Variations of fungal communities in lead–zinc tailings located in Northwestern China

Abstract Investigation of microbial community structures is critical for situ-remediation of heavy metal contaminated soils. For this purpose, we used high-throughput 18S rRNA sequencing method to analyze soil fungal communities in the Shanping village (SPC) and Yanzibian (YZB) tailings situated in northwestern China. The relationships of the fungal community and the environmental factors, including heavy metals and soil chemical properties were analyzed. The results showed that the fungal community richness and diversity in the soils were both trending as control soil > contaminated soil. The compositions and relative abundances of the fungal populations in the two tailings were different, however the dominant fungal phyla of which were almost the same, mainly including Ascomycota, Basidiomycota, Calcarisporillomycota, Mortierellomycota, and Rozellomycota. Further, the Canonical correlation analysis and spearman correlation analyses revealed that Cd, Pb, Cu, Zn, Cr, TP, NO3 −-N, TN, and/or pH were predominantly positive correlation factors for the most abundant fungal phyla Ascomycota and Basidiomycota. Together, we have sorted out certain fungal species, such as s_unclassified_p_Ascomycota, s_unclassified_c_Sordariomycetes, s_Talaromyces_solicola, and s_Cutaneotrichosporon_curvatus, which are probably tolerant to heavy metals in these specific tailing soils. Overall, our results have provided an initial perspective of the fugal community variations of these specific tailing ponds.


Introduction
Heavy metal contamination in soil is one of the most serious environmental problems in the world (Li et al. 2015(Li et al. , 2016. The development of social economy has accompanied by the enhancement of mining activities, as a result, serious environmental problems were triggered. In the mining area, a series of processes, such as mining, metallurgy, transportation and waste treatment, would generate tailing dams, slag heap, etc. Due to washing of rainwater, heavy metals in the tailing dams might be released and transferred, resulting in heavy metal pollutions in the nearby environment (Mohammed and Suresha 2017;Barsova et al. 2019; Ove cka and Tak a c 2014). These heavy metals may finally be enriched in human bodies through various food chains, thus are detrimental threats for humankinds (Ayangbenro et al. 2017;Zhao et al. 2022). The accumulation of heavy metals in these soils may also change the abundance, diversity and activity of microorganisms, leading to extensive environmental degradation (Chen et al. 2014). Thus, heavy metal pollution has become one of the most serious problems in modern society.
Heavy metal pollution in soils may also affect the surrounding ecosystem (e.g., biomass and species diversity), and soil chemical properties Li et al. 2021b). On the other hand, soil microorganisms are components of the soil ecosystem, secretions (enzymes) of microorganisms may promote decomposition of organic matters and formation of humus in soils, such that changes of microbial community structures can in turn reflect the pollution extent of soils. Therefore, the characteristics of the microbial community can be used as potential environmental indicators for the contaminated soils (Di Cesare et al. 2020). In recent years, many studies have shown the evolution of the microbial community caused by heavy metals under long-term contamination (Lin et al. 2019;Gong et al. 2021;Wen et al. 2022). Song et al. (2018) found that the microbial communities changed with heavy metals (Cd, Cu, Zn) concentrations and soil physicochemical properties (pH, TN, TC); Li et al. (2020) found that Cr, Pb, and Zn all negatively affected the abundance of Nitrospirae, Bacteroidetes, and Verrucomicrobia. Studies of soil microorganisms in metal-mining areas were mainly focused on the effect of heavy metals on the microbial community structures (Beeck et al. 2015;Deng et al. 2018). For example, Kuperman et al. (1997) found that in soils with multiple heavy-metal contaminations, the number of microbial species was 80% lower than that in non-polluted soil, also there were studies showing that low concentration of heavy metals could stimulate the growth of microorganisms, while high concentration of which could lead to significant decrease of soil microbial biomass (Arshadi et al. 2020;Bai et al. 2021). Wu et al. (2022) found that soil pH, total phosphorus, total nitrogen, available potassium, available phosphorus, and the bacteria abundance and diversity all gradually increased with the decrease of the heavy metals.
Various techniques have been tried to remove the heavy metals in the soils, including chemical, physical, electrical, bioremediation and phytoremediation methods (Zhao et al. 2022, Li et al. 2016). Among which bioremediation using microorganisms as the "cleanser" is an environment-friendly and economic biological remedial technique. It is worthy to note that microbes cannot degrade the heavy metals directly, however they can use different mechanisms such as precipitation, biosorption, and enzymatic transformation to change heavy metals to less toxic forms, which are more stable, less mobile or inert (Geetesh 2017;Chang et al. 2008;Wuana et al. 2011;Ojuederie et al. 2017). Fungi are widely used as biosorbents for the removal of toxic metals, due to their excellent capacities for metal uptake and recovery (Dursun et al. 2003;Fu et al. 2012). Many studies have shown that active and lifeless fungal cells could both play significant roles in adhesion of inorganic chemicals (Tiwari et al. 2013;. In this study, we used 18S rRNA high-throughput sequencing technique to investigate the fungal community structures in SPC and YZB lead-zinc tailings located in the Qinling Mountains, Ningqing county, Shaanxi province of northwestern China. The YZB and SPC tailings are derived from one lead-zinc mine, which has a mining history of 37 years. The main components of the two tailings are shales and phyllites, which are loose, highly water-permeable and prone to be eroded. Further, vegetation coverage in both tailing ponds was extremely low . From these backgrounds, it could be speculated that the two tailings have been exposed to long-term and severe heavy metal pollutions. We were aimed to explore whether and how the fungal communities could be changed by the heavy metals, and we expected that through this study, the most heavy-metal tolerant fungi could be screened out, thereby providing a theoretical basis for the upcoming bioremediations of these tailing ponds.

Collection of soil samples
Soils were sampled on December 21, 2019 from Shanping village (SP) and Yanzibian valley (YZB) tailings situated in Ningqiang county, Shaanxi province. The pollution source of the two tailings is from the nearby lead-zinc reservoirs which is approximately 4 km away from them. When the lead-zinc metals were refined, the slag heaps were transported and piled to the Shanping village and Yanzibian valley thus forming lead-zinc tailings. The linear distance between the two YZB and SPC sampling locations is about 3 km. Specifically, in this work, a total of 7 sites in and around the tailing ponds were chosen for sample collections ( Figure 1; Table 1). As exhibited by other similar studies Gao et al. 2021aGao et al. , 2021b, we also chose a nearby unpolluted farmland soil (i.e., YZB_NT) as a control soil. For each site, the soil sample was collected at a depth of approximately 20 cm; 3 equal amount of soil samples (i.e., 3 biological replicates) were randomly collected at each sampling site. After the soil, plant root and debris were thoroughly mixed, a portion of the soil was air-dried, ground and screened, and then the soil chemical traits and heavy metal concentrations were determined. The other portion was stored in À80 freezer and used for further metagenomic sequencing.

Soil chemical property and heavy metal concentration measurements
Soil pH values were measured by a pH meter (PHS-3E, Inesa Instrument, China) (w/v ratio of soil and water was set at 1:2.5). Soil organic matter (SOM) was measured by titration with oxygen butyrate; total nitrogen (TN) was determined by Kjeldahl digestion method. Total phosphorus (TP) was determined by sodium hydroxide alkali fusion molybdenum antimony method using a (UV-1800, Shimadzu, Japan). Total potassium (TK) was determined by a flame photometer(FP640, Shjingmi, China). Soil nitrate nitrogen (NO 3 À -N) and ammonium nitrogen (NH 4 þ -N) were extracted by 2 M KCl solution (1:5 w/v) for 30 min, and the concentrations were measured using a flow injection autoanalyzer (Flowsys, Systea, Italy). For heavy metal measurements, the soil samples were air-dried and sieved with a 2-mm polyethylene mesh to remove large debris. As demonstrated by other similar studies (Gao et al. 2021b;Chen et al. 2020;Deng et al. 2015), we only analyzed the relationship between the total contents of the heavy metals and the fungal community, the bioavailability of heavy metals in the soil was not considered. The heavy metal elements (Zn, Cu, Pb, Cr, and Cd) were detected by a flame atomic absorption spectrophotometer (TAS-990F, Persee, China), based on the principal that different metal concentrations have different absorptions of the resonance radiation corresponding to their gaseous ground state atoms.

DNA extraction and high-throughput sequencing
Total metagenomic DNA in each soil sample was prepared by using a Fast DNA Spin Kit for soil (MP, Santa Ana, USA), according to the manufacturer's protocol. The  concentration of the metagenomic DNA was detected using NanoDrop 2000 (Termo Scientifc, Wilmington, USA), and the DNA integrity was assessed by 1% (w/v) agarose gel electrophoresis. The primer pairs, i.e., ITS1F and ITS2R were used to amplify the ITS regions of the metagenomic DNAs. The sequence informations are as follows, ITS1F: 5 0 -CTTGGTCATTTAGAGGAAGTAA-3 0 , and ITS2R: 5 0 -GCTGCGTTCTTCATCGATGC-3 0 . The PCR reaction system was consisted of 5 lM of ITS1F and ITS2R primers, 0.2 lL of rTaq polymerase, 0.25 mM of dNTPs, 0.2 lL of BSA, 2 lL of 10 Â polymerase buffer, 10 ng of DNA template, with ddH 2 O supplemented to a final volume of 20 lL. The PCR reactions were run on a ABI GeneAmp 9700 thermocycler PCR system, the program used was as follows: 95 C, 3 min for initial denaturation; for each cycle, 95 C for 30 s, 55 C for 30 s, 72 C for 45 s, in total 35 cycles; 72 C,10 min for final extension. The PCR products showing bright main bands could be used for further experiments. PCR products were mixed in equidensity ratios according to the GeneTools Analysis Software (Version4.03.05.0, SynGene). Then, mixed PCR products were purified with EZNA Gel Extraction Kit (Omega, USA). Sequencing libraries were generated using NEBNextV R Ultra TM DNA Library Prep Kit for IlluminaV R following manufacturer's recommendations. The library quality was assessed on the Qubit @ 2.0 Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2100 system. The libraries were sequenced on an Illumina Hiseq 2500 platform and 250 bp paired-end reads were generated.

Bioinformatic and statistical analysis
By using USEARCH software (8.1.1831), initial sequencing reads were categorized into different operational classification units (OTUs), on the basis of ! 97% sequence similarities. The representative OTU sequences were then taxonomically classified using RDP classifier (V2.2, http://sourceforge.net/projects/rdp-classifier) according to the information from Silva data base (Silva 132, http://www.arb-silva.de), the threshold was defined at 0.7, if the classification values were lower than the threshold, they were categorized into "unclassified" . The indexes of Sobs, Shannon, Simpson, Ace, Chao1 and Coverage were used to evaluate the richness and diversity of soil fungal community, which were calculated by Mothur rank sum test (http://www.mothur.org/ wiki). The column charts showing fungal community composition of each soil sample were generated by R language tool. Non-metric multidimensional scaling (NMDS) analysis was performed to visualize discrepancies of the fungal communities among and within the sampling sites. This NMDS analysis was performed in the R environment (version 2.4-2) by using the subsampled OTU. The pluralistic relationships among the fungal community, the heavy metal concentrations, and the soil chemical characteristics were evaluated by Canonical correlation analysis (CCA) by using Canoco 5.0 software. Spearman correlation analysis was used to determine the correlation between the most abundant 12 fungal phyla and the chemical properties, or the heavy metals of the soil samples, and it was performed by QIIME 2 software. The significant differences of the soil chemical properties, the heavy metal concentrations, the fungal community diversity and richness and the relative abundance of predominant fungal phyla/species of the soil samples were analyzed by SPSS (version 23.0), respectively. When p < 0.05, the results were considered as statistically significant different.

Accession number
Pyrosequencing reads are deposited in Genbank under accession number PRJNA750323.

Heavy metal contents of the soil samples
The heavy metal concentrations of the soil samples, including Yanzibian farmland soil (YZB_NT), Shanping village (SPC) and Yanzibian (YZB) tailing soils, are shown in Table 2. Overall, the heavy metal contents were different to some extent of the 7 sampling sites. The YZB_1 site had the most severe heavy metal pollution extent. Specifically, in this site, the Zn, Cu, Pb, and Cd concentrations were all significantly higher than the other soil samples. YZB_2 was considered as the second severely contaminated site, in which the Pb content was the highest, and the Zn, Cu, and Cr contaminations were all second-ranked. Further, although less serious than those of the YZB_1 and YZB_2 sites, the heavy metals contents of YZB_3, YZB_4, SPC_1, and SPC_2 also demonstrated considerable levels of pollutions, e.g., Zn and Cd contents in SPC_1 and SPC_2; Pb and Cd contents in YZB_3; and Cd content in YZB_4 were all times higher than the other soil samples. Intriguingly, the Cr contents in both SPC and YZB sites were remarkably lower than the other soil samples, implying that Cr might not be extensively generated during mining activities of these areas, or might have been absorbed by the surrounding ecosystem.

The chemical properties of the soils
In consistent with the data of the heavy metal concentrations, the chemical characteristics of the soil samples were also significantly differed (p < 0.01) (  values of respectively 0.37 and 1.13 g kg À1 , although they were geographically closed. On the whole, the differences of NH 4 þ -N, NO 3 À -N and TN among these samples were considerably small compared with those of other chemical properties. Moreover, besides that of TK, the values of all the chemical properties of the YZB_NT soil were inbetween those of the SPC and YZB tailing samples.

Fungal community diversity and richness of the soil samples
In total, 5068 OTUs were revealed after the 18S rRNA sequencing, the fungal community diversity and richness of the soils were then assessed according to the abundance of the resulting OTUs. The Chao1 and Ace indexes were used to estimate the number of OTUs in the community, which reflected the species richness of the community. The Shannon index positively reflected the alpha diversity of the community, while the Simpson index reflected the dominant level of the most common microbes in the whole community, and the greater the Simpson index, the lower the community diversity. Coverage represents the coverage rate of each sample sequence in the library. The results were summarized in Table 4. The highest Sobs, Ace, Chao1 and Shannon indexes were recorded in YZB_NT, which were significantly higher than those of the tailing soil samples (p < 0.05). Overall, the values of Sobs, Ace and Chao1 implied that the fungal community richness of these samples was ranked as: Based on the values of Shannon and Simpson (i.e., species diversity indexes), the fungal community diversity in these soils was ranking as follows: YZB_NT > YZB_1 > YZB_3 > YZB_2> SPC_1 > SPC_2 > YZB_4. Therefore, the fungal community diversity and richness in  the tailing soils were both inferior to those of the control farmland soil, which implied that the heavy metals have changed the fungal community structures.

Fungal community compositions of the soils
In total of 12 phyla, 45 classes, 118 orders, 297 families, 623 genera and 980 species were summarized from the 21 soil samples. The fungal community compositions and variations of these soils were subsequently assessed on the phylum level. The mean relative abundances of the predominant fungal phyla which mainly included Ascomycota, Basidiomycota, Calcarisporiellomycota, Mortierellomycota, Rozellomycota and Glomeromycota were significantly different (p < 0.05) of the 7 sites (Figure 2A). The most abundant fungal phylum in the YZB tailing soil was Ascomycota, with an average value of 76.24% ±18.38% detected; and this phylum was extremely abundant in the YZB_1 and YZB_4 sites, accounting for 85.79% ±16.44% and 92.26%±13.18%, respectively (64.46%±4.64% was detected in the control soil) (Table S1). Notably, except for Ascomycota, YZB_2 contained a large amount of Calcarisporillomycota (28.42% ±14.32%), which were nearly undetectable in other soil samples. In sharp contrast to the YZB tailing soils, Basidiomycota became the most abundant fungal phylum of the SPC soils, which reached 54.18%±30.19% and 74.47%±22.89% in SPC_1 and SPC_2 soils, respectively (only 26.17%±3.94% was detected in the control soil) ( Figure 2B; Table S1). Moreover, on the species level, the average relative amount of some species especially those of the Ascomycota and Basidiomycota phyla in the YZB and/or SPC soils were significantly increased ( Figure S1; Table S2). For example, s_Cutaneotrichosporon_curvatus and s_Cutaneotrichosporon_cutaneum respectively accounted for 26.29%±19.29% and 21.08%±9.78% of the whole fungal community in SPC_1, which were significantly higher than 0.00%±0.00% of which found in YZB_NT. 16.76%±12.36% and 12.55%±7.42% of s_unclassified_c_Sordariomycetes and s_Paraphoma_radicina were respectively in YZB_1, by contrast, only 0.96%±0.30% and 0.07%±0.05% of which were found in the control soil. The relative average abundance of s_unclassified_p_Ascomycota and s_Mortierella_alpina were respectively 20.85%±10.13% and 7.15%±1.42% of the YZB_2 site, which were significantly higher than 1.07%±0.51% and 0.26%±0.14% of which found in YZB_NT. Further, astonishingly, s_Talaromyces_solicola reached 71.42%±25.52% in YZB_4 (Table S2). Taken together, the above data indicated that the heavy metal contaminations have directly or indirectly altered the fungal community structures of the tailing soils, especially those of the YZB_1 and SPC_2 soils, as contaminations of these two sites were the most severe in the respective tailing ponds (Table 2).

Fungal community variations of the soils
Non-metric multidimensional scaling (NMDS) was used to show the combined effects of soil chemical properties and heavy metal contents on the diversity of fungal communities, which would imply community variations among the soils. In agreement with the above analysis, the fungal communities of the same sampling site had apparent clustering and high similarity, and those among the YZB tailing soils, the SPC tailing soils, and the YZB_NT soil were strikingly different, thus they could be classified into 3 clusters (Figure 3). In general, the fungal community structures of the YZB tailing soils were more similar to those of the YZB_NT soils (Figure 3). The similarities of the fungal community structures were most prominent between SPC_1 and SPC_2, since a certain extent of similarities were recorded from the negative axis. Further, in agreement with the above results regarding soil chemical properties and heavy metal contents (Tables 2, 3), the fungal community structures of the YZB soils, especially those of YZB_2 and YZB_3 were most widely distributed. Together with the results presented in Tables 2 and 3, it was inferable that the fungal community structures were affected by both heavy metals and other chemical characteristics of the soils.

Correlations between the fungal community and environmental factors
The above results suggested that the fungal communities in both tailing soils were significantly changed, probably triggered both by heavy metals and other soil chemical parameters, therefore it was necessary to reveal to which degree that individual environmental parameter could change the fungal community structures. For this purpose, canonical correlation analysis (CCA) was performed. The result is summarized in Figure 4. A total of 26.20% of fungal community diversity variation was explained (the first and second dimensions were 13.95 and 12.25%, respectively). The main influence factors included Cd, Pb, pH, SOM, TN, and TP. In detail, both the fungal community diversities of SPC_1 and SPC_2 were positively correlated with TP, SOM, pH, TK, Zn, Cu, Cr, and Cd, and negatively correlated with Pb, TN, NO 3 À -N and NH 4 þ -N.
Compared with those of the SPC tailing, the environmental parameters that influenced the fungal community diversities of the 4 YZB tailing soils were more diversified. In the situations of YZB_1, YZB_2, and YZB_3, the fungal community diversities were positively correlated with TN, TK, NO 3 À -N, Pb, Cu, Cr, Zn, and Cd, and negatively affected by TP, pH, SOM and NH 4 þ -N. However, for YZB_4, only one positive correlation factor, i.e., NH 4 þ -N was detected. Since the soil chemical properties and heavy metal contents were most similar between SPC_1 and SPC_2, and YZB_1 and YZB_2 (Tables 2,  3), the above result indicated that the fungal community structures in the tailing soils were synergetic results of all the environmental elements, and the closer the environmental properties were, the more similar the fungal community structures. Moreover, to uncover the correlation between the relative amount of individual predominant fungal phylum and each specific environmental factor, a Spearman heatmap correlation analysis was executed. The results demonstrated that all the heavy metals and soil chemical parameters had significant correlation with the most dominant fungal phyla (i.e., Basidiomycota and Ascomycota; Figure 5). Generally, all the environmental factors (including heavy metals and soil chemical parameters) rendered strongly positive effects on both of the Basidiomycota and Ascomycota phyla. In detail, except for Talaromyces, Acidomyces, Fusicollal, and Chaetomium, a majority of other fungal genera (e.g., Cladosporium, Paraphoma, Periconia, Cyphellophora, and Emericellopsis) belonging to the Ascomycota phylum were positively correlated with Cd, TP, Pb, Cu, Zn, and Cr ( Figure S2). Additionally, some genera were significantly positively correlated with NO 3 À -N and TN (e.g., Cyphellophora, Emericellopsis, Didymellaceae and unclassified Eurotiales; Figure S2). Further, except for that of Boletales, the significant correlation factors of the genera (e.g., Vishniacozyma and Cutaneotrichosporo) in the Basidiomycota phylum were almost the same as those of Ascomycota ( Figure S2). Overall, consistent with the CCA analysis results, the environmental factor NH 4 þ -N was the relatively weaker factor which only influenced a minority of fungi.

Discussion
In this study, comprehensive profiles of Chemical properties, heavy metals, fungal communities in 2 lead-zinc tailing (i.e., SPC and YZB tailings) ponds, farmland samples from tailings in Ningqiang County, Shaanxi Province were created. Our results revealed that the heavy metal pollution in the tailing ponds affects the chemical properties and the fungal community structure of the soil. Although SPC and YZB tailings reservoirs are located in the same metallogenic belt of Ningqiang County, there are significant differences in chemical properties and heavy metal concentration of the two mining areas due to different smelting processes (Tables  2, 3). The average contents of TN, TP, and TK were much higher in the YZB tailing than those in the SPC tailing, and the distribution trend is similar to the content distribution of Cu, Zn, Cr, and Pb in tailings ponds. Therefore, it can be inferred that the presence of Cu, Zn, Cr, and Pb plays a key role in determining the soil chemical properties at the sampling sites of these tailings. Similarly, Gao et al. (2021a) showed that the average SOM, TN, TP, and TK contents were much higher in the SP tailing than those in the LJP tailing, and this variation trend was resembled to that of the Cd content. Further, it was obvious that the average heavy metal concentrations of YZB_1 and YZB_2 were higher than those of other sampling site, by contrast, the average pH of these two samples is relatively low. This indicates that heavy metal content is significantly negatively correlated with soil pH. This result was in a high agreement with some former studies Gong et al. 2021;Rogiers et al. 2021). For example, Li et al. (2021) showed that pH value of sediments was negatively correlated with Cu, Zn, Cr, Ni, Pb (p < 0.05); Gong et al. (2021) demonstrated that pH value had a significant (p < 0.001) negative relationship with Cu, Zn, Cr, Cr(VI) in soils which were suffered from long-term heavy metal contaminations. Moreover, our results also implied that when the pH values of the tailing ponds (e.g., YZB_1 and YZB_2) were acidic, pH was negatively correlated with SOM and positively correlated with TN, NH 4 þ -N, and TP. It has also been shown that pH value was negatively correlated with SOM, and the correlation coefficient was À0.354 (p < 0.05) (Hu et al. 2021). Studies in contaminated soils nearby an abandoned realgar mine have shown that there was a negative correlationship between pH and SOM in acidic samples, but no significant correlation between pH and SOM in neutral or alkaline samples (Xiao et al. 2022). In addition, the values of all the chemical parameters of the control farmland soil except for TK were lower than those of the YZB and SPC tailing soils (Table 3), as can be seen from the analysis results of heavy metals (Table 2). It seemed that different levels of the heavy metal pollution may have changed the soil chemistry to some extent. At the same time, it may be that heavy metals in tailings reservoirs cause the change of soil microbial community, and then lead to the change of soil chemical properties. It's well-known that the soil microbial community can be affected by a variety of factors, including geographical factors, soil chemical characteristics such as pH, TN, TP, TK, SOM, composition of microbes and other biological factors Wen et al. 2020). Normally if these factors remained stable, the microbial community structure would remain stable. In this work, the soil chemical properties, heavy metal concentrations and fungal community structures of the YZB and SPC tailing soils were respectively analyzed. The results showed that the fungal community structure in these heavy-metal contaminated soils changed significantly compared with the control farmland soil (Tables 4, S1; Figures 2-5). We found that the diversity and richness of fungi in tailings soil were lower than those in control farmland soil (Table 3; Figures 2, 3), indicating that the heavy metal pollution have changed the community structures of fungi. However, the relative abundance of fungi phylal Ascomycota and Basidiomycota in SPC and YZB tailings soils was significantly increased compared with the control farmland soil (Figure 2). Usually, Ascomycota is the most abundant and widely distributed soil fungi in the world (Al-Sadi et al. 2017;Wang et al. 2019;Lin et al. 2019), and their abundance will tend to be stimulated by environmental factors. For example, Lin et al. (2019) found that the fungal phyla Ascomycota, Basidiomycota, Chytridiomycota and Zygomycota showed strong correlation with heavy metal contaminations in paddy fields; in which, Ascomycota was found to have the highest tolerance to heavy metals, and its abundance increased significantly even under moderate levels of heavy metal pollution. This research also found that Ascomycota was positively correlated with Cu and Pb at moderate and light heavy-metal contamination levels, however was in a significant negative correlation with Cr at severe level (Poorter et al. 2016). The study on the microbial communities of an abandoned copper mine at Colmenarejo in the northwest of Madrid (Central Spain) also showed that the fungi in the Ascomycota phylum especially those in Aspergillus genus, have increase significantly in areas with high Cd, Cu, Fe, Mn, Pb, Zn, and As concentrations (Nemergut et al. 2011;Zeng et al. 2020). Therefore, our results were in high agreement with previous studies (Yuan et al. 2019;Yang et al. 2020;Cao et al. 2020). Regarding the mechanisms of the heavy-metal tolerant ability of Ascomycota and Basidiomycota, it has been suggested that species in both of the fungal phyla are able to produce high level of phytochelatins, which will thus protect cells against heavy metals, such as cadmium (Mart ınez-Soto et al. 2020). Moreover, the Ascomycetes fungi were shown to be able to express ZIP protein families, which were important for zinc transport (Chodak et al. 2013). It would be the reason why Ascomycota was found to increase significantly at high level of Zn contamination in our study, namely, YZB_1 contained the highest level of Zn and the most abundant Ascomycota among all the samples. In summary, our results strongly suggest that heavy metals in YZB and SPC tailings cause changes in fungal community structure, and further strengthened the suggestion that fungal phyla Ascomycota and Basidiomycota had strong tolerance to heavy metals.
Regarding the impact of individual environmental parameter on the fungal communities of the YZB and SPC tailing soils, the Canonical correlation analysis (CCA) and Spearman heatmap correlation analysis results showed that all of the 12 tested heavy metals and soil chemical properties had significant correlation with certain dominant fungal genera belonging to the Ascomycota and Basidiomycota phyla, in which Cd, Pb, Cu, Zn, Cr, TP, NO 3 À -N, TN, and/or pH were predominant correlation factors ( Figures   4, 5). These results indicated that both soil chemical and heavy metals traits will alter microbial community structures. Specifically, in our study, we found that the average relative abundance of Ascomycota in the YZB tailing soils was much higher than that in the SPC tailing soils, while the opposite result was observed for Basidiomycota.
Noticeably, the average pH in YZB was much lower than that in SPC. Thus, as demonstrated by the Spearman heatmap correlation analysis ( Figure 5), there were significant positive and negative correlationships between pH and the relative abundances of Basidiomycota and Ascomycota, respectively. This result was in high consistent with certain previous studies (Frossard et al. 2017;Passarini et al. 2022;Xiao et al. 2022). For instance, fungal phyla such as Basidiomycota and Ascomycota in an ongoing antimony mine area were also shown to be postively and negatively responding to the pH value, respectively (Sey and Belford 2021;Sharp et al. 2014). In fact, soil pH has been considered to be a main factor that would drive microbial community structure changes in the mining area. However, it is worthy to note that the effects of pH on the microbial community are comprehensive and might not be consistent in different sampling site (Leff et al. 2018;Ren et al. 2018). In addition to pH, soil fungal community have also been shown to be sensitive to changes of environmental variables such as SOM and TN (Lin et al. 2019). In our analyses, SOM was found to be negatively correlated with Calcarisporiellomycota, Zoopagomycota and Glomeromycota. Similar results were also revealed in some former studies, for example, Nie et al. (2018) studied the effect of different fertilizers on fungal communities, and found that higher TN and SOM contents in soils would decrease the abundance of Ascomycota, but increase that of Basidiomycota. Chen et al. (2020) found that TN and SOM were negatively correlated with the fungal phyla Cosmospora, but positively correlated with Pyrenochaetopsis and Myrothecium.
As expected, except for soil chemical properties, heavy metals were strong driving forces of the fungal community variations in the tailing soils. As illustrated by the CCA analysis, the heavy metals Cd, Cr, and Pb were the most prominent factors of the fungal community diversities in YZB_1, 2, and 3 sampling sites. Further, the Spearman heatmap correlation analysis showed that Zoopagomycota was shown to be negatively responding to the Cu, Zn, and Cr. Our results were in high agreement with some previous reports. Lin et al. (2019) reported that Cr, Cd, and Cr had adverse effects on the relative abundances of soil fungal phyla, such as Basidiomycota and Glomeromycota. Fern andez-Calviño and Bååth (2016) suggested that the fungal growth was not negatively affected by Cu addition in short term; however certain fungal phyla such as Ascomycota and Basidiomycota would be enriched with the increasement of Cu concentrations.  found that the fungal phyla Acremonium, Guehomyces and Gibberella showed strong negative correlation with Cd in paddy fields. Narendrula-Kotha and Nkongolo (2017) reported that Ascomycota and Zygomycota were negative with Cu and Ni in a mining region under long-term metal exposure. Taken together, our results suggested that heavy metals in the tailing soils would probably change the soil chemical properties, and that heavy metals themselves and soil chemical properties would jointly shape the fungal communities (Jia et al. 2010;Jiang et al. 2019).

Conclusions
In conclusion, our work demonstrated that: (1) there are significant differences in heavy metal content between YZB and SPC tailings, in which YZB_1 and SPC_2 were the most severely contaminated sites for the respective tailing; consistently, the soil chemical properties in these two tailings were also strikingly different, implying that the heavy metals have influenced the soil chemical properties; (2) the fungal community diversity and richness in the same tailing were similar, however they were prominently different between the two tailings, which were projecting as YZB > SPC, this result was echoing to the fact that the YZB tailing was more severely contaminated and exhibited extreme soil conditions; (3) The relative abundances of Ascomycota, Basidiomycota, Calcarisporillomycota, Mortierellomycota, and Rozellomycota in contaminated YZB and SPC soil was significantly increased compared with that in control farmland soil, indicating that the soil chemical properties and heavy metals have jointly changed the fungal communities, specifically, Cd, Pb, Cu, Zn, Cr, TP, NO 3 À -N, TN, and/or pH were predominantly positive correlation factors for the most abundant fungal phyla Ascomycota and Basidiomycota. Together, our analyses have sorted out the most tolerant fungi in the two specific heavy-metal contaminated tailings, which will provide useful insights for further bioremediations of these areas. Overall, in this manuscript, we conducted a preliminary study on the ecosystems of YZB and SPC tailings soils, and successfully identified the heavy metal resistant fungal species of Ascomycota and Basidiomycota, e.g., s_unclassified_p_Ascomycota, s_unclassified_c_Sordariomycetes, s_Paraphoma_radicina, s_Cutaneotrichosporon_curvatus, and s_Cutaneotrichosporon_cutaneum.

Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.