Introduction

Marine fungi are a taxonomically diverse group and are ubiquitous on substrata within coastal environments [21]. They serve as major decomposers of organic matters in marine ecosystems [19] but have also been reported as parasites, pathogens, and mutualists with other marine organisms [1, 23, 36, 37]. Deep-sea environments are recognized as an extreme marine habitat, which is characterized by an absence of sunlight, low nutrient input, extreme low temperatures (<4 °C, except near hydrothermal vents), and high hydrostatic pressure (up to 110 MPa) [33]. Burgaud et al. [4] suggested that the deep-sea floor and especially hydrothermal vent environment are under-explored habitats for fungi, which may possess unique fungal communities.

Although the isolation of deep-sea fungi was first reported approximately 50 years ago from the Atlantic Ocean [39], knowledge of “marine fungi” in deep-sea regions of the ocean is still limited. The recognition and importance of fungal organisms in marine ecosystems has increased over the last several decades, and recent studies have focused on fungal diversity in various deep-sea habitats, such as hydrothermal vents [5, 6, 15, 26], seawater and sediments [2, 9, 10, 12, 41–43], oxygen-depleted regions [20], and methane hydrate-bearing sediments [25, 31, 46, 47]. Most fungi isolated from deep-sea environments belong to Phylum Ascomycota, with a few species being affiliated with Basidiomycota [32]. In addition, Chytridiomycota and other basal fungal lineages have also been recovered from deep-sea environments, which were identified using culture-independent techniques [2, 26, 31, 32].

Earlier studies revealed the diversity of fungi from deep-sea sediments by conventional approach wherein several culture media and isolation techniques were used in conjunction with molecular identification [10, 15, 41]. With the development of culture-independent molecular techniques, the presence of fungi has also been directly reported from deep-sea DNA samples. These techniques involve an initial polymerase chain reaction (PCR) step to amplify genes of interest and a DNA separation technique, such as single-strand conformation polymorphism (SSCP), amplified rRNA intergenic spacer analysis (ARISA), restriction fragment length polymorphism (RFLP), terminal-RFLP, thermal and denaturing gradient gel electrophoresis (TGGE and DGGE), and length heterogeneity PCR (LH-PCR) analysis. However, universal PCR primers for all eukaryotic organisms were employed in most of these studies; hence, our knowledge on the diversity of fungi in the deep-sea environments is fragmentary [11, 28, 27, 29, 46]. Only recently, various fungal-specific PCR primers have been designed and used to report fungal communities in these environments [2, 25, 26, 31, 33]. Bass et al. [2] applied a fungal-specific primer set of the 18S rRNA gene to investigate the fungal diversity in deep-sea samples collected from various oceanic locations (water depths ranging from 1,500 to 4,000 m). Nagano et al. [33] investigated the diversity of fungal communities in deep-sea sediments collected at locations off the Japanese islands in the Pacific Ocean (depths ranging from 1,200 to 10,000 m) using three fungal-specific primer sets of the ITS of rRNA gene. Nagahama et al. [32] used two PCR primer pairs of the SSU rRNA to study the fungi inhabiting the methane cold seeps at Sagami Bay (depths 850–1,200 m). All of these studies revealed a wide occurrence of previously unidentified fungal taxa, which occupy a wide variety of niches in deep-sea environments. In recent years, quantitative PCR (qPCR) has emerged as a promising tool for quantifying microbial communities [14]. qPCR approach has been shown to be highly useful in assessing the abundance of specific phylogenetic groups of microorganisms in soils [22, 24, 44], which is also applied to fungal abundance in some environment samples [3, 14, 26].

Although the use of culture-independent molecular techniques with fungal-specific primers has revealed the presence of diverse fungi, including various unknown species in deep-sea environments, the distribution and diversity of fungal communities within the deep-sea floor sediments are still largely unknown, especially in regions where water depths are greater than 5,000 m. In the present study, we examined the composition and abundance of fungal communities in the deep-sea sediment samples from five locations of the Pacific Ocean (water depths ranging from 5,000 to 7,000 m) using universal fungal and phylum-specific primer sets, targeting the ITS1-5.8S-ITS2-28S rRNA regions, in order to advance our knowledge of fungal communities in deep-sea ecosystems.

Materials and Methods

Sediment Sampling

The deep-sea sediment samples were collected at five locations (water depths ranging from 5,017 to 6,986 m) in the Pacific Ocean during the cruise of DY115-23 of R/V ‘Hai-Yang-Liu-Hao’ from June to October 2011, including a sample from the deepest known part of the world’s oceans, the Mariana Trench. Details of the collected samples are described in Table 1. All sediment samples were immediately kept at −20 °C on the ship and brought to the laboratory for molecular analysis. Sediment temperature was measured with a portable electronic thermometer, while pH and salinity were determined using the pore water of sediments with pH meter and salinity meter, respectively.

Table 1 Details and physicochemical characteristics of the sediment samples collected from the Pacific Ocean
Full size table

DNA extraction, PCR Amplification, and Clone Library Construction

Total DNA was extracted from 0.5 g of sediment sample (wet weight) using a FastDNA® Spin Kit for Soil (MP Biomedicals, USA) according to the manufacturer’s protocol. Extractions were performed in triplicate for each sample. DNA extracts were pooled and then quantified by a NanoDrop1000 spectrophotometer (Thermo Scientific, USA) according to the manufacturer’s protocols.

Fungal-universal primer pair ITS1/ITS4 and five taxon phylum-specific primer sets ITS5/ITS4Asco (Ascomycota), ITS1F/ITS4B (Basidiomycota), ITS5/ITS4Chytrid (Chytridiomycota), ITS5/ITS4Oo (Oomycota), and ITS5/ITS4Zygo (Zygomycota) were used for ITS region amplification, respectively (Table 2). Due to the weak band obtained on 1 % agarose gel electrophoresis for the ITS1F/ITS4B primer set (Basidiomycota), a nested PCR approach using ITS1F/ITS4B as the first PCR primer set and ITS1/ITS4B as the second PCR primer set was applied. No products were obtained when using the nested PCR for the phylum of Chytridiomycota, Oomycota, and Zygomycota. PCR reaction was carried out for an initial 5 min at 95 °C, 32 cycles of denaturation at 94 °C for 1 min, annealing at 55 °C for 1 min and extension at 72 °C for 2 min, and a final extension of 7 min at 72 °C. Triplicate PCR reaction products were pooled together to minimize PCR bias. The amplified products were gel-purified and ligated into PGEM-T easy vectors (Progema, USA), which were transformed into Escherichia coli DH5α competent cells (Takara, China) to construct clone libraries. Recombinant transformants were selected using Luria–Bertani (LB) indicator plates supplemented with ampicillin (100 mg l−1).

Table 2 Details of the primers used in this study
Full size table

Screened clones were subjected to colony PCR using the vector M13F/RV primers. Over 100 positive clones with expected size from each library were subjected to RFLP analysis with restriction enzymes Afa I and Msp I (Takara, China). The digested products were size-fractioned on a 3 % (w/v) agarose gel electrophoresis and grouped according to their restriction patterns. The representative clones with different digestion profiles were selected for sequencing by Majorbio (Shanghai, China) using ABI 3730XL sequencer (ABI, USA).

Community Structure Analysis

The ITS gene sequences were edited using DNAstar software package (DNASTAR, USA) to clip the redundant sequences. All sequences from the environmental libraries were checked using Check_Chimera utility (Ribosomal Database Project) to eliminate chimeric sequences [7]. Representative sequences from each clone library and their closest relative matches retrieved from the GenBank were aligned using Clustal W (http://www.clustal.org/). Operational taxonomic units (OTUs) were defined as groups of sequences that differed ≤3 %. The library coverage values were calculated using the formula [1 − (n/N)], in which n is the number of OTUs represented only by a single clone and N is the total number of clones in the library. Rarefaction curves to the numbers of observed OTUs were calculated within the fungal assemblage for each dataset. Phylogenetic trees were constructed by selecting a representative sequence of each OTU in different samples and the reference sequences retrieved from the GenBank using neighbor-joining method in MEGA software (version 5.01). The relative confidence of the tree topologies was estimated by performing 1,000 bootstrap replicates. Gene sequences reported in this study have been deposited in GenBank under accession numbers KJ194276–KJ194455.

Quantitative PCR Analyses

Primers targeting the nuclear LSU rRNA gene fragment of the fungi (universal primers), Ascomycota, and Basidiomycota were chosen for an abundance analysis by qPCR (Table 2). Standard curves were generated using a tenfold serial dilution of a plasmid containing 102–108copies of either one of universal fungal, Ascomycota, and Basidiomycota LSU rRNA gene fragment. Concentrations of the plasmid DNA were measured using a NanoDrop1000 spectrophotometer (Thermo, USA). Three primer sets, NL1F/LS2R, LROR/ITS4Asco, and NL1F/ITS4B, were used to amplify the LSU rRNA gene of total fungi, Ascomycota, and Basidiomycota, respectively (Table 2). The PCR protocols were performed as follows: 95 °C for 10 min, followed by 45 cycles of 20 s at 95 °C, 20 s at 55 °C, and 20 s at 72 °C. Negative controls containing no template DNA were subjected to the same qPCR procedure to exclude any possible contamination. qPCR reactions were run in triplicate with the DNA extracted from each sediment sample. The specificity of the qPCR amplification of correct size was determined by the melting curve and checked with 1 % agarose gel electrophoresis. Cycle thresholds were determined by comparison with standard curves constructed using a tenfold serial dilution standard sample. Relative copy numbers among target groups were quantified via Pico Green analysis using the Rotor-Gene 6000 software.

Statistical Analysis

The weighted Jackknife environmental clustering analysis and principal coordinate analysis (PCoA) were conducted using the online software UniFrac (http://bmf2.colorado.edu/unifrac/) to evaluate the community similarity based on the gene sequence data. Comparison of the microbial communities was carried out using genetic distances based on the gene sequence data [30]. Environment cluster trees were projected in MEGA 5.01.

Results

Selectivity of Different Primer Sets

Six fungal-specific primer sets including one fungal-universal primer pair and five phylum-specific primer sets were applied to amplify the fungal ITS regions from the five sediment samples in the Pacific Ocean. PCR products were successfully obtained for all samples using the fungal-universal, Ascomycota- and Basidiomycota-specific primers, while no products were obtained for the primers specific for Oomycota, Zygomycota, and Chytridiomycota. Clone libraries were constructed for a total of 15 amplified PCR fragments (five samples, three primer sets), with 1,947 positive clones selected for restriction enzyme analysis including 766, 600, and 581 clones from the fungal-universal, Ascomycota-, and Basidiomycota-specific ITS clone libraries, respectively (Table 3). A total of 140, 86, and 62 restriction enzyme patterns was respectively obtained from the fungal-universal, Ascomycota-, and Basidiomycota-specific ITS clone libraries, which were subsequently subjected for sequencing. After analysis by BLASTN search in GenBank, six unique RFLP patterns (4.29 %) obtained from the fungal-universal primer set turned out to be of non-fungal origin. These six non-fungal sequences were affiliated with the genera of Aspidisca and Engelmanniella in the Alveolata [18]. No non-fungal taxa were retrieved from the phylum-specific primer sets.

The remaining 282 fungal sequences resulted in 48 fungal OTUs based on 3 % gene sequence distance cutoff: 16 (OTU1 to OTU16) in fungal-universal, 17 (OTUA1 to OTUA17) in Ascomycota-specific, and 15 (OTUB1 to OTUB15) in Basidiomycota-specific clone libraries (Table 3). The calculated coverage values showed that more than 95 % of the ITS diversity was captured in all clone libraries, which was further confirmed by rarefaction analysis.

Table 3 Biodiversity and richness indices of fungal ITS sequences from different clone libraries
Full size table

Fungal Composition

Sixteen OTUs were obtained from the fungal-universal ITS primers; 12 OTUs were referred to four classes of Ascomycota—Eurotiomycetes, Saccharomycetes, Dothideomycetes, Sordariomycetes, while the other four OTUs (OTU9, OTU12, OTU15, and OTU16) belonged to the classes Agaricomycetes, Exobasidiomycetes, and Tremellomycetes of Basidiomycota (Table 4 and Fig. 1). Twelve Ascomycete OTUs belonged to six genera (Aspergillus, Aureobasidium, Candida, Exophiala, Fusarium, and Periconia) and an uncultured deep-sea fungal group (DSF-group 1), in which OTU3 with a sequence similarity of 77 % to Fusarium spp. and OTU5 with a sequence similarity of 99 % to Candida metapsilosis were the most prevalent OTUs appearing in all five clone libraries. Eight OTUs were only recovered from single clone libraries, within which five (OTU10–OTU13 and OTU15) were represented by a single clone. Compared with Ascomycota, there was a lower diversity of Basidiomycota recovered. Only four Basidiomycete OTUs were obtained from two samples (JK and WS), which belonged to the genera Malassezia, Cryptococcus, Pleurotus, and Auricularia. The first three Basidiomycetous genera were obtained from WS, while Auricularia was only recovered from JK. These Basidiomycetous genera accounted for less than 1 % of the total clones in individual clone libraries except Pleurotus which formed a major portion (61.7 %) of the clones in sample WS.

Table 4 Molecular identification of fungal OTUs in this study
Full size table
Fig. 1
figure 1

Neighbor-joining phylogenetic tree for the fungal OTUs constructed based on ITS gene sequences obtained from three fungal primers sets. OTU 1-16 were recovered from fungal-universal primer set ITS1/4, OTU A1-17 were recovered from Ascomycota-specific primer set ITS5/4Asco, and OTU B1-15 were recovered from Basidiomycota-specific primer set ITS1F/4B. Numbers below branches indicate bootstrap values (>50 %) from 1,000 replicates. Novel sequences which showed low similarity (≤97 %) with available fungal sequences in the GenBank are marked with a triangle

Full size image

More diverse OTUs were recovered from the Ascomycota- and Basidiomycota-specific primers. The Ascomycota-specific primer amplified 17 OTUs belonging to five different classes: Sordariomycetes, Eurotiomycetes, Saccharomycetes, Dothideomycetes, and Leotiomycetes (Table 4 and Fig. 1). The number of OTUs obtained from each sample ranged between 4 and 11. Among 17 Ascomycete OTUs, 16 OTUs were referred to 14 named fungal genera (Fusarium, Ramichloridium, Chaetomium, Cladosporium, Aureobasidium, Candida, Gloeotinia, Meyerozyma, Lecanicillium, Paecilomyces, Aspergillus, Simplicillium, Xylaria, and Penicillium), while the remaining one (OTUA2) was affiliated with an unclassified Pleosporales. OTUA1 showing a sequence similarity of 99 % to Fusarium solani and OTUA5 with a sequence similarity of 99 % to Candida metapsilosis were the most prevalent OTUs, which occurred in all five clone libraries. Basidiomycota-specific primers recovered 15 OTUs belonging to two different classes: Agaricomycetes and Exobasidiomycetes (Table 4 and Fig. 1). OTUB3 related to Trichosporon insectorum with a 98 % sequence similarity was the most prevalent OTU which occurred in all five samples. These 15 Basidiomycete OTUs were affiliated with four known Basidiomycetous genera (Trichosporon, Cryptococcus, Pleurotus, and Trametes) and four unclassified or uncultured fungal group (unclassified Polyporales, unclassified Lycoperdaceae, uncultured Trechisporales, and uncultured Thelephoraceae).

Among the 48 fungal OTUs recovered in this study, 11 were unique in the fungal-universal, 13 in the Ascomycota-specific, and 14 in the Basidiomycota-specific ITS clone libraries. Four OTUs occurred in both fungal-universal and Ascomycota-specific ITS clone libraries, while only one OTU was shared in both fungal-universal and Basidiomycota-specific ITS clone libraries (Fig. 2). The number of singletons in libraries constructed with fungal-universal, Ascomycota- and Basidiomycota-specific ITS primer pairs was five, seven, and two, respectively (see table note in Table 4).

Fig. 2
figure 2

Venn diagram showing the number of OTUs that were detected simultaneously by two of the primer sets and the number of OTUs amplified exclusively with a single primer set

Full size image

Diversity Patterns of Fungal Communities in the Pacific Ocean

A total of 48 OTUs affiliated with eight fungal classes were recovered from five sediment samples in this study. Species richness of fungal taxa differed at different sampling sites: 17, 16, 13, 23, and 27 OTUs were respectively recovered from sites CQ, CW, JK, W, and WS, belonging to 14, 11, 10, 18, and 20 genera, respectively. The highest richness of fungal diversity was recorded at Central Pacific Ocean area (sites W and WS), followed by Northwest Pacific Ocean area (sites CQ and CW), and finally the Mariana Trench area (site JK).

The weighted Jackknife environmental clustering analysis based on ITS sequence diversity of fungal OTUs clearly distinguished three groups of fungal assemblages in the deep-sea sediments of the Pacific Ocean (Fig. 3a). Samples CQ and CW from the Northwest Pacific Ocean areas clustered together and were separated from those collected at sites W and WS in the Central Pacific Ocean areas, while the sample JK was distinct from other Pacific Ocean samples (Fig. 3a). The fungal community at the Mariana Trench area site (JK) showed the highest dissimilarity to all other samples. Similar classification was revealed in the PCoA analysis, and the separation was explained along the first principal coordinate (P1) by 54.41 % of the total community structure variation among all samples (Fig. 3b).

Fig. 3
figure 3

Weighted Jackknife environmental clustering (a) and PCoA (b) analyses for fungal ITS sequences in this study

Full size image

Fungal Abundance in Sediment Samples

Fungal abundance in sediment samples was estimated by qPCR assays targeting the LSU rRNA gene fragment of the total fungi (using universal primers), Ascomycota, and Basidomycota, respectively. For each qPCR assay, more than 50 positive clones were reamplified and sequenced. No nontarget sequence was obtained from these three individual qPCR assays. The coefficient of the real-time PCR assays with three primer sets was confirmed by the strong linear inverse relationship between the threshold cycle (C T) and the log value of gene copy number for both primers sets (r 2 ≥ 0.99). The PCR amplification efficiencies were 98.9, 94.5, and 91.2 % for the total fungi, Ascomycota, and Basidiomycota, respectively.

Quantitative PCR analysis showed that fungal abundance varied slightly within samples from different sites. The total fungal LSU rRNA gene copy numbers ranged from 3.52 × 106 to 5.23 × 107 copies/g sediment (wet weight), and the highest abundance occurred at sample WS which was five times over the lowest one (sample JK) (Fig. 4). The Ascomycota LSU rRNA gene copy numbers ranged from 2.89 × 106 to 1.58 × 107 copies/g sediment (wet weight), and the highest (sample WS) and the lowest (sample JK) differed by 5.5-fold. The Basidiomycota LSU rRNA gene copy numbers ranged from 5.67 × 104 to 4.22 × 105 copies/g sediment (wet weight), and the highest abundance was found in sample W and differed from the lowest one (sample JK) by 2.5-fold. The ratios of LSU gene abundance for Ascomycota and Basidiomycota to total fungi were 0.82–0.91 and 0.01–0.02, respectively. As a general observation, the highest abundance of total fungi and two dominant fungal taxa occurred in the Central Pacific Ocean areas, followed by Northwest Pacific Ocean area, and Mariana Trench area.

Fig. 4
figure 4

Abundance of the fungal, Ascomycete, and Basidiomycete LSU rRNA gene copies in the deep-sea sediments of the Pacific Ocean. Error bars represent standard deviations of triplicate analyses

Full size image

Discussion

The present study revealed 48 fungal OTUs from deep-sea sediments of the Pacific Ocean, including 27 phylotypes, which showed a low similarity (<97 % sequence similarity) with known fungal sequences in the GenBank database. The fungal community was dominated by Ascomycota, followed by Basidiomycota, which was consistent with previous studies on fungal diversity of deep-sea environments [32]. No sequences of Chytridiomycota and the LKM11 clade (a clone group mainly retrieved from oxygen-depleted aquatic environments including deep-sea environments) were detected in sediment samples in this study, suggesting low abundance at the sampling sites examined.

All five different classes of the Ascomycota recovered from the deep-sea environments in previous studies were identified in the present study. The most frequently detected Ascomycete class was Sordariomycetes, followed by Dothideomycetes, Saccharomycetes, Eurotiomycetes, and Leotiomycetes. The Leotiomycetes class has rarely been reported from the deep-sea environments except in the sediments of the Pacific Ocean [31, 33]. The most prevalent Ascomycetous genera recovered in this study were Candida, Fusarium and uncultured fungal group DSF-group 1. Candida occurred in all sediment samples in this study, showing a good agreement with previous deep-sea reports that it is one of the most common genera in deep-sea environments [2, 25, 31, 33, 41–43, 45]. Fusarium was represented by a number of OTUs and two of them were affiliated with Fusarium oxysporum, which was known as a denitrifying fungus in a previous study [20]. Sequences of the DSF-group 1 formed a major cluster in the Ascomycota with Metschnikowia/Candida within the Saccharomycotina as its closest described taxa. This group was described by Nagano et al. [33] with fungal sequences from the deep-sea sediments collected at several locations off the Japanese islands based on ITS and LSU rRNA clone library analysis. Analysis using SSU rRNA sequences also recovered this group from other deep-sea regions [2, 32] and oxygen-depleted environments [46]. Apart from Fusarium, other sordariomycetous genera including Periconia, Xylaria, Lecanicillium, Chaetomium, Simplicillium, and Paecilomyces were also discovered in this study, challenging the traditional view that the diversity of Sordariomycete phylotypes was low in the deep-sea environments [32].

Only three classes within the phylum Basidiomycota were recovered in this study, including Agaricomycetes, Tremellomycetes, and Exobasidiomycetes. Agaricomycete fungi in marine environments were mostly reported from mangroves [21, 32]. Although molecular studies have shown the ubiquitous presence of Agaricomycete phylotypes in deep-sea environments [2, 26, 31, 32, 40], only one culture (Fomitopsis sp.) has been isolated from a deep-sea hydrothermal seafloor mat [8]. A diverse range of Agaricomycete phylotypes retrieved in this study indicated that deep-sea sediments may harbor diverse Agaricomycete fungi, and improved culturing methods are needed to culture them. The Tremellomycete OTUs in this study were affiliated with two genera, Crytococcus and Trichosporon, which have been widely detected in cold environments such as deep-sea and Polar Regions [2, 13, 25, 32]. The only OTU belonging to Exobasidiomycetes was related to the genus Malassezia, which was ubiquitously recovered in the deep-sea environments based on molecular methods. Malassezia fungi have been known as the causative agents of skin diseases in marine mammals [31]. As the members of this genus have also been detected in terrestrial soil nematodes [38], it was speculated in some reports that Malassezia spp. were associated with small marine invertebrates inhabiting in deep-sea sediments, such as nematodes or polychaetes [31, 32].

Basidiomycota was only successfully detected in two out of five sediment samples and was much less diverse than Ascomycota in each clone library when using a fungal-universal primer set. Similar results were also observed in previous reports using fungal-specific SSU and ITS primers to investigate fungal diversity in marine habitats [2, 26, 33, 41]. This may be due to the high percentage of Ascomycota in deep-sea fungal community, which the fungal-universal primers preferentially amplified. Our qPCR results supported this speculation and showed that Ascomycota was about two orders of magnitude more abundant than Basidomycota in deep-sea sediment samples. A nested PCR approach with the Basidiomycota-specific primer set successfully amplified basidomycete sequences from all sediment samples and a high ratio of novel phylotypes (13/15) was observed in the Basidiomycota clone libraries, suggesting that there are diverse Basidomycota in the deep-sea floor and it is necessary to use improved culture-independent methods for enhanced detection of fungi in deep-sea environment.

In this study, 2.08 % of sequences were of non-fungal origin. These non-fungal sequences were only recovered from the fungal-universal primers, indicating that taxon-specific primers had a higher specificity to fungi compared with fungal-universal primer set. Moreover, the Ascomycota- and Basidiomycota-specific primer sets recovered more diverse fungal genera (15 and 8, respectively) than the fungal-universal primers (eight Ascomycetous and four Basidiomycetous genera), demonstrating that different fungal primer sets may detect different subsets of the fungal community, and the taxon-specific primer sets would be helpful to amplify the less abundant groups of fungi. Therefore, using multiple primer sets, especially the taxon-specific primers, may be essential to investigate the true fungal diversity in deep-sea environments.

Although information is lacking on the abundance of fungi in the deep-sea floor so far, the only report by Le Calvez et al. [26] showed that the fungal SSU rRNA gene copy numbers ranged from 1.91 × 105 to 1.35 × 107 copies/μg genomic DNA in hydrothermal vent ecosystems. Our study recovered a relatively high abundance of fungi (3.52 × 106–5.23 × 107 LSU rRNA gene copies/g wet weight sediment) in the deep-sea sediments of the Pacific Ocean, suggesting that fungi might be involved in important ecological functions in the deep-sea environments.

Mariana Trench, the deepest part of the ocean, is a unique environment on earth. With a depth of almost 11,000 m, a temperature of 2.5 °C, and a pressure of around 111.79 MPa, a comprehensive study of its microecology is technically difficult [17]. Nagano et al. [33] collected a sample from the Mariana Trench, but they failed to get any fungal DNA sequences from it, possibly because the sample predominantly comprised rough sand. This study successfully detected fungi from the sample collected at Mariana Trench area, although the diversity of fungi was low in comparison with that in other sampling sites of the Pacific Ocean. Community similarity analyses also suggested that the fungal community structure at the sample of Mariana Trench was distinct from that at samples from other locations of the Pacific Ocean. The effects of depth, temperature, and hydrostatic pressure on fungal communities may explain these differences, but these need to be studied further.

In conclusion, this study suggests a high diversity and abundance of fungi in deep-sea sediments of the Pacific Ocean, including a sample from the deepest ocean depth, the Mariana Trench. We discovered a high number of unknown fungal species, which accounted for 56.25 % of the total phylotypes, indicating that deep-sea environments may harbor a diverse taxonomically unknown fungal community than we expected. Future investigation should look into the ecological roles of fungi in deep-sea sediment using fluorescent in situ hybridization (FISH) and macro transcriptome approaches. This eventually will improve our understanding of fungal communities responding to extreme conditions, how they interact with other components of the ocean ecosystems, and the ecological role of deep-sea-derived fungi.