By: Virginia Sebastián, Carlos Garrido-Allepuz, José María Lezcano, Marcos Garayar and Manuel Li of Helix BioS; Pedro Delvasto of the Universidad Industrial de Santander (Colombia) and Guillermo Shinno of Anabi.

Abstract

In this paper, 13 water and soil samples were taken at sites affected and unaffected by mining activity, in a real mining environment in the Peruvian highlands, specifically in the Anama Pit of the Anabi S.A.C. mining company located in the district of Huaquirca, province of Antabamba, Department of Apurimac. A distinction was made between samples directly affected by mining activity (pit) and samples from areas not subject to mining activity (natural environment), but within the vicinity of the mine. Genetic material (environmental DNA) was extracted from the samples and subjected to NGS, specifically 16S ribosomal RNA, which allows identification of bacteria. Using bioinformatics techniques based on metagenomics, the bacteria present in the samples were taxonomically identified to genus and species level. Using biostatistical procedures, diversity indices, bioindicators and other ecological data of interest were established in the samples. The results show the presence of iron and sulfur oxidizing microorganisms, associated with acid water generation processes within the mine, as well as numerous genera with the capacity to work in anaerobiosis, which can act as sulfate-reducing and fermenting microorganisms, generating possible synergies, which could be used for the design of bioremediation procedures within the mine itself. The study demonstrates the importance of applying metagenomics within the set of environmental actions that accompany mining activity.

Introduction 

Before undertaking any mining project, it is essential to carry out a thorough environmental impact assessment, including a biological baseline. This comprehensive process spans from the exploration stage to mine closure, and includes reserve calculation, design of mining operations and mitigation of negative impacts(Rodríguez-Luna et al., 2022). These detailed assessments identify potential impacts on local ecosystems and provide a framework for implementing preventive and corrective measures, thus ensuring responsible and sustainable management of mining. 

In the context of an environmental impact assessment for mining projects, biological aspects play a crucial role in the evaluation of potential impacts. In developing a biological baseline, it is essential to address aspects such as the diversity and distribution of plant and animal species, including endangered or endemic species. In addition, key habitats, such as forests, wetlands or aquatic ecosystems, as well as biological corridors and breeding or migration areas, should be considered. Studies should also include analyses of water, soil and sediment quality, as well as monitoring of flora and fauna during different seasons and cycles. Identification and assessment of the direct and indirect effects of mining activities on these biological aspects are essential to establish effective mitigation and conservation strategies (Gwimbi & Nhamo, 2016). 

Ignoring or overlooking the microbiological aspects of an ecosystem when establishing a baseline study for a mining project can have significant implications, usually of a negative nature. Microorganisms play a vital role in biogeochemical cycles, organic matter decomposition, soil fertility and water quality (Dabolkar et al., 2023; Ezeokoli et al., 2020; Sari et al., 2023; Yuan et al., 2021). In addition, some microorganisms can be indicators of environmental health and the presence of contaminants. By not considering microbiological aspects, there is a risk of underestimating or ignoring the potential impacts of mining activities on the microbiota of soil, water and aquatic ecosystems. This could lead to a lack of understanding of the actual effects and failure to implement adequate mitigation measures. In addition, the consequences may extend to human health and the long-term sustainability of the mining project. Therefore, it is critical to include a microbiological assessment in the baseline for a comprehensive and accurate understanding of the environmental impacts of mining. 

Metagenomics is a scientific discipline that focuses on the study of genetic material present in environmental samples, such as soils, sediments and water. Through advanced DNA sequencing techniques, metagenomics makes it possible to analyze and characterize microbial diversity in an ecosystem without the need to individually culture each microorganism (Offiong et al., 2023). In the context of baseline studies in mining projects, metagenomics is an invaluable tool. It allows to detect and describe the composition and function of microbial communities in ecosystems affected by mining, as well as to identify resistant microorganisms or indicators of contamination(Romero et al., 2021). This provides essential information for assessing environmental impacts and developing effective mitigation strategies, while providing a more complete and accurate picture of ecosystem health and resilience to mining activities. Metagenomics, by unveiling the microbial potential and its response to mining, becomes a valuable tool to promote responsible and sustainable management of these projects(Sari et al., 2023).

However, metagenomics not only allows for the refinement of biological baseline studies in the early stages of the mining project. It is worth asking whether it can be applied to existing projects and what useful information it can provide when making environmental management decisions during mining operations. In the context of mining operations, continuous environmental monitoring and the implementation of mitigation measures are essential to minimize impacts on ecosystems (Offiong et al., 2023). In addition, making critical decisions during progressive mine closure and environmental remediation is vital to ensure effective reclamation. In this sense, metagenomics presents itself as an invaluable tool by providing detailed and up-to-date information on microbial diversity and function at mining sites. 

Using next-generation DNA sequencing techniques, metagenomics makes it possible to identify heavy metal-resistant microorganisms, assess natural remediation capacity and design customized ecosystem reclamation strategies (Dabolkar et al., 2023). This accurate information helps to make informed decisions regarding the treatment of contaminated water, the implementation of soil covers or technosoils and the optimization of remediation practices. By leveraging metagenomics, mining companies can raise their level of innovation and commit to improving the environment and surrounding society. The recovery of ecosystem services becomes more effective and sustainable, thus promoting responsible mining and contributing to long-term sustainable development(Garris et al., 2016).

In a previously published paper, we demonstrated the power of metagenomics to explain, from a comprehensive metagenomic perspective (Sancho et al., 2022). A proprietary data analysis methodology was established, which in turn allowed the analysis of environmental DNA data collected from the water treatment system of a Chinese copper mine (Yuan et al., 2021) and from the soils of a former coal mine in South Africa (Ezeokoli et al., 2020). The analysis presented (Sancho et al., 2022) indicated how the microbiological populations of the ecosystem stand as faithful markers of its health and, in addition, how the increase in the capacity to provide ecosystem services of the soils of an area rehabilitated after mining can be measured quantitatively, using metagenomic tools. Despite this good background, to date there is still no data available on biodiversity or ecosystem analysis of microenvironments in mining areas of Peru.

Taking into account the above, this applied research work presents, for the first time, the microbiological characterization of different microenvironments within a mining operation in full production in Peru. Specifically, this is the metagenomic analysis of water and soil sampled directly from mining disturbed areas and surrounding unaffected areas, located in the Anama Pit of the Anabi, mining company located in the district of Huaquirca, province of Antabamba, Department of Apurimac (Peru). With samples from unaffected areas, we intend to establish a microbiological baseline, which will allow to differentiate the effects of mining in this particular environment. The results obtained made it possible to establish various existing ecosystemic relationships and to identify bioindicators of the environmental contamination process within the pit. The study shows how metagenomics is a novel tool for effectively visualizing microbial ecosystems in a mining-affected environment and for realistically assessing the health of these ecosystems. 

Objectives 

Given the above, the purpose of this work was the application of metagenomics in a real mining environment of the Peruvian highlands, considering the following three specific objectives: 

(I) Validate the applicability of metagenomics as a powerful tool to determine the impact of mining activity on the microbiological biodiversity of soils and waters within the operation.

(II) Identify relevant native microorganisms involved in geochemical alteration processes leading to the generation of acid waters.

(III) Identify, through bioprospecting, those native microorganisms that can be used for the design of in situ bioremediation solutions.

Development and data collection

Samples were taken of both contaminated and uncontaminated water and soil. The "uncontaminated" category was considered to be areas not affected by mining activity but located within the mine site, while the samples classified as "contaminated" belonged to sites directly affected by mining activity. For contaminated water (AC), samples were taken from two different sites in duplicate, and for uncontaminated water (ANC), samples were taken from a single site in triplicate. For the contaminated soil (SC), three different pit sites were sampled, while for the uncontaminated soil (SNC) samples, the same area was sampled in triplicate. Thus, a total of 13 samples were obtained for processing. The sample volume was 50 ml for soils and 250 ml for water. Table 1 shows the list of samples collected, as well as the coordinates of the collection site both in the original coordinates (UTM) and in the coordinates in Degrees (DMS). Figure 1 illustrates the sampling sites and their characteristics.

DNA extraction was carried out by means of specific kits for environmental samples, using the Qiagen DNeasy PowerSoil and DNeasy PowerMax Soil kits, following the protocols established by the manufacturer in the respective manuals. Qiagen's kits for environmental samples include the Inhibitor Removal Technology, capable of removing different types of acids, polyphenolic compounds, heavy metals and other types of elements that may be present in the samples.

Bacterial metagenomic studies are performed by analyzing the hypervariable regions of the prokaryotic 16S rRNA gene. The regions that best discriminate the microbiota, in general, are hypervariable region 3 (V3) and hypervariable region 4 (V4)(Wang & Qian, 2009). For this project, the region from the beginning of V3 to the end of V4, the amplicon known as V3-V4, was used. To carry out the construction of libraries for both regions, we proceeded according to the standard Illumina protocols and library kits. The sequencing process was carried out in two phases, followed by quality control. The first phase is sequencing in Illumina Miseq 2x300, where the libraries were loaded in cartridges (flow cells), in order to obtain an average result between 100,000 to 150,000 reads per sample and amplicon, verifying that the average number of reads per sample in this project was 102,235, as a quality control of the process. Finally, more than 90% of the bases and readings far exceeded the threshold of 20 on the Phred scale, with the average quality being close to 30 in this project. Subsequently, the data were subjected to bioinformatic analysis, leading to the creation and clustering of taxonomic units present in the samples(Callahan et al., 2017; Edgar, 2018), also called ZOTU (zero-radius operational taxonomic unit). These ZOTU are an approximation to the diversity and composition of the bacterial populations in the sample and each one represents, individually, each of the possible taxa present in the samples. The taxonomic assignment of the ZOTU is fundamental, as this step allows to really see which bacterial groups are present in the sample. Traditionally, in classical 16S analyses, sequences are compared with others present in a general database, predicting to which taxa each ZOTU may belong (predictive analysis). For this, the ZOTU obtained are compared with pre-established entries in these databases. In the case of this project, two databases were used, one for general use (GreenGenes) (DeSantis et al., 2006) and another one specific for mining(Helix Bioinformatics Solutions S.L, 2021), which is a Helix BioS in-house development. This double filtering of taxonomic data allowed to obtain the best possible taxonomic resolution. 

Next, two data cleaning steps were performed. The first step was to exclude from the analyses both ZOTU whose sequences belonged to chloroplasts or mitochondria, in order to keep those sequences genuinely prokaryotic and those ZOTU not identified at the phylum level. And the second step consisted of filtering in an unspecific way all those ZOTU whose total abundance was less than or equal to 25 reads in the set of samples, thus reducing the technical noise generated by the sequencing artifacts. From the set of filtered ZOTU, subsequent analyses of microbiological diversity and composition were performed (Garris et al., 2016; Hugerth & Andersson, 2017). For the description-analysis of the biodiversity and microbial composition of samples, the statistical software R v3.6.2 (R Core Team, 2019) and the libraries vegan 2.5-7 (Dixon, 2003), phyloseq 1.30.0 (McMurdie & Holmes, 2013)were used, while the DESeq2 1.26 library was used for the analysis of bioindicators(Love et al., 2014). The following alpha diversity indices were calculated with the samples: Richness (S), Chao1 Index (Chao1), Shannon-Wiener Index (H), Phylogenetic Diversity Index (PD), Inverse Simpson's Index (1/D). 

An important aspect in the analysis of the microbiota is the study of the microbial composition, i.e., which taxa and how often they are present in the samples. For this purpose, we worked with two types of abundances: the absolute observed abundance or count (n) and the relative observed abundance or proportion (%).

Presentation and discussion of results 

Diversity is an emergent property of biological communities that relates to the variety within them. This attribute is the expression of two components: the number of species present in the community, richness, and equity, which describes how the abundance of individuals is distributed among the different species that make up the community (Tuomisto, 2012). In such a context of amplicon or metabarcoding analysis, these indices are estimated from the cleaned and filtered table of ZOTU. Table 2 shows the results of the biodiversity analysis using various alpha diversity indices. The first two indices conceptualize the term richness, these indices were observed richness (S) or total number of ZOTU counted and an index based on estimated richness, Chao-1, specifically, on average richness was estimated at 72 (SD 140) ZOTU and at 72 (SD 140) ZOTU for Chao-1. The sample with the highest richness, clearly superior to the rest, was SNC1-2 with 531 ZOTU, the second and third richest samples were: AC2-1 with 95 ZOTU and SC1-1 with 75 ZOTU. The samples with the lowest observed richness were SC2-1, SC3-1 and ANC1-1 with 10, 10 and 14 ZOTU, respectively. The rest of indices represent the diversity in the sample, for these indices the following mean values were obtained for the set of samples: Shannon index (H) was 2.70 (SD 1.00); for the PD index, which also takes into account the phylogenetic distances between ZOTU, it was 12.36 (SD 11.95); finally, for 1/D it was 16.69 (SD 33.87). Also the same sample that presented a high richness, SNC1-2, also presented the highest values for the rest of the diversity indices studied. As for the multivariate analysis with the alpha diversity indices with respect to the study conditions, first, the normality of data was checked, and when this requirement was not met, non-parametric techniques such as Kruskal-Wallis and Wilcoxon were applied to evaluate possible associations between these biological indices and the experimental conditions, correcting for multiple comparisons. Statistically significant differences between contaminated samples (AC+SC) and non-contaminated samples (ANC+SNC) were evaluated independently of the sample support material (water or soil), and no significant index was found. Therefore, we proceeded to compare within the same support, looking for differences with respect to contamination, finding no differences either in richness or diversity between AC vs ANC and between SC vs SNC.

Apart from alpha diversity indices, there are other ecological indices, called beta diversity indices, which evaluate biodiversity among groups of samples defined by experimental variables (Gotelli & Colwell, 2001). In this case, the possible association between the microorganisms detected and the type of sample (AC/ANC/SC/SNC) was studied in order to detect differences in terms of microbial profile using the weighted UniFrac phylogenetic dissimilarity index. With the distance matrix, the Permanova test was performed, carrying out 999 permutations and correction for multiple comparisons by FDR (p<0.05). Likewise, the homoscedasticity of variances was tested multivariate and, once the distance matrix was obtained, the ordination method known as principal coordinate analysis or PCoA was applied. The groups were homogeneous (p = 0.340), however, no differences were detected between the microbial communities of the four conditions evaluated (p = 0.550), as corroborated in Figure 2, which shows the arrangement of the samples in the PCoA ordination system (Figure 4), no groups of samples are observed according to their characteristics, as can be seen the SC samples are far from each other showing that they have diverse microbial profiles, as for the water samples (AC and ANC) except for one duplicate, they are grouped together but together with non-contaminated soil samples (SNC). The maximum explained variability was 60%. Although the variability explained by the first two components was high, this does not generate clear sample groupings. In conclusion, the richness obtained is to be expected for this type of ecosystems, characterized by not having many different species, it would be expected that in the uncontaminated samples it would be higher since, in principle, they are "natural" samples, however, this is not the case, there is a lack of richness except for sample SNC1-2, which, in this context, could be considered as an outlier. The same happens with diversity, since no differences between conditions are observed and some of these indices are associated to a particular condition and can be used as ecological indicators. The same result was obtained from the study of biodiversity using the Weighted UniFrac index, ruling out the possibility that the categories present a sufficiently differential microbial profile.

An important aspect of microbiota analysis is the study of microbial composition, i.e., which taxa and how often they are present in the samples (Lin & Peddada, 2020). For this purpose, we worked with two types of abundances: the absolute observed abundance or count (n) and the relative observed abundance or proportion (%). The absolute abundance observed was calculated by taking into account the count of readings for each taxon detected at the different taxonomic levels evaluated. This count is obtained by summing the readings of ZOTU belonging to the same taxon. The relative abundance observed was obtained by estimating the proportions of each taxon, that is, by dividing the count obtained for each taxon per sample by the total count of readings in that sample. For the total calculation of readings per sample, taxa without informative taxonomy were excluded from the calculation, i.e., the estimated proportions were the valid proportions. The relative abundance observed in mean proportions was represented visually by stacked bar charts, each bar being a sample, for the seven taxonomic orders, in percent. For the taxonomic orders of phylum, class, order, family, genus and species, a new category was also created that included all those taxa that presented frequencies of less than 1% in the totality of the samples, so the new category was named "Others: <1%". Figures 3 to 8 below show the results of the microbial composition of the 13 samples analyzed.

At the kingdom level, the abundance of readings was distributed between two kingdoms: Archea (0.03%) and Bacteria (99.97%) taking almost all of the reads. Archaea were only detected in the uncontaminated soil samples, specifically in the SNC1-2 sample, the one with the highest richness. At the phylum level, the most abundant were Actinomycetota (37.55%), Bacillota (=Firmicutes) (34.24%), and Proteobacteria (16.99%), the rest of the phyla had a representation of less than 5% in the total set of samples. A total of 23 different phylla were detected. The AC and SC samples presented a higher mean percentage of abundance of the phylum Bacillota (41.14% and 34.66%, respectively) being the first phylum in these categories, unlike the uncontaminated water and soil samples (ANC and SNC) whose first phylum was Actinomycetota with 47.8% and 32.56%, respectively. In terms of class, up to 58 different classes were detected, highlighting: Actinomycetes (36.95%), Bacilli (28.49%), Gammaproteobacteria (8.4%), Alphaproteobacteria (3.76%), Clostridia (3.56%), Betaproteobacteria (3.47%). The rest showed mean abundances below 3%. A total of 77 different orders were detected, with Actinomycetales (38.02%), Caryophanales (17.64%), Lactobacillales (5.78%) and Pseudomonadales (4.87%) standing out in average abundance; the rest showed an average proportion of less than 5%. At the family level, the families Pseudonocardiaceae (28.62%), Bacillaceae (10.98%), Nocardiopsaceae (8.22%) and Lactobacillaceae (4.73%) stood out. A total of 109 families were detected. The rest of the families showed mean abundances below 5%. In terms of genera, a total of 98 different genera were detected, the most abundant being the genus Prauserella (36.87%), present in AC samples (45.30%), in ANC samples (60.25%) and in SNC samples (37.22%), but in very low abundance in SC samples (1.91%). Followed by other genera but with an overall lower mean abundance, Lactobacillus (5.77%), this genus being higher in the SC samples (19.14%) but not in the AC, ANC and SNC samples (3.71%, 0.89% and 0.00%, respectively); the genera Citrobacter (4.42%), Propionibacterium (3.49%), Clostridium (3.36%), Corynebacterium (3.20%), Hydrogenophaga (3.16%), Pseudomonas (3.02%), the remaining genera represent less than 3% on average in the total set of samples. Finally, a total of 97 different species were identified, being the species Citrobacter farmeri (14%) the one with the highest average abundance, followed by the species Rubrobacter bracarensis (11.08%) and the species Hydrogenophaga soli (7.36%), the rest of the species obtained average abundances lower than 5%. It should be noted that, at the species level, the degree of taxonomic assignment was 25%, which is why the genus level has been chosen for the rest of the abundance analysis. 

As for interesting genera, related to acidified environments, that could be detected were: Acidibrevibacterium (0.25%) [acidophilic isolated from mine drainage water], Acidiferrimicrobium (0.25%) [acidophilic isolated from coal mine drainage water], Acidiphilium (0.29%) [acidophilic isolated from mine drainage water], Acidisphaera (0.24%) [acidophilic isolated from mine drainage water and hot springs], Acidithiobacillus (0.87%, specifically the species A. ferrivorans as a sulfate-oxidizer), Anoxybacillus (0.89%) [isolated from acidic hot springs], Desulfosporosinus (0.04%) [sulfate-reducing strict anaerobe], Gallionella (0.15%, species of this genus are able to fix CO₂ from saline mineral solutions containing iron sulfide), Metallibacterium (0.16%) [acid-tolerant facultative anaerobe isolated from an acidic biofilm of a pyrite mine] and Sulfobacillus (0.01%) [acidophilic isolated from sulfide-bearing gold concentrates].

Abundance analysis is one of the main methods used to determine differences in microbial composition between conditions or groups of samples and to identify microbial taxa associated with certain environmental, biological and/or clinical factors; such highly differentiating taxa are called bioindicators. The DESeq2 library was applied to detect possible candidate phylotypes as bioindicators of the AC, ANC, SC, SNC conditions, by comparing categories two by two to determine which genera were found to be differentially abundant in the condition of interest with respect to the study condition, applying a negative binomial model fitted to massive amplicon sequencing data, with an estimation of the effect measure using the geometric mean, applying a local model with a poscount, with correction for multiple comparisons using FDR (p<0.05) and setting the LFC (logarithm of the Fold-Change [FC] as effect measure) to zero (Table 3; Figures 9, 10 and 11) (Lin & Peddada, 2020).

Figure 12 shows the mean relative abundance (%) for each category (AC/ANC/SC/SNC) of the genera that were found to be significant in the two previous statistical analyses for the contaminated areas, as well as certain genera associated with mining environments of interest.

In conclusion, through abundance analysis it has been possible to identify certain genera significantly associated with contaminated areas, which could be proposed as candidates as bioindicators of these environmental conditions and which will have to be evaluated with greater precision with a greater number of samples, as well as collecting samples from unsampled areas that present similar characteristics to the samples from Anama Pit and "natural" areas away from the radius of action of the mining activity in the area.

Comments and implications of these results for a more effective environmental management in the mining system

The results shown, although they correspond to the analysis of a few samples from limited environments of the Anama Pit, show the power of metagenomics to describe the ecology and biodiversity of the microenvironments of a mining system. The implications of these findings are multiple, since they allow (Yuan et al., 2021) to identify bioindicators to describe undesirable ecosystemic effects in the mining system, to glimpse how this knowledge can be accompanied by the design of new remediation processes for environments affected by mining, to better follow up on mine closure processes or to complement the environmental monitoring currently being carried out.

In the first instance, interesting results were found regarding the biodiversity of the mining environments analyzed. In general terms, the richness and diversity of these environments were low, possibly because they are oligotrophic (poor in nutrients), psychrophilic and extreme in altitude and acidity. Despite what might be thought a priori, most of the samples showed similar biodiversity, regardless of whether they were "contaminated" or "uncontaminated" soil or water samples, as explained in section previous and shown in Figure 2. The greatest difference between samples was verified in the "contaminated soil" samples taken from different points of the Anama Pit, which reflects the great distance between the different sampling points, as shown in Figure 1. However, differences were observed in some of the duplicate samples, obtained in principle from the same points, as was the case of samples AC2-1 and SNC1-2, the latter being the one with the highest biodiversity. These differences could be due to the sampling technique or the selection of the specific sampling site. Therefore, these divergences indicate that, in future projects, the number of samples should be substantially increased to establish the statistical significance of this type of variation and, per protocol, to identify and define in a cohesive, coherent and systematic manner the classification of zones (e.g., both "contaminated" and "non-contaminated" or "natural"), in order to reduce the inherent observational bias of a subjective classification of the sampling site. This last recommendation can be extended to all types of variables or factors that involve an observer or observers as a method of classification. 

In addition, the technique allowed the identification, with sufficient statistical differentiation, of various microbiological indicators, which, in the absence of data related to the physicochemical conditions of the samples received (e.g. pH, mineralogy, chemical composition, among others), give an account of the acidification processes that could be taking place in the waters of the Tajo Anama. It is worth mentioning that the "contaminated water" (AC) samples analyzed in this study come from the pit water collection system, which converge to two ponds or pools built in the lower part of the system, as shown in the red dots in Figure 1. 

These samples showed the presence of acidophilic microorganisms responsible for metal sulfide oxidation processes, which work in consortium to catalyze the oxidation of pyrites and other minerals containing reduced sulfur, which consequently generates sulfuric acid, which when diluted in water produces acidity and solubilization of the heavy metals present in the surrounding rocks (Mendez-Garcia et al., 2015). Thus, as can be seen in the lower left part of the satellite image shown in Figure 1, the point where these waters finally converge is altered by oxidation phenomena, typically caused by the flow of acidic waters of mining origin. Table 8, which summarizes the bioindicators identified in this study, shows that both in the "contaminated soils" (SC), i.e. solid samples taken directly from Anama Pit, and in its associated "contaminated waters" (AC), there is the presence of bacteria of the Acidithiobacillus genus, directly responsible for the oxidation of pyrite, and Metallibacterium, a genus highly adaptable to mining systems and responsible for the formation of biofilms that facilitate the colonization of rocks rich in sulfides (Haferburg et al., 2022). Both types of microorganisms could catalyze the oxidation of the pit walls once the mine is depleted and increase the generation of acid water. Figure 11 shows graphically the significant presence of these two genera in contaminated soils and waters of the Anama Pit. It should be noted that the rest of the bacterial genera identified as bioindicators of contaminated sites have not been reported in mining environments before, so their role in these poorly studied ecosystems is still unknown. Other bacterial genera identified were not indicated as bioindicators after statistical analysis, however, they are known to play active roles in the sulfide-type mineral alteration processes(Mendez-Garcia et al., 2015), as is the case of the genera Gallionella (iron oxidizer) and Sulfobacillus (sulfur oxidizer), both present in the "contaminated waters" (AC samples).

Figure 9 also shows that there are few characteristic and common bioindicators for both "uncontaminated waters" (ANC) and "uncontaminated soils" (SNC), these are the bacterial genera Fusobacterium, Leuconostoc and Acidisphaera. Of these three, the last one (Acidisphaera) has indeed been found in acid water systems(Kadnikov et al., 2016), while the other two, Fusobacterium and Leuconostoc, are anaerobic or facultative anaerobes, which actively participate in fermentative processes related to the lactic acid cycle (Ouamba et al., 2022), and are typical of the rumen and intestines of livestock (cattle, sheep, horses, etc.). These last characteristics coincide with the fact that the sampling of "uncontaminated" soils and water was carried out in an area that, upstream, shows evidence of livestock activity, since corrals of a cattle ranch were identified, as shown in Figure 1.

The metagenomics technique does not stop there, for as shown in the series of Figures 3 through 8, the relative abundance of microorganisms in each sample can be accurately known, from the phylum level down to the genus and species levels. These results can be thought of as an "X-ray" showing most of the players in the microecosystem of the samples. This opens the door to what is known as bioprospecting, that is, the identification of microenvironments within the mine itself that can serve as "donors" of microbes with interesting functionalities to carry out bioremediation procedures (Romero et al., 2021). The remediation procedures carried out in this way are considered to have a low environmental impact, since they involve species native to the site, making it unnecessary to bring or purchase commercial microbial inoculums for such work. However, it should be clarified that these bioprospecting processes should be carried out in a large number of sites in the mine and its surrounding areas not affected by the mining process, in order to increase the chances of finding useful microorganisms. In the case of this pilot study, only 4 sampling points and a total of 13 samples were analyzed.

Although there are many ways to perform bioremediation operations of acid waters or to avoid their generation, most of these activities are carried out without an accurate knowledge of the microorganisms available in the system. In the particular case of this study, a multitude of heterotrophic bacterial genera were identified, capable of functioning in anaerobiosis or microaerophilia (Anoxybacillus, Clostridium, Turicibacter, Fusobacterium, Corynebacterium, Prevotella, among others), which could be consorted with other genera, also identified in the samples, which are capable of fermenting organic nutrients in these environmental conditions (Lactobacillus, Leuconostoc, etc.) and which are capable of reducing sulfates to sulfides (Desulfosporosinus). With this combination of microorganisms, it is possible to consider the design of water treatment systems in anaerobic systems (Anekwe & Isa, 2023), to facilitate the precipitation of heavy metals and avoid water pollution; for example, artificial wetlands or technosoil systems, fully adapted to the microbiology of the site.

Conclusions 

1. This pilot study demonstrated the potential of metagenomic analysis based on environmental DNA extraction in samples from a real mining environment to establish the biodiversity of these environments, diagnose the tendency to generate negative environmental impacts (e.g. acid waters), establish biomarkers between sites affected and not affected by mining activity and, finally, perform bioprospecting in order to design ecological solutions to environmental problems that must be addressed at the time of the closure of mining operations. 

2. However, it is necessary to deepen this study by increasing the number of samples and establishing comparisons with the physicochemical parameters of the sampled site, a key aspect to describe in detail the analyzed ecosystems, which were not addressed in this study. 

3. Studies with these characteristics are key for mining companies to be able to establish more sustainable technologies and environmental management systems, with which they can efficiently accompany their mine closure processes. In the same way, the use of this environmental sampling methodology is recommended to enrich environmental impact studies, so that the presence or absence of sulfur and iron oxidizing microorganisms in mining areas that remain unexploited can be compared much more effectively, in order to establish their possible future behavior once mining begins and help prevent environmental problems in future mining projects.

References

Anekwe, I. M. S., & Isa, Y. M. 2023. Bioremediation of acid mine drainage – Review. Alexandria Engineering Journal, 65, 1047–1075. https://doi.org/10.1016/j.aej.2022.09.053

Callahan, B. J., McMurdie, P. J., & Holmes, S. P. 2017. Exact sequence variants should replace operational taxonomic units in marker-gene data analysis. The ISME Journal, 11(12), 2639–2643. https://doi.org/10.1038/ismej.2017.119

Dabolkar, S., Furtado, I. J., & Kamat, N. M. 2023. Pioneer Studies on Metagenomic Evaluation of Diversity of Microbial Community in Banded Iron Formation (BIF) from Iron Ore Mining Belt of Goa, India. Geomicrobiology Journal, 40(5), 427–433. https://doi.org/10.1080/01490451.2023.2184883

DeSantis, T. Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E. L., Keller, K., Huber, T., Dalevi, D., Hu, P., & Andersen, G. L. 2006. Greengenes, a Chimera-Checked 16S rRNA Gene Database and Workbench Compatible with ARB. Applied and Environmental Microbiology, 72(7), 5069–5072. https://doi.org/10.1128/AEM.03006-05

Dixon, P. 2003. VEGAN, a package of R functions for community ecology. Journal of Vegetation Science, 14(6), 927–930. https://doi.org/10.1111/j.1654-1103.2003.tb02228.x

Edgar, R. C. 2018. Updating the 97% identity threshold for 16S ribosomal RNA OTUs. Bioinformatics, 34(14), 2371–2375. https://doi.org/10.1093/bioinformatics/bty113

Ezeokoli, O. T., Bezuidenhout, C. C., Maboeta, M. S., Khasa, D. P., & Adeleke, R. A. 2020. Structural and functional differentiation of bacterial communities in post-coal mining reclamation soils of South Africa: bioindicators of soil ecosystem restoration. Scientific Reports, 10(1), 1–14. https://doi.org/10.1038/s41598-020-58576-5

Garris, H. W., Baldwin, S. A., Van Hamme, J. D., Gardner, W. C., & Fraser, L. H. 2016. Genomics to assist mine reclamation: A review. Restoration Ecology, 24(2), 165–173. https://doi.org/10.1111/rec.12322

Gotelli, N. J., & Colwell, R. K. 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters, 4(4), 379–391. https://doi.org/10.1046/j.1461-0248.2001.00230.x

Gwimbi, P., & Nhamo, G. 2016. Benchmarking the effectiveness of mitigation measures to the quality of environmental impact statements: lessons and insights from mines along the Great Dyke of Zimbabwe. Environment, Development and Sustainability, 18(2), 527–546. https://doi.org/10.1007/s10668-015-9663-9

Haferburg, G., Krichler, T., & Hedrich, S. 2022. Prokaryotic communities in the historic silver mine Reiche Zeche. Extremophiles, 26(1), 2. https://doi.org/10.1007/s00792-021-01249-6

Helix Bioinformatics Solutions S.L. 2021. Base de datos ambiental para estudios de metagenómica de bacterias (16S) y Hongos (ITS). IDENTIFICADOR 2105187862705.

Hugerth, L. W., & Andersson, A. F. 2017. Analysing Microbial Community Composition through Amplicon Sequencing: From Sampling to Hypothesis Testing. Frontiers in Microbiology, 8. https://doi.org/10.3389/fmicb.2017.01561

Kadnikov, V. V., Ivasenko, D. A., Beletsky, A. V., Mardanov, A. V., Danilova, E. V., Pimenov, N. V., Karnachuk, O. V., & Ravin, N. V. 2016. Effect of metal concentration on the microbial community in acid mine drainage of a polysulfide ore deposit. Microbiology, 85(6), 745–751. https://doi.org/10.1134/S0026261716060126

Lin, H., & Peddada, S. Das. 2020. Analysis of microbial compositions: a review of normalization and differential abundance analysis. Npj Biofilms and Microbiomes, 6(1), 60. https://doi.org/10.1038/s41522-020-00160-w

Love, M. I., Huber, W., & Anders, S. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550. https://doi.org/10.1186/s13059-014-0550-8

McMurdie, P. J., & Holmes, S. 2013. phyloseq: An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data. PLoS ONE, 8(4), e61217. https://doi.org/10.1371/journal.pone.0061217

Mendez-Garci¬a, C., Pelaez, A. I., Mesa, V., Sanchez, J., Golyshina, O. V., & Ferrer, M. 2015. Microbial diversity and metabolic networks in acid mine drainage habitats. Frontiers in Microbiology, 6. https://doi.org/10.3389/fmicb.2015.00475

Offiong, N.-A., Edet, J., Shaibu, S., Akan, N., Atakpa, E., Sanganyado, E., Okop, I., Benson, N., & Okoh, A. 2023. Metagenomics : an emerging tool for the chemistry of environmental remediation. Frontiers in Environmental Chemistry, 4(May), 1052697. https://doi.org/10.3389/fenvc.2023.1052697

Ouamba, A. J. K., Gagnon, M., LaPointe, G., Chouinard, P. Y., & Roy, D. 2022. Graduate Student Literature Review: Farm management practices: Potential microbial sources that determine the microbiota of raw bovine milk. Journal of Dairy Science, 105(9), 7276–7287. https://doi.org/10.3168/jds.2021-21758

R Core Team. 2019. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Rodríguez-Luna, D., Encina-Montoya, F., Alcalá, F. J., & Vela, N. 2022. An Overview of the Environmental Impact Assessment of Mining Projects in Chile. Land, 11(12). https://doi.org/10.3390/land11122278

Romero, M. F., Gallego, D., Lechuga-Jiménez, A., Martínez, J. F., Barajas, H. R., Hayano-Kanashiro, C., Peimbert, M., Cruz-Ortega, R., Molina-Freaner, F. E., & Alcaraz, L. D. 2021. Metagenomics of mine tailing rhizospheric communities and its selection for plant establishment towards bioremediation. Microbiological Research, 247 (September 2020). https://doi.org/10.1016/j.micres.2021.126732

Sancho, F., Sebastián, V., Garrido-Allepuz, C., Lezcano, J. M., Acosta, Á., Garayar, M., & Delvasto, P. 2022. La metagenómica como técnica novedosa para el análisis de impactos ambientales por efluentes y el seguimiento en el tiempo de la rehabilitación del suelo en zonas mineras desde una perspectiva microbiológica integral. Revista Minería, 539 (Agosto), 8–27. https://revistamineria.com.pe/pageflipx/mineria/539/8/

Sari, E., Nugroho, A. P., Retnaningrum, E., & Prijambada, I. D. 2023. Literature Review and Experiment: Diversity of Bacteria in Forest, Revegetated Post-Mining Land, and Active Tin Mining with A Metagenomic Approach. Indonesian Journal of Science and Technology, 8(1), 19–48. https://doi.org/10.17509/ijost.v8i1.50737

Tuomisto, H. 2012. An updated consumer’s guide to evenness and related indices. Oikos, 121(8), 1203–1218. https://doi.org/10.1111/j.1600-0706.2011.19897.x

Wang, Y., & Qian, P.-Y. 2009. Conservative Fragments in Bacterial 16S rRNA Genes and Primer Design for 16S Ribosomal DNA Amplicons in Metagenomic Studies. PLoS ONE, 4(10), e7401. https://doi.org/10.1371/journal.pone.0007401

Yuan, Q., Wang, P., Wang, C., Chen, J., Wang, X., & Liu, S. 2021. Indicator species and co-occurrence pattern of sediment bacterial community in relation to alkaline copper mine drainage contamination. Ecological Indicators, 120 (September 2020), 106884. https://doi.org/10.1016/j.ecolind.2020.106884