Biodiversity of fungi in hot desert sands

Abstract The fungal community of six sand samples from Saudi Arabia and Jordan deserts was characterized by culture‐independent analysis via next generation sequencing of the 18S rRNA genes and by culture‐dependent methods followed by sequencing of internal transcribed spacer (ITS) region. By 18S sequencing were identified from 163 to 507 OTUs per sample, with a percentage of fungi ranging from 3.5% to 82.7%. The identified fungal Phyla were Ascomycota, Basal fungi, and Basidiomycota and the most abundant detected classes were Dothideomycetes, Pezizomycetes, and Sordariomycetes. A total of 11 colonies of filamentous fungi were isolated and cultured from six samples, and the ITS sequencing pointed toward five different species of the class Sordariomycetes, belonging to genera Fusarium (F. redolens, F. solani, F. equiseti), Chaetomium (C. madrasense), and Albifimbria (A. terrestris). The results of this study show an unexpectedly large fungal biodiversity in the Middle East desert sand and their possible role and implications on human health.


| INTRODUCTION
Hot deserts are characterized by exceptionally limited availability of water and nutrients, extreme temperatures with wide day/night excursion, strong winds, and high ultraviolet (UV) radiation (Makhalanyane et al., 2015). Desert microbial communities are adapted to grow under hostile environmental conditions and have a foremost role in ecosystem bioprocesses (Pointing & Belnap, 2012). Compared to biomes of moderate or tropical regions, desert microbial communities differ in composition, function and display a low level of phylogenetic diversity (Fierer et al., 2012). Among eukaryotes, fungi are the most stress-resistant organisms, able to grow in water-scarcity and to tolerate desiccation through the formation of spores (Sterflinger, Tesei, & Zakharova, 2012).
Fungi comprehend a heterogeneous group of saprobes, parasites or symbionts microorganisms and can be pathogens or endophytes (Gadd, 2007). Different groups of fungi are adapted to the desert environment, comprising terricolous fungi, fungi associated with plants, hyphomycetes, yeasts, and microcolonial fungi (Sterflinger et al., 2012).
Although molecular studies of fungal diversity have started to emerge in recent years (Buée et al., 2009;Jumpponen, Jones, David Mattox, & Yaege, 2010;Zimmerman & Vitousek, 2012), the fungal communities of deserts are still poorly uncovered, especially with respect to human pathogenic fungi. In fact, desert mycological studies are mainly based on culture-dependent methods, which detect only a fraction of the fungal community (Sterflinger et al., 2012). As deserts are frequently visited by tourists and locals, the presence of pathogenic fungi might pose a risk to human health. Moreover, fungi that are usually saprophytes may act as opportunistic pathogens, especially in immunocompromised patients, or as allergens in atopic hosts (De Hoog et al., 2000). Furthermore, fungi and bacteria can be transported for long distance by the desert dust movement, including trans-pelagic migrations across continents, posing concerns for human health (Goyer, 2004;Rosselli et al., 2015;Sano et al., 1997).
The geography of Saudi Arabia and Jordan is mostly represented by the Arabian deserts, associated semi-deserts and scrubland. The climate of Saudi Arabia deserts is represented by extremely high daytime temperatures and an extreme temperature drop during nights, while Jordan has moderate temperature. Average summer temperature in Saudi Arabia is around 45°C but maximum values could reach 54°C, while the average temperature in Jordan is 20-35°C. The amount of annual rainfall is very low in these countries (Peel, Finlayson, & McMahon, 2007). Internationally, these sites are recognized for their natural and geomorphological value and have high conservation value for research and eco-tourism.
So far, only dated culture-based studies have been carried out on Saudi Arabia desert fungi (Abdel-hafez, 1981;Abdel-Hafez, 1982a,b,c,d), whereas the fungal communities of Jordan deserts have not yet been investigated. Therefore, the aim of this study was to improve the knowledge of the biodiversity of the filamentous fungi in these deserts sand by culture-dependent and independent methods. The identification of these fungi could be useful in the recognition of new sources for bioactive metabolites and in the determination of hazards posed by desert fungi on the development of human infection diseases.

| Sampling sites
Overall six sand samples were collected, four from Saudi Arabia and two from Jordan deserts. In detail, two samples were collected in Central Saudi Arabia (Riyadh Province) from the White desert

| Scanning electron microspopy
Fine grained sediment was mounted on aluminum stubs and viewing a low vacuum in a FEI QUANTA 200 SEM at 20 keV beam accelerating voltage, 0° tilt angle and 17 mm working distance. The image analysis program (ImageJ 1.38e) was used to measure the grain size of sand (Prakongkep, Suddhiprakarn, Kheoruenromne, & Gilkes, 2010).

| Venn diagrams and biodiversity indices
Sequences were categorized by aspects of interest, analyzed, and displayed by use of Venny 2.161 (http://bioinfogp.cnb.csic.es/tools/ venny/index.html). The alpha and beta diversity was calculated using QIIME. In particular, were calculated Shannon and Chao1 indexes plus the observed OTUs for alfa diversity. The samples were also clustered in a tree using Unweighted Pair Group Method with Arithmetic mean (UPGMA) for beta diversity.

| Isolation and identification of filamentous fungi
An amount of 0.1 g of each sand sample was aseptically weighed and inoculated into plates of Sabouraud dextrose agar (SDA) with cloramphenicol (0.05 g/L) and gentamicin (0.1 g/L) (Microbiol. diagnostici, Uta, Italy). Plates were incubated at room temperature for 7-15 days. Identification of fungi was done based on the macroscopic and microscopic characteristics using standard mycological methods (Hawksworth et al., 1995). DNA of isolated fungi was extracted from homogenized samples (mycelium) using MagPurix Plant DNA extraction kit (Zinexts Life Science Corp, New Taipei City, Taiwan) with the automatic extractor SaMag ™ System (Sacace Biotechnologies, Como, Italy), as described by the manufacturer. Molecular species identification was performed by PCR with fungal-specific primers ITS1F and ITS4 (White, Bruns, Lee, & Taylor, 1990) targeting the internal transcribed spacer (ITS) region of the ribosomal rRNA gene with the following conditions: initial denaturation at 94°C for 5 min, followed by 30 cycles of 94°C for 1 min, 50°C for 90 s, 72°C for 90 s, and final extension at 72°C for 7 min. PCR products were purified using QIAquick PCR purification Kit (Qiagen, Hilden, Germany) and sequenced (CERSAA, Albenga, Italy). The sequences were aligned with BLAST against the GenBank database (Altschul, Gish, Miller, Myers, & Lipman, 1990). Sequences were deposited in NCBI (acc. numb: MG679518-24). Moreover, the ITS sequences obtained were clustered with reference sequences and it was built a neighbor joining phylogenetic tree using Geneious 4.8.5 with 1000 bootstrap replication ((http://www.geneious.com, Kearse et al., 2012)

| Scanning Electron Microscopy
Scanning Electron Microscopy (SEM) observation of the six desert sand samples showed that the sands can be classified according to ISO 14688-1:2002 as "Coarse soil" with fine, medium, and coarse sand depending on the site (Figure 1). Samples 2_SA from Red desert (Riyadh Province) and 6_J from Wadi Rum were constituted of fine, regular sand grains with size average of 234 ± 89 μm and 384 ± 105 μm, respectively. Samples 4_SA from Mada'in Saleh and 5_J from Wadi Rum were composed of medium sand with irregular appearance and grain size average of 660 ± 201 μm and 773 ± 372 μm, respectively. Samples 1_SA from White desert (Riyadh Province) and 3_SA from Al Ula consisted of coarse, irregular sand grains with size average of 1.110 ± 420 μm and 1.410 ± 398 μm, respectively (Figure 1).

| Richness and diversity of fungi by cultureindependent 18S sequencing analysis
The presence of fungi was detected by 18S sequencing analysis in all the six samples analyzed.

| Eukaryotic richness and diversity in the sand samples
A total of 831 OTUs were identified in all the six samples, ranging from 163 in sample 2_SA to 507 in samples 1_SA and 6_J (Table 2).
We compared the richness calculated by the Chao1 estimator, which ranged from 212 to 636 OTUs per sample. Moreover, the measured diversity, estimated using the Shannon index, ranged from 4.0 to 6.8 per sample. All these alpha diversity indices showed that the sample with the highest diversity level was 1_SA, while the sample with the lowest diversity level was 2_SA, both collected in the Riyadh Province (Table 2). Venn diagrams were calculated on shared OTUs between the three locations: Riyadh (samples 1_SA and 2_SA), Mada'in Saleh (3_SA and 4_SA), and Wadi Rum (5_J and 6_J). The prevailing occurrence of a common core of shared taxa is shown in detail by the overlapping Venn diagrams representation in Figure 3.
The highest proportions of taxa were those shared by all three loca-  Figure 4).

| Isolation and taxonomic characterization of fungi
On the basis of cultivable approach, 11 filamentous fungi were isolated from the six desert sand samples. Three colonies grown from both the Jordanian samples, two colonies grown from the Saudi Arabian sample 3_SA, and three colonies grown from three other remaining Arabian samples. ITS sequencing allowed to trace the presumed identities (expressed as top-scoring homologies to GenBank records) of seven filamentous fungi, all belonging to the class Sordariomycetes. Fusarium was the most represented genus with three bona fide species, followed by Chaetomium (one species) and Albifimbria (one species), as showed in Table 3. Four isolates (not listed in Table 3) were not identified. A phylogenetic tree constructed using the ITS sequences obtained showed that our isolates clustered with the correspondent reference species (Figure 5).

| DISCUSSION
The life-limiting hot desert sands of the Middle East represent a harsh but extremely extended facet of our planet. In this study, we sampled a transect of three locations from Jordan to Central Saudi Arabia (Wadi Rum, Mada'in Saleh and Riyadh), exploiting for the first time eukaryotic population diversity of surface desert sand samples, with particular reference to fungal communities, using a metagenomic DNA sequencing approach as well as culture-based analyses.
A total of 612 taxa were detected with 18S rRNA sequencing analysis. In three of the samples analyzed, fungi represented more of the  Pezizomycetes is a diverse class with a role in ecological processes and symbioses including animal and plant pathogens and existing in aquatic and arid environments, but mostly occurring in hot regions or at high altitudes (Spatafora et al., 2006). As confirmed in our study, this class of fungi is common in tropical regions (Hansen & Pfister, 2006). high mortality rate (70-100%) (Chakrabarti et al., 2006;Skiada et al., 2011).  (Tazaki et al., 2004). By that way, opaque masses of air dust containing sand particles and fungi could be transported by strong winds and assume a role in the spread of fungal diseases (Rosselli  et al., 2015). As the sand from deserts surrounding populated places could be a source for mycoses, the establishment of fungal biodiversity in sand is of profound epidemiological importance.
It has been suggested that the presence of fungi is related to direct or indirect contamination originating from organic sources and climate variations. The principal microbial risk to human health encountered with sand in desert is similar and arising from contact with animal excreta, mostly camels, used as touristic attraction, or other smaller desert animals.
As expected, we gained more information about fungal communities in desert sand by metagenomic analysis compared to traditional culturable method. However, with the culturable method we were able to isolated several fungi whose ecology is linked to plants.  (Guarro, 2013;Salah et al., 2015;Tortorano et al., 2014). The genus Chaetomium, together with more common genera Penicillium and Trichoderma, belong to phylum Ascomycota (Wang, Houbraken, groenewald, Meijer, et al., 2016).
Chaetomium is cosmopolitan species occurs in soil, dung, wide plant materials, air and marine environments (Wang, Lombard, Groenewald, Li, et al., 2016). Chaetomium has also been reported to directly infect humans, and is most commonly associated with onychomycosis, but there is also one report from Indian authors presenting the case of corneal ulcer due to a coelomycetes (Kim et al., 2013;Reddy, Venugopal, Prakash, & Kamath, 2017). Members of family Stachybotriaceae include important plant and human pathogens, as well as several species used in industrial and commercial applications as biodegrades and biocontrol agents. Previous study of multi-locus phylogenetic showed that the genera Myrothecium is polyphyletic resulting in the introduction of 13 new genera with Myrothecium-like morphology. Among others, previously known all under the name of Myrothecium species, Albifimbria is for the first time recognized as independent species with distinct morphology . However, the impact of Albifimbria presence in sand and its implication on the human health is still insufficiently researched.
In conclusion, our results evidence an unexpectedly large fungal biodiversity in the Middle East desert sand, predominantly consisting of fungi adaptive for survival in the extreme environmental conditions prevailing in deserts, indicating the robustness and endurance of microorganisms such as the fungal classes Dothideomycetes, Pezizomycetes, and Sordariomycetes, which are worthy of further investigation. This study suggests that desert sand and plants may be a rich source of fungi, tolerant to extreme environmental conditions that could be examined for plant growth promotion activity in agriculture.
Moreover, this study might be helpful to resolve the strategies adopted by microbes in response to extreme conditions to defeat draining by induction of various types of solutes. Furthermore physico-chemical analyses of sand could uncover a more complete screenplay of the abiotic factors and fungal interplay conditions existing in the desert areas.

ACKNOWLEDGMENTS
We acknowledge Emanuela Sias for her technical assistance.

CONFLICTS OF INTEREST
Authors declare that they have no conflicts of interest.