentic and lotic environments. According to Shearer et al. [24], the composition of freshwater ascomycetes species is similar between Florida, tropical South America (Peru), and tropical Southeast Asia.
Cancellidium applanatum, found in Lake Juá and Maicá, was also reported by Zelski et al. [22] in a survey of ascomycetes in wood submerged in freshwater habitats in Peru.
is species is considered a generalist and can occur in herbaceous and woody substrates of lotic and lentic habitats.
is species, as well as Monodictys sp., can be considered resistant to dry periods [47], and its presence is not surprising considering that the Lake Juá also undergoes considerable periods of drought. In addition, Acrogenospora sphaerocephala has recently been reported on leaf material in rivers and streams in the Caatinga biome as well as in submerged wood in a lake in Manaus, Amazonas [5,28]. Nine taxa were identified as new records for the Brazilian Amazon, as shown in Table 1.
e Lakes Juá and Maicá presented a diversity of H′ index � 2.65 and 2.82, respectively. Hu et al. [41] in ailand compared an artificial lake and a stream obtaining a diversity index of 2.34 and 3.68, respectively, indicating that factors such as the composition of riparian forest, damming of rivers, and anthropogenic actions may affect the community and the diversity of fungi in freshwater habitats. Cortez [28] studied the diversity in a lake in the state of Amazonas, making seasonal comparisons, and obtained indices of 2.60 in the rainy season and 2.13 in the nonrainy season. is study inferred that seasonality may affect fungal diversity. In the Nile, Abdel-Aziz [43] also performed seasonal comparisons, with indices of 3.08 and 2.44 for winter and summer, respectively. Hyde et al. [48] emphasized that the number of taxa per wood sample is higher in the tropics than in temperate habitats. According to Graça et al. [49], the variation in low diversity rates in tropical and subtropical areas may be related to ecological and methodological issues. e diversity indexes of the present work are similar to those found in other lentic environments (mean ∼ 2.8) and corroborate the assertion that diversity in lotic environments is greater (mean ∼ 4.60) [41,[50][51][52].
Sørensen's similarity index (0.3673) revealed low similarity among the studied lakes.
is may be due to the structural and physicochemical differences found during the study that are responsible for the composition of the mycota in aquatic ecosystems [37,38,53]. According to Leff [54], the microbial composition is affected in streams and rivers by the unidirectional flow of water (which is reflected in organisms' life strategies), mixing and aeration of water, and the importance of allochthonous materials.
Studies on fungi that decompose submerged wood in aquatic environments are lacking in the Amazon region. us, the present work was motivated to contribute to the knowledge of the diversity of aquatic fungi in this region. e results show that there is a rich and balanced diversity. is study provides more insight into the dynamics of freshwater fungi in clear water habitats and indicates how the community is distributed in the Amazonian lakes. It is noteworthy that further studies are needed to strengthen this line of research and increase available information on biogeography, ecology, and taxonomy of freshwater ascomycetes species in the Brazilian Amazon.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.