Rhizoglomus venetianum, a new arbuscular mycorrhizal fungal species from a heavy metal-contaminated site, downtown Venice in Italy

Rhizoglomus venetianum, a new arbuscular mycorrhizal fungal species, has been isolated and propagated from a heavy metal-contaminated site in Sacca San Biagio island, downtown Venice, Italy. Interestingly, under the high levels of heavy metals occurring in the site, the new fungus was able to grow only intraradically. In greenhouse trap and single species cultures under low heavy metal levels, the fungus produced innumerous spores, clusters, and sporocarps extraradically, which were formed terminally on subtending hyphae either singly, in small spore clusters, or, preferably, in loose to compact non-organized sporocarps up to 2500 × 2000 × 2000 μm. Spores are golden-yellow to bright yellow brown, globose to subglobose to rarely oblong, 75–145 × 72–140 μm in diameter, and have four spore wall layers. Morphologically, the new fungus is similar to R. intraradices, and phylogenetically, it forms a monophyletic clade next to R. irregulare, which generally forms irregular spores and lacks, like R. intraradices, the flexible innermost wall layer beneath the structural/persistent third wall layer. A key for the species identification is presented comprising all 18 Rhizoglomus species, so far described or newly combined.


Introduction
The number of species of arbuscular mycorrhizal (AM) fungi (AMF, Glomeromycota, Tedersoo et al. 2018) greatly increased during the last decades, passing from about 120 to ca. 300 species since the 1990s (Schenck and Pérez 1990;Giovannetti et al. 1990;Oehl et al. 2011a;Błaszkowski 2012;Krüger et al. 2012). The continuous progress in the morphological identification of AMF species (e.g., Sieverding et al. 2014;Błaszkowski et al. 2015Błaszkowski et al. , 2018 and in the setup of suitable molecular tools based on more informative regions of nuclear rDNA (Silva et al. 2006;Krüger et al. 2009), resolving even very closely related taxa Krüger et al. 2012), allowed the discovery and separation of new species (ca. 10 per year in the last 20 years), covering diverse genera and families, some ubiquitous, but others rare or associated with particular plants or habitats (Turrini and Giovannetti 2012). However, the huge AMF diversity, evidenced on the basis of environmental DNA sequences, which do not correspond to formally described species (Öpik et al. 2010, 2014), is still far to be fully described. Actually, it was calculated that probably less than 5% of existing AMF in the world have formerly been described so far (Krüger et al. 2009), most of which have not yet been cultivated in pure cultures. Indeed, the difficulty in cultivating AMF out of their original environment still represents one of the major limits to the isolation and description of new species. Nevertheless, AMF research focused on invaluable Bhot spots^for species diversity (Myers et al. 2000), distributed around the world, together with studies on AMF occurring in extreme habitats (high-altitude habitat, highsalinity habitat, thermal habitat, desert, wetland, polluted soils) would probably lead to the discovery of novel endemic genetic resources (Turrini and Giovannetti 2012;Sousa et al. 2018). In very recent years, new species were isolated from extreme habitats, i.e., Rhizoglomus melanum from wetland (Sudová et al. 2015), Diversispora omaniana, Septoglomus nakheelum and Rhizoglomus arabicum from hyper arid environments of the Arabian Peninsula (Symanczik et al. 2014(Symanczik et al. , 2018, and Acaulospora pustulata and Acaulospora tortuosa from a high-altitude habitat in Sierra Nevada, Spain (Palenzuela et al. 2013).
Among extreme habitats, those contaminated by heavy metals (HM) represent invaluable sites for the recovery of peculiar AMF isolates and species. HM are extremely toxic to microbial communities, modifying the structure of some essential enzymes, causing oxidative damage and DNA injury, and altering plasma membranes (Hassan et al. 2011).
During the last years, HM-polluted habitats have been largely studied in relation to AMF occurrence, due to the their beneficial effects in enhancing plant tolerance to such toxic habitats (Pawlowska et al. 1996;Hildebrandt et al. 2007). It has been shown that AMF play a fundamental role in HM phytostabilization and phytoextraction (Göhre and Paszkowski 2006) by trapping HM in their hyphae and spores and, consequently, reducing metal availability for the plants ( González-Chávez et al. 2002Cornejo et al. 2013). Many studies have described different AMF communities occurring in HMpolluted soils (Turnau et al. 2001;Vogel-Mikuš et al. 2006;Regvar et al. 2010;Hassan et al. 2011;Ban et al. 2015;Yang et al. 2015;Sánchez-Castro et al. 2017) and in the roots of plants spontaneously growing in HM-polluted areas (Whitfield et al. 2004;Vallino et al. 2006;Zarei et al. 2008;Bedini et al. 2010). Molecular analyses of AMF communities of HM-polluted sites detected several sequence types new to science (Vallino et al. 2006;Sánchez-Castro et al. 2017), including sequences from Sacca San Biagio, an ash disposal island located downtown Venice (Italy; Bedini et al. 2010).
Sacca San Biagio, which can be considered a model system for the study of remediation and requalification of polluted environments, has been spontaneously colonized by pioneer plants, animals, and microorganisms during the past 30 years, since the incinerator plant's closure in 1984. Most of the plants occurring on the island showed a high level of AM fungal colonization (67%), but no spores were retrieved in bottom ashes. The molecular analysis, carried out on the roots of three mycorrhizal plants growing on the island, evidenced the occurrence of sequences corresponding to the BRhizoglomus intraradices/Rhizoglomus fasciculatum^complex, together with sequences detected so far only in planta and sequences new to science (Bedini et al. 2010).
The goals of the present work were to obtain in pure culture AMF spores, after transplanting the host plants from Sacca San Biagio island to a non-polluted environment, and to describe a new AM fungal species associated to the ruderal host plant Senecio inaequidens. Both morphological and molecular techniques were used to characterize the AM fungus, which was named Rhizoglomus venetianum.

Study site
The new glomeromycotan species originated from Sacca San Biagio island, downtown Venice (latitude 45°42′ N, longitude 12°30′ E). The island is a tideland that was elevated after the Second World War, filling the lagoon sand with inert construction waste, which has been subsequently covered by bottom ashes produced by an incinerator plant of municipal solid waste (MSW), operating for more than 10 years, since 1973 up to 1984. Ashes form layers of about 1.5-3 m and occupy a volume of about 60,000 m 3 (Bedini et al. 2010). They are highly toxic, since the incineration process enhances the concentration of some pollutants, especially heavy metals, which represent a risk for human health and the environment (Clijsters et al. 2000). Ashes contain high concentrations of different heavy metals: Al (25,941.6 mg kg −1 ), As (24.8 mg kg −1 ), Cd (3.4 mg kg −1 ), Cr (101.6 mg kg −1 ), Cu (1820.1 mg kg −1 ), Ni (83.6 mg kg −1 ), Pb (2000.8 mg kg −1 ), Zn (3210 mg kg −1 ) (Bedini et al. 2010). Moreover, they contain 85.1% sand, 10.1% silt, 4.8% clay, 5.1% total organic C, a pH (KCl) of 8, and the following total nutrient concentrations: 2.8 ‰ total N, 1215.3 mg kg −1 total P, and 24.6 mg kg −1 available P (Olsen) (Bedini et al. 2010). After the year 1984, the incinerator was demolished and no activities were carried out on the isle, which has been naturally colonized by both herbaceous and woody plants for more than 30 years. Plant communities detected on the isle were mostly ruderal and nitrophilous, (class Artemisietea vulgaris, including perennial ruderal xerophilous phytocenoses, typical of temperate or Mediterranean regions) (Mucina 1997). Among plant species occurring in Sacca San Biagio island, S. inaequidens plants, belonging to the Asteraceae family, were selected and collected with their intact root system. They are ruderal plants originating from South Africa, very adaptable to different environments, invasive, and spreading in many countries also in Europe. Senecio inaequidens, widely distributed on the isle at the sampling time, was chosen for spore isolation, since it was shown to be highly mycorrhizal and able to host in its roots different AMF species (Bedini et al. 2010).

Sampling, establishment, and growth of trap and pure cultures
Senecio inaequidens plants and rhizosphere toxic ashes were collected in Sacca San Biagio island in May 2011 (Figs. 1 and 2). Plants and ashes were transplanted in 8-L pots filled with a steam-sterilized 1:1 volumetric mixture of agricultural field soil collected at the Interdepartmental Centre for Agrienvironmental Research Enrico Avanzi (CIRAA), University of Pisa, S. Piero a Grado, Pisa, Italy (latitude 43°40′ N, longitude 10°19′ E) and TerraGreen (calcinated clay; OILDRI, Chicago, IL, USA), then transferred in greenhouse to induce AMF sporulation of fungi occurring in S. inaequidens roots, which were used as Btrap cultures^for the isolation of AM fungal spores from the heavy metal toxic island ( Fig. 3). At different time points, after 6, 9, and 12 months of culture, rhizosphere soil (50 g) from different trap cultures was sieved through a set of nested sieves down to a mesh size of 50 μm (Gerdemann and Nicolson 1963) for spore collection and analysis and the spores were extracted by a water-sucrose gradient and centrifugation (Sieverding 1991). Spores investigated in this study were obtained from spore clusters (max 5 spores connected with a common hypha). Pure cultures were achieved making use of the Bsandwich system,^useful to allow the interaction between AMF and plant roots (Giovannetti et al. 1993). Briefly, small spore clusters were let to germinate for 10 days in sterile microwells containing sterile water, in a growth chamber at 24°C in the dark. Germinated spores were then placed in contact with the roots of Trifolium alexandrinum plantlets placed on 47-mm diameter cellulose ester MilliporeTM membranes (0.45-μm-diameter pores). Another 47-mm membrane was placed on spores and root systems to close the sandwich. The Bsandwich system^was buried in sterile 10-cm pots containing steamsterilized quartz grit, then maintained in sun-transparent bags (Sigma, Milan, Italy) under controlled conditions (18-24°C, 16-8-h photoperiod of irradiance 100 μEm −2 s −1 , 60% RH). Plants were supplied weekly with 10 ml half strength Hoagland's solution. After 6 weeks, the Bsandwich systemsŵ ere opened, plants were gently removed, and the occurrence of extraradical mycelium was assessed under a stereomicroscope (Wild, Leica, Milan, Italy) (Figs. 4 and 5). When colonization check was positive, plants were transferred to 700cm 3 plastic pots containing a 1:1 mixture of soil and Terragreen and moved to the greenhouse. Pots were regularly checked for spore formation, as described above. Sporulation was obtained after 4 months of culture (Fig. 6). The new species was maintained and renewed under greenhouse conditions together with T. alexandrinum and Lactuca sativa as host plants in the International Microbial Archive (IMA) collection of the Department of Agriculture Food and Environment, University of Pisa, Pisa, Italy.

Morphological analyses
Spores were isolated from the pot-culture soil by wet sieving and decanting through a set of nested sieves down to a mesh size of 32 μm (Sieverding 1991), then transferred into Petri dishes and examined under a stereomicroscope (Wild, Leica, Milan, Italy). Spores were isolated by using capillary pipettes, mounted on microscope slides in polyvinyl alcohol lactoglycerol (PVLG; Koske and Tessier 1983), in PVLG + Melzer's reagent (1:1, v:v; Brundrett et al. 1994) and in water (Spain 1990). Qualitative spore traits (sporocarp and spore shape, color and size, spore wall structure, including color Production of extraradical mycelium from Trifolium alexandrinum roots used as bite plants in the sandwich system (root diameter ca. 500 μm). 6 Spores of R. venetianum produced in the type culture (single spore diameter ca. 80-140 μm) and size of each wall layer of the spores and their subtending hyphae) were examined on > 25 sporocarps and > 100 spores.
The terminology of the spore structure basically is that presented for species with glomoid spore formation in Oehl et al. (2011b), Sieverding et al. (2014), and Błaszkowski et al. (2018). Photographs were taken with a digital camera (Leika DFC 295) on a stereomicroscope (Leica S8APO) and a compound microscope (Leica DM750) using Leica Application Suite Version V4.1 software. Specimens mounted in PVLG and a (1:1) mixture of PVLG and Melzer's reagent were deposited at Z + ZT (ETH Zurich, Switzerland) and at the Botanical Garden of the University of Pisa, Italy.

Molecular analysis and phylogeny
Intact, healthy spores were manually collected with a capillary pipette under a dissecting microscope (Wild, Leica) and cleaned by sonication (120 s) in a B-1210 cleaner. After three rinses in sterile distilled water (SDW), spores were surface sterilized with 2% chloramine T supplemented with streptomycin (400 μg/ml) for 20 min and rinsed five times in SDW. Intact sterilized single spores were selected under a laminar flow hood, individually transferred into Eppendorf PCR tubes, crushed with a glass pestle, and their DNA directly amplified using the nested protocol of Krüger et al. (2009), focused on a fragment of about 1500 bp covering partial SSU, the whole ITS and the D1 and D2 variable regions of the LSU sequences of rDNA. In the first PCR reaction, crushed spores were amplified in 25 μl of PCR reaction mix using 0.125 U GoTaq Flexi DNA Polymerase (Promega, Milan, Italy), 0.4 μM of each primer (SSUmAf1 and LSUmAr3, Krüger et al. 2009), 0.2 mM (each) dNTPs, 1.5 mM MgCI 2 , and 1× manufacturer's reaction buffer. The thermal cycler was programmed as follows: a manual hot start at 95°C for 3 min, 35 cycles at 95°C for 30 s, 60°C for 1 min, 72°C for 2 min, and a final extension step at 72°C for 10 min. The nested PCR reactions were performed by diluting (1:100) the first PCR amplicons and using 2 μl of dilutions as template for the second reaction in a final volume of 50 μl. The primer pair, SSUmCf1-LSUmBr3 (Krüger et al. 2009;0.4 μM), was added to the PCR mix. Taq DNA polymerase, dNTPs, buffer, and MgCl 2 concentrations were the same as those described above. Amplification conditions were as follows: a manual hot start at 95°C for 3 min, 35 cycles at 95°C for 30 s, 63°C for 45 s, 72°C for 1.5 min, and a final extension step at 72°C for 10 min. PCR products (10 μl) were separated on 1% agarose gels containing ethidium bromide (0.5 μg/ml).
Amplified DNA fragments of SSU-ITS-LSU regions were purified by Wizard SV Gel and PCR Clean-Up System according to the manufacturer's instructions (Promega), with a final elution volume of 20 μl, and purified products (2 μl) were quantified by a BioPhotometer (Eppendorf). Purified products were cloned into pGem®-T Easy vector according to the manufacturer's instructions (Promega). Putative positive clones were screened by standard SP6/T7 amplifications, followed by a nested PCR using the SSUmCf1-LSUmBr3 primer pair. Concentration of PCR mix components and PCR conditions were the same as those described above for PCR reactions. Six clones from single spores (two clones/ spore) were purified by Wizard® Plus SV Minipreps (Promega). Recombinant plasmids were sequenced using SP6/T7 vector primers at GATC Biotech (Köln, Germany). Sequences reported in this study were deposited in the European Nucleotide Archive (http://www.ebi.ac.uk/ena) under the accession numbers LS974594-LS974599. The alignment for the tree obtained in this study was deposited at TreeBase under the ID: 23045.
Sequences were edited in MEGA6 (Tamura et al. 2013) and their similarities determined using the Basic Local Alignment Search Tool (BLASTn) provided by NCBI. Then, they were aligned with those corresponding to the closest matches from GenBank as well as with some sequences from the Glomeraceae family covering the entire SSU-ITS-LSU region or portion thereof, using MUSCLE as implemented in MEGA6. Phylogenetic tree was inferred by Bayesian and maximum-likelihood analyses. The Bayesian analysis was carried out in MrBayes version 3.2 (Ronquist et al. 2012), using the General Time Reversible sequence evolutionary model and branch support values corresponded to the posterior probabilities of two Markov chain Monte Carlo samplings over 500,000 generations and a tree sampling every 100 generations after discarding the initial 10%. For the maximum-likelihood analysis, the evolutionary rate differences among sites were computed in MEGA6 using a discrete Gamma distribution + G method. The confidence of branching was assessed using 1000 bootstrap resamplings. The generated phylogenetic tree was drawn in MEGA6 and edited in Adobe Acrobat XI.
Etymology: venetianum referring to the city of Venice to which the island Sacca San Biagio belongs.
Type: Holotype, deposited at Z + ZT (accession ZT Myc 58970), derived from a pure culture established on the host plant Trifolium alexandrinum in the greenhouse of the Microbiology Laboratory in Pisa, at the Department of Agriculture, Food and Environment, University of Pisa, Italy. Mycorrhizal mycelia and intraradical mycorrhizal structures, occurring in the roots of Senecio inaequidens plants, were collected in Sacca San Biagio island in the central lagoon of Venice Italy (45°25′ 36″ N; 12°18′ 34″ E). The island hosted a municipal solid wastes' incinerator operating since 1973 up to 1984, producing bottom ashes that were disposed all over the island area. Since 1984, no significant activities were carried out on the island and in 2003 the furnace was demolished. In the last 30 years, the highly polluted soil of the island has been naturally colonized by spontaneous vegetation. Collector of the host plant S. inaequidens was A. Turrini and collection date 31.05.2011. Isotypes were deposited at Z + ZT (ZT Myc 58971) and at the Botanical Garden of the University of Pisa (PI-MH-Z11 to Z18). The living culture is currently maintained in the International Microbial Archives in Pisa under the accession number IMA10.
Subtending hyphae (SH) of spores cylindrical to slightly funnel-shaped, sometimes recurved, 9.1-13.2 μm broad and 15-100 μm long, and without introverted wall thickening toward the spore base. The base generally is not closed by a septum formed by SWL3 or SWL4 but open. Such septa can more often be found in some distance from the spore bases within the intrasporocarpic hyphae (ISH), which are 8-15 μm thick, golden yellow to bright yellow brown. SWL2-4 Figs. 7-12 Rhizoglomus venetianum. 7-10 Sporocarps in water isolated from the rhizosphere of Trifolium alexandrinum from the single species culture of the type. Intrasporocarpic hyphae (ish) are generally pigmented, while extrasporocarpic hyphae (esh) are hyaline to subhyaline. 11 Sporocarp fragment mounted in PVLG. 12 Sporocarp fragment mounted in PVLG + Melzer's reagent. Several septa (sp) can be observed on intrasporocarpic hyphae and especially at the transition between extra-and intrasporocarpic hyphae continue in the ISH. The extrasporocarpic hyphae (ESH, or mycelia hyphae) are hyaline, 5-12 μm thick, and have 1-2 hyphal wall layers. In the transition zone between ISH and ESH, hyphae are slightly pigmented and generally show several septa within a rather short distance of 100 to 200 μm.
Mycorrhiza formation: forming AM associations with Trifolium alexandrinum L. and Lactuca sativa as plant host in pot cultures. The mycorrhizal structures consist of arbuscules, vesicles, and intra-and extraradical hyphae and stain dark blue in 0.05% trypan blue.

Key to the species in Rhizoglomus
Here, we are presenting an updated morphological identification key for all 18 species belonging to the genus Rhizoglomus according to Sieverding et al. (2014), including the recently described or newly combined species R. dunense, Figs. 13-18 Rhizoglomus venetianum. 13-15 Uncrushed and crushed spores in PVLG. Spores with four layers (SWL1-4), cylindrical to slightly funnelshaped subtending hyphae (SH) and open pores at the spore base. Sometimes a septum (sp) can be detected at some distance to the spore base. 16-18 Spores mounted in PVLG + Melzer's reagent. SWL1 and SWL2 stain pinkish to pinkish purple, while SWL3 stain purple R. vesiculiferum, and R. neocaledonicum (e.g., Sieverding et al. 2014;Al-Yahya'ei et al. 2017;Błaszkowski et al. 2018

Discussion
The new AM fungus Rhizoglomus venetianum can be easily distinguished from all other Rhizoglomus spp. by the combination of spore size, color, and spore wall structure. It forms a four-layered spore wall, such as R. natalense, R. custos, and R. proliferum, which form either smaller spores (R. proliferum; < 75 μm), spores of a different spore wall structure (R. custos), or less-pigmented spores (R. natalense; light yellow to bright yellow spores; see identification key above). Morphologically, it resembles R. intraradices, which lacks, as also known for R. irregulare, the innermost, fourth spore wall layer, which is always present in R. venetianum. R. venetianum is a fungus isolated from a very toxic environment, containing high levels of heavy metals, especially Zn, Pb, and Cu, the latter occurring at very high concentrations, compared with other metal polluted environments in which studies on AMF diversity were performed (Turnau et al. 2001;Vallino et al. 2006;Abdel-Azeem et al. 2007;Khade and Adholeya 2009;Long et al. 2010;Alguacil et al. 2011;Hassan et al. 2011;Krishnamoorthy et al. 2015). Interestingly, also another new species in the genus Rhizoglomus (R. custos) was isolated from a naturally heavy metal-polluted environment, i.e., the bank side of the Rio Tinto River in southern Spain (Cano et al. 2009), showing high levels of Fe (2 g/L), Mg (1.3 g/L), Cu (390 mg/L), Zn (280 mg/L), and Mn (100 mg/L). R. custos, as R. venetianum, forms a multi-stratified wall structure, a trait that might represent a possible barrier to toxic elements for the spores and might help the survival of the species in very toxic environments. Other species in the genus Rhizoglomus, such as R. clarum or R. intraradices, have been shown to possess different attributes linked to the ability of developing strategies, such as avoidance, compartmentalization, or resistance to stress, allowing them to live and grow in environments with high levels of heavy metals (Ferrol et al. 2009). Such traits, accompanied with the attitude of Glomeraceae to colonize roots by hyphal fragments or pieces of mycorrhizal roots rather than by germinated spores, might have helped also R. venetianum to be more competitive and widespread in the roots of S. inaequidens plants collected on the island. Interestingly, on Sacca San Biagio, R. venetianum appeared to grow entirely intraradically, since no spores of any AMF species were previously detected within ashes (Bedini et al. 2010). We can speculate that AM fungal species, living in very disturbed soil, may complete their life cycle in a privileged ecological niche (within plant roots), avoiding the direct contact with toxic compounds. In our previous study on AMF diversity occurring on Sacca San Biagio (Bedini et al. 2010), the most abundant sequence types detected in the roots of three plant species collected on the island corresponded to R. irregulare/R. fasciculatum and to R. intraradices. Since SSU sequences encompassing the V3-V4 region were not analyzed in this work, we are not able to confirm the occurrence of R. venetianum within the VeGlo8 cluster of our previous work. The genus Rhizoglomus, to which the identification key to species is proposed in this work, is the prevalent genus in studies on AMF diversity occurring in HMcontaminated sites, since it was dominant in colonized roots of Fragaria vesca L. collected in zinc wastes in southern Poland (Turnau et al. 2001), in a HM-contaminated soil along South Tyne River (Whitfield et al. 2004), in a chemical industry contaminated site in Northern Italy (Vallino et al. 2006), and in soils of mines in Asia (Zarei et al. 2008;Yang et al. 2015;Sun et al. 2016;Wu et al. 2010;Wei et al. 2015;Park et al. 2016) and Europe (Alguacil et al. 2011;del Mar Montiel-Rozas et al. 2017;Sánchez-Castro et al. 2017).
In conclusion, Sacca San Biagio represents an interesting site, contributing to our knowledge of AMF diversity in extreme habitats. Future works will disclose the concentrations of heavy metals able to inhibit the growth of the isolate R. venetianum IMA10, compared with other AMF species and isolates, and the occurrence of the new fungus in other environments.