Assessment of Physicochemical, Microbiological and Toxicological Hazards at an Illegal Landfill in Central Poland

This study aimed to assess the physicochemical, microbiological and toxicological hazards at an illegal landfill in central Poland. The research included the analysis of airborne dust (laser photometer), the number of microorganisms in the air, soil and leachate (culture method) and the microbial diversity in the landfill environment (high-throughput sequencing on the Illumina Miseq); the cytotoxicity (PrestoBlue) and genotoxicity (alkaline comet assay) of soil and leachate were tested. Moreover, an analysis of UHPLC-Q-ToF-UHRMS (ultra-high-performance liquid chromatography-quadrupole-time-of-flight ultrahigh-resolution mass spectrometry) was performed to determine the toxic compounds and microbial metabolites. The PM1 dust fraction constituted 99.89% and 99.99% of total dust and exceeded the threshold of 0.025 mg m−3 at the tested locations. In the air, the total number of bacteria was 9.33 × 101–1.11 × 103 CFU m−3, while fungi ranged from 1.17 × 102 to 4.73 × 102 CFU m−3. Psychrophilic bacteria were detected in the largest number in leachates (3.3 × 104 to 2.69 × 106 CFU mL−1) and in soil samples (8.53 × 105 to 1.28 × 106 CFU g−1). Bacteria belonging to Proteobacteria (42–64.7%), Bacteroidetes (4.2–23.7%), Actinobacteria (3.4–19.8%) and Firmicutes (0.7–6.3%) dominated. In the case of fungi, Basidiomycota (23.3–27.7%), Ascomycota (5.6–46.3%) and Mortierellomycota (3.1%) have the highest abundance. Bacteria (Bacillus, Clostridium, Cellulosimicrobium, Escherichia, Pseudomonas) and fungi (Microascus, Chrysosporium, Candida, Malassezia, Aspergillus, Alternaria, Fusarium, Stachybotrys, Cladosporium, Didymella) that are potentially hazardous to human health were detected in samples collected from the landfill. Tested leachates and soils were characterised by varied cyto/genotoxins. Common pesticides (carbamazepine, prometryn, terbutryn, permethrin, carbanilide, pyrethrin, carbaryl and prallethrin), quaternary ammonium compounds (benzalkonium chlorides), chemicals and/or polymer degradation products (melamine, triphenylphosphate, diphenylphtalate, insect repellent diethyltoluamide, and drugs (ketoprofen)) were found in soil and leachate samples. It has been proven that the tested landfill is the source of the emission of particulate matter; microorganisms (including potential pathogens) and cyto/genotoxic compounds.


Introduction
In Poland, almost 123 million tonnes of solid waste were generated in 2020, of which 10.7% was municipal solid waste (MSW) (13.1 million tonnes), which is 342 kg of this type of MSW per capita [1]. Despite the growing number of supporters of the circular economy, striving to minimise the amount of waste generated and the use of unavoidable waste as a recycling resource, the problem of vast amounts of MSW continues to be urgent. Following European Union legislation [2], proper waste management includes a fivestep waste management hierarchy. First of all, waste should be prevented, reused, then recycled, other forms of recovery, and, finally, neutralised (e.g., by landfilling). Despite these guidelines, a large proportion of business waste is still landfilled. The amount of waste stored in Poland in 2020 amounted to 46 million tonnes and occupied a total area of almost 8 thousand hectares. MSW was stored in 271 dedicated landfills. At the same time, over 2000 existing illegal "wild" landfills with a total area of almost 2 km 2 were inventoried [1]. A similar problem with illegal landfilling has been reported in Estonia, Italy and France. More facilities of this type are located in France, the Czech Republic, Croatia, Russia and Hungary. On the other hand, statistical data show that the largest number of illegal landfills in Europe is in Albania (9046), Slovakia (8334) and Romania (7174) [3]. Illegal landfill sites pose a potential hazard of severe pollution. Unlike the orderly municipal landfills, they are totally unprotected. Neither a geomembrane layer separates them from the substratum, waterproof sealing, nor a separate protection zone [4]. Illegal landfill sites arise most frequently in the vicinity of MSW management plants, wastelands, roadsides, byways, unfenced private plots (particularly on the city outskirts), in forests and by the water courses [5].
The literature data indicate that the problem of illegal MSW disposal is global [6][7][8]. The harmfulness of such objects seems obvious. It manifests itself mainly in the contamination of surface and ground water's soils, the destruction of aesthetic and landscape values and the exclusion from the use of agricultural and forest areas used for MSW storage [9][10][11]. The adverse effects depend mainly on the type and quantity of MSW and the duration of its residence. Illegal landfill sites are a source of threat resulting from the migration of toxic substances from deposited MSW (or leachates) into the soil profile and due to the blowing and settling of dust from this place [5,12]. Illegal landfill sites are also the source of soil pollution with heavy metals [4]. Elevated concentrations of copper, nickel, zinc, chromium and mercury were registered in the soil in their vicinity [13,14]. Individual reports indicate that the spreading of pathogenic microorganisms carried by birds, rodents and other small animals, which seek food in such places, is also dangerous [5]. Among the numerous works on illegal landfill sites, it is difficult to find studies assessing the microbiological quality of air. Similarly, no comprehensive analyses would assess the content of harmful substances in soil, leachate and air in illegal landfills.
The research aimed to comprehensively assess the physicochemical, microbiological and toxicological hazards at an illegal landfill, including determining airborne dust, the number and types of microorganisms in the air, soil and leachate using classical methods of high-throughput sequencing on the Illumina Miseq platform. The cytotoxicity of soil samples and drainage leachate from the landfill against three human cell lines were tested. Moreover, analysis of a targeted and untargeted UHPLC-Q-ToF-UHRMS (ultra-highperformance liquid chromatography-quadrupole-time-of-flight ultrahigh-resolution mass spectrometry) was performed to determine the toxic compounds and microbial metabolites in soils and leachate samples from an illegal landfill site. The metagenomic analysis of air from illegal landfills has been performed in the presented study for the first time. Hitherto, metabolomic and toxicological studies for soil and leachate from a landfill environment have not been carried out so comprehensively.

Tested Landfill Sites
Air, water and soil samples were collected at illegal landfills in a town 25 km away from Łódź (area 21.6 km 2 , population approx. 12,500; central Poland, Łódź Province)). The area is a former clay excavation site in which there has been a MSW landfill since the 1970s; it closed around 2015 and was recultivated. The area was used illegally to store ballast from the sorting plant. MSW has been stored without appropriate permits for about 10 years.
The heap of MSW has a height of 218 m above sea level and a weight estimated at >5000 mg. The MSW is mixed with a predominance of plastics. The measurement sites are marked in Figures 1 and 2  1970s; it closed around 2015 and was recultivated. The area was used illegally to store ballast from the sorting plant. MSW has been stored without appropriate permits for about 10 years. The heap of MSW has a height of 218 m above sea level and a weigh estimated at >5000 mg. The MSW is mixed with a predominance of plastics. The measurement sites are marked in Figures 1 and 2 and described in Table 1.   Table 1. Table 1. Characteristics of sampling sites in an illegal dumpling.  1970s; it closed around 2015 and was recultivated. The area was used illegally to store ballast from the sorting plant. MSW has been stored without appropriate permits for about 10 years. The heap of MSW has a height of 218 m above sea level and a weight estimated at >5000 mg. The MSW is mixed with a predominance of plastics. The measurement sites are marked in Figures 1 and 2 and described in Table 1.   Table 1. Table 1. Characteristics of sampling sites in an illegal dumpling.

No. Description Sample No
A The pile of MSW on the hill on the south side. Numerous streams of leachates. The leachates were black, and the water was flowing down the hill towards the ponds (clay pits) Air 1 Soil 1 Soil 2 Soil 3 Figure 2. Sampling points at an illegal landfill; (A-F) described in Table 1. Table 1. Characteristics of sampling sites in an illegal dumpling.

No. Description Sample No
A The pile of MSW on the hill on the south side. Numerous streams of leachates. The leachates were black, and the water was flowing down the hill towards the ponds (clay pits) located approx. 12 m below. Water in ponds was class I purity for surface waters.

Microclimate and Airborne Dust Concentration Analysis
A VelociCalc ® Multi-Function Velocity Meter 9545 (TSI, Shoreview, MN, USA) thermoanemometer was used to establish the temperature, relative humidity and air-flow rate at the selected workstations. The particulate matter (PM) concentration was measured using a DustTrak™ DRX Aerosol Monitor 8533 portable laser photometer (TSI, USA). The detection range for particles ranging from 0.1 to 15 µm in size was between 0.001 and 150 mg m −3 . The measurements were obtained in triplicates for each location at the height of 1.5 m from the ground level. The sampling rate was set at 3 L min −1 and the sampling interval at 5 s.

Sampling of Air, Leachate and Soil
Air samples for the determination of the number of microorganisms were collected using a MAS-100 Eco Air Sampler (Merck Life Science, Warsaw, Poland) according to the EN 13098 standard [15]. Three separate air samples of 50-100 L each, were collected at a height of 150 cm from the ground. Atmospheric air samples were also collected 250 m from the landfill and 10 km (external background). The air for the determination of biodiversity was collected using AirPort MD 8 (Sartorius, Goettingen, Germany) from place C. This location was crucial due to the potential spread of bioaerosol to nearby areas (presence of a stocked lake at the bottom of the escarpment). Sterile gelatine filters (80 mm, 0.3 µL Sartorius, Germany) were subjected to continuous air-flow of 8000 L and then sent for DNA extraction. Leachates and soils were collected in sterile 120 mL plastic containers. The leachates were collected directly into the containers, and the soils were sampled with a metal spatula from a depth of 5-20 cm.

Determination of Microorganism's Number
The media summarised in Table 2 were used to determine the microbiological contamination of the air.
Moreover, the samples of soil (10 g) and leachate (10 mL) were suspended in 99 mL of sterile saline solution (0.85% NaCl). Dilutions were prepared from 10 −1 to 10 −7 in triplicates and then plated onto the above-mentioned media. In addition, TSC (Tryptose Sulfite Cycloserine Agar, Merck Life Science, Warsaw, Poland) was used to quantify the number of sulfate-reducing anaerobes.
Following incubation, the colonies were counted. The results were expressed in CFU m −3 (air) and CFU g −1 (soil) CFU mL −1 (leachate). The arithmetic mean of three independent repetitions was reported as a result.

Determination of Biodiversity
A modified method based on Genomic Mini AX Bacteria + set (A&A Biotechnology, Gdańsk, Poland) was used to extract genomic DNA. A FastPrep-24 device was used to perform additional mechanical lysis of the samples. Additional DNA treatment was performed with the Anti-Inhibitor Kit (A&A Biotechnology, Gdańsk, Poland) post-extraction. The Real-Time PCR was used to confirm the presence of bacterial DNA in the tested samples. The Real-Time PCR reaction was carried out in Mx3000P thermocycler (Stratagene, La Jolla, CA, USA), using SYBR Green dye as a fluorochrome.
The extracted DNA concentration was between 2 and 30 µg mL −1 . Universal primers amplifying the 16S rRNA bacterial gene's fragment and fungal ITS regions were used in the reaction [16,17]. The DNA eluates had been checked for their quality and quantity prior to library preparation. The libraries of V3-V4 and ITS amplicons were prepared according to the guidelines of 16S Metagenomic Sequencing Library Preparation Part #15044223 Rev. B. The Herculase II Fusion DNA Polymerase Nextera XT Index Kit V2 was used for the two-step PCR.
The libraries were performed and quality-checked according to the Illumina qPCR Quantification Protocol Guide and the sequencing Macrogen (Seoul, Korea). Then, the sequencing was performed using the paired-end technology on the Illumina MiSeq (2 × 300 bp) platform (Macrogen; Seoul, Korea).
Qualitative and quantitative taxonomic identification was carried out as described in [18], using CLC Genomic Workbench v. 12  100 µg/mL streptomycin/100 IU/mL penicillin (Merck Life Science, Warsaw, Poland) and 0.1 U/mL insulin (IEC-6). Caco-2, A-549 and Hep-G2 were purchased from Cell Line Service GmbH (Eppelheim, Germany), while IEC-6 from DSMZ German Collection of Microorganisms and Cell Cultures GmbH (Germany). The cells were then incubated at 37 • C with 5% CO 2 in the humidified atmosphere for 7-10 days to reach 80% confluence. Two to three times per week, the cells were washed with 0.1 M PBS (pH 7.2), and the medium was renewed. TrypLE TM Express (Thermo Fisher Scientific, Waltham, MA, USA) was used to detach the cells from the culture (37 • C, 6-12 min). Then the cells were centrifuged (307× g, 5 min) and decanted. Then, the pellet was re-suspended in a fresh culture medium. The cells were ready to use after performing a cell count by hemacytometer and determining cell viability by trypan blue test (it must have been at least 90%).

Preparation of Samples
Soil samples (1.0 g) were suspended in 10 mL of the appropriate cell culture-ready medium, mixed and then extracted for 40 min (160 r.p.m.) at ambient temperature. The primary soil concentration in each extract was 100 mg/mL. The pH of each extract was adjusted to neutral (pH 7.0 ± 0.2) to eliminate the influence of this parameter on the toxicity. Sterile 0.22 µm syringe filters (Membrane Solutions, Kent, WA, USA) were used to filter each extract twice. The leachate samples were only filtered through sterile 0.22 µm syringe filters. A series of dilutions were prepared.

Cytotoxicity and IC 50 Determination
The cytotoxicity of the water-soluble fractions of the soil samples and the leachate samples was assessed with PrestoBlue. A total of 5000 (Caco-2, Hep-G2, IEC-6) or 1000 (A-549) cells/well were placed in 96-well black flat-bottom plates and incubated for 24 h at 37 • C in 5% CO 2 . The next day, the medium was removed from the cells, and then dilutions of leachate samples and soil extracts were added. The final concentrations of the analysed soil extracts were (mg/mL): 3.13; 6.25; 12.5; 25; 50; 100, while for the leachate samples (mg/mL): 31.3; 62.5; 125; 250; 500. Cells in the culture medium were negative control. The samples were exposed for 72 h at 37 • C in an atmosphere of 5% CO 2 . After this, the samples were aspirated, and PrestoBlue (10% solution in PBS) was added to each well and incubated for a further 2 h at 37 • C in 5% CO 2 . A microplate reader (TriStar2 LB 942, Berthold Technologies GmbH & Co. KG, Bad Wildbad, Germany) was used to measure fluorescence (λ ex 560 nm; λ em 590 nm). IC 50 values were estimated from the resulting curves.

Single Cell Gel Electrophoresis Assay
The tested samples/dilutions were introduced into Eppendorf tubes along with A-549 cells in the amount of 1 × 10 5 cells/sample (the final volume was 1 mL). The negative control contained cells suspended in the culture medium. Cells incubated with 25 µM H 2 O 2 were a positive control. The samples were incubated for 60 min at 37 • C, centrifuged (15 min, 4 • C, 182× g), decanted and LMP (Low Melting Point) agarose was added at 37 • C. The suspension was spotted on warm NMP (Normal Melting Point) double-layered slides and covered with coverslips (hot plate ZF6 Premiere Slide Warmer). The samples were placed on a Chilling Plate for Comet Assay Slides (Cleaver Scientific) and allowed to solidify. Then, alkaline lysis was performed with the buffer (2.5 M NaCl, 1% Triton X-100, 100 mM EDTA, 10 mM Tris, pH 10) and incubated (60 min, 4 • C). The lysis buffer was decanted, and then the slides were flooded with the unwinding buffer (300 mM NaOH, 1 mM EDTA) (20 min, 4 • C); next, they were placed in an electrophoresis apparatus (CSL-COM20, Cleaver Scientific). The electrophoresis was performed in an electrophoretic buffer (300 mM NaOH, 1 mM EDTA, pH > 13) for 20 min at a voltage of 21 V and a current of 29 mA. The slides were neutralised in leachate, allowed to dry, and then were stained for 60 min at 4 • C with DAPI (1 µg/mL); then, comet analysis was performed under a fluorescence microscope (Nikon) at a magnification of 200× equipped with a camera (Nikon Digital Sight DS-U3) and with Lucia Comet v.7.0 software (Laboratory Imaging, Prague, Czech Republic). In each trial, 50 randomly selected comets were analysed based on the parameter determining the percentage of DNA in the comet's "tail". Results are presented as mean ± S.E.M.

Instrumentation
UHPLC-Q-ToF-UHRMS analysis was performed on a Bruker Elute UHPLC system operated by Hystar 3.3 software and a Bruker Impact II (Bruker Daltonik GmbH, Bremen, Germeny) mass spectrometer of ESI QToF-MS type. The column for the Elute system was Bruker Intensity Solo with C18 silica modification, 1.8 µm particles, 100 × 2.1 mm (length × diameter). The UHPLC column was thermostated and held at 40 • C. The first mobile phase was water:methanol 99:1 (v/v) with 5 mM NH 4 HCOO and 0.01% HCOOH. Phase B was methanol 5 mM NH 4 HCOO and 0.01% HCOOH. The injection volume was 5 µL. For Targetscreener bbCID measurements the B percentage was: 18.3% (1 min), 50% (2.5 min), 99.9% (14-16 min), 4% (16.1-20 min). The solvent flow was 0.2 mLmin −1 from 0 to 1 min, gradually changing from 0.2 to 0.223 mL min −1 from 1 to 2.5 min and from 0.223 to 0.400 µLmin −1 from 2.5 to 14 min; after this, the flow was 0.4 to 0.48 mL min −1 from 14 to 19 min and back to 0.2 mL min −1 from 19.1 to 20 min. The UHPLC separations for the AutoMSMS experiments were conducted similarly, flows and B phase percentages were identical as above, with the exception that 99.9% of B was held up to 25 min at 0.48 mL min −1 and 4% B was from 26.1 to 30 min. Internal calibration was based on ions from 10 mM sodium formate pumped into the ESI ion source with a syringe pump at an infusion flow rate of 0.12 mL h −1 (solvent: water: isopropanol 1:1 v/v). The calibration was performed automatically in Metaboscape using a high precision calibration (HPC) mode. The analyses in positive bbCID mode were made with the following parameters: pre-pulse storage time: 8 µs; transfer time: 60 µs m/z: 50-1000; capillary voltage: 4 kV; nebuliser: 4 bar; dry gas: 8 L min −1 ; drying gas temperature: 200 • C; hexapole voltage: 30 Vpp; funnel 1: 300 Vpp; funnel 2: 300 Vpp. In autoMSMS mode, the m/z range was 50-1500, and the CID (Collision-Induced Dissociation) energy value was 30 eV. The CID settings were: absolute area threshold: 5000 cts; active exclusion 2; isolation window: for m/z = 100-4, 300-5, 500-6, 1000-8. Targeted identification was made in TASQ (ver. 2021b) with the use of built-in libraries containing spectral details and additionally retention times. The untargeted annotations were performed in Metaboscape (ver. 2021b) with a criterion of mass deviation (∆ m/z) under 2 ppm and a mSigma value under 15 Statistical analysis was carried out with Statistica 13.1 (Statsoft, Tulsa, OK, USA). Descriptive statistics were calculated for all variables of interest. The numbers of microorganisms in the air, soil, and leachate were compared between the tested samples (ANOVA, α = 0.05). Tukey's post hoc procedure was used to compare the means (α = 0.05) in case a statistical difference was detected (p < 0.05).

Toxicological Analysis
A two-way analysis of variance (ANOVA) was conducted using OriginPro 6.1 (Northampton, MA, USA) software to evaluate the experimental data for genotoxicity testing. The signifi- cant differences between the means were compared using Scheffe's multiple comparison test, and they were regarded as significant at p < 0.05.

Metabolite and Pathway Analysis
All metabolites were analysed using the advanced MetaboAnalyst 5.0 online software to perform multivariate statistical analysis [21]. Metabolite profiles of leachate and soil samples from UHPLC-HRMS analysis were subjected to unsupervised Principal Component Analysis (PCA). The separation between two groups of leachates (2, 3 vs. 1, 4, 5) and two groups of soils (2, 4, 5 vs. 1, 3) was examined using Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA). The overall quality of the OPLS-DA models was assessed by examining the goodness of fit (R 2 Y) and the predictive ability of the models (Q 2 ). Potential features for group separation of leachate samples were subsequently identified by S-plot loading analysis of corresponding OPLS-DA models and based on the significance criterion of |p(corr)| > 0.5 and |p| > 0.05. The data were analysed using an independent t-test and fold-change analysis. p-values and false discovery rates (FDR; q-value) less than 0.05 and |Log 2 FC| > 1 were considered statistically significant.
All 2666 (for leachate) and 2413 (soil samples) identified compounds by LC-MS were analysed using MetaboAnalyst 5.0 software to identify the most relevant pathways. However, targeted pathway analysis was performed for 471 (leachate) and 413 (soil) compounds that were successfully well-annotated by HMDB or KEGG ID. Metabolic pathways with values > 0.1 were considered to be significant. Pathway enrichment analysis was performed based on the main class compounds.

Microclimate and Airborne Dust Concentration
The microclimate parameters of the illegal dumping are summarised in Figure 3. At location B, the lowest temperature (9.2 • C) and the highest relative humidity (51.5%) and air-flow rate (2.62 m s −1 ) were observed. The highest temperature (20.8 • C) and the lowest relative humidity (28.7%) and air-flow rate (0.28 m s −1 ) were observed for location G, which was an external control. The microclimate conditions in different locations were generally similar (for locations A-D, no statistical differences in all tested parameters were detected, and for E, only the temperature was different from locations A-D). Statistically significant differences were detected between tested locations and internal and external control (temperature between A-D vs. F and G, relative humidity for B vs. F and G, and air-flow rate between B and D vs. F and G). The diversified values of the microclimate parameters in the tested locations and both control sampling sites suggest that microorganism development conditions might differ in those locations.
The PM 1 size fraction (Figure 3d), i.e., particles with dimensions below 1 µm, constituted almost all the measured dust at the tested locations. Its share in the total quantity of the measured dust was between 99.89% and 99.99%. The PM 1 dust concentration at tested locations was similar, but statistically significant differences occurred either way. The highest values were observed at location B (0.0475 mg m −3 ) and the lowest ones at location F, which constituted an internal control (0.0436 mg m −3 ). Such results can be associated with air-flow velocities that were much higher for location B than for location F. Exposure to air pollution can cause significant harm to humans. To improve air quality, the European Union has developed a set of health-based standards and objectives for various air pollutants [22]. According to the Directive's requirements, the annual average concentration of dust with dimensions below 2.5 µm (i.e., the fraction containing the PM 1 fraction) should not exceed 0.025 mg m −3 . It means that the measured PM 1 concentration was almost twice as high as the environmental threshold independent of the sampling site.
with air-flow velocities that were much higher for location B than for location F. Exposure to air pollution can cause significant harm to humans. To improve air quality, the European Union has developed a set of health-based standards and objectives for various air pollutants [22]. According to the Directive's requirements, the annual average concentration of dust with dimensions below 2.5 μm (i.e., the fraction containing the PM1 fraction) should not exceed 0.025 mg m −3 . It means that the measured PM1 concentration was almost twice as high as the environmental threshold independent of the sampling site.

Number of Microorganisms in the Air, Leachate and Soil
The number of microorganisms in the air at the landfill is shown in Figure 4 and Table S1 (Supplementary Materials).

Number of Microorganisms in the Air, Leachate and Soil
The number of microorganisms in the air at the landfill is shown in Figure 4 and Table S1 (Supplementary Materials).
In the air in the tested landfill, the total number of bacteria ranged from 9.33 × 10 1 (G, external control) to 1.11 × 10 3 CFU m −3 (place C). The number of bacteria at site C was statistically greater (p = 0.05) than in the other investigated locations. The number of fungi ranged from 1.17 × 10 2 (F, internal control) to 4.73 × 10 2 CFU m −3 (C).
The number of xerophilic fungi was similar, from 1.17 × 10 2 to 4.47 × 10 2 CFU m −3 . A lower number was found in the case of mannitol-positive staphylococci (1.33 × 10 1 to 2.33 × 10 1 CFU m −3 ) and hemolytic bacteria (6.67 × 10 0 to 1.40 × 10 2 CFU m −3 ). Actinomycetes were the least numerous in the tested air samples (33.3 × 10 0 and 4.67 × 10 1 CFU m −3 ). However, their statistically significant dominance (p = 0.05) at place C was observed compared to the other examined places. No presence of Enterobacteriaceae bacteria, which could be evidence of faecal contamination of the air, or Pseudomonas fluorescens bacteria were recorded at the examined locations. Regardless of the type of microorganisms, the greatest number were recorded at site C (pile of MSW on the hill on the west side) and the lowest in the samples from external control (site G) ( Figure 5, Table S1). In the air in the tested landfill, the total number of bacteria ranged from 9.33 × 10 1 (G, external control) to 1.11 × 10 3 CFU m −3 (place C). The number of bacteria at site C was statistically greater (p = 0.05) than in the other investigated locations. The number of fungi ranged from 1.17 × 10 2 (F, internal control) to 4.73 × 10 2 CFU m −3 (C).
The number of xerophilic fungi was similar, from 1.17 × 10 2 to 4.47 × 10 2 CFU m −3 . A lower number was found in the case of mannitol-positive staphylococci (1.33 × 10 1 to 2.33 × 10 1 CFU m −3 ) and hemolytic bacteria (6.67 × 10 0 to 1.40 × 10 2 CFU m −3 ). Actinomycetes were the least numerous in the tested air samples (33.3 × 10 0 and 4.67 × 10 1 CFU m −3 ). However, their statistically significant dominance (p = 0.05) at place C was observed compared to the other examined places. No presence of Enterobacteriaceae bacteria, which could be evidence of faecal contamination of the air, or Pseudomonas fluorescens bacteria were recorded at the examined locations. Regardless of the type of microorganisms, the greatest number were recorded at site C (pile of MSW on the hill on the west side) and the lowest in the samples from external control (site G) ( Figure 5, Table S1).
The assessment of the microbiological contamination of the leachates from the dumping area is presented in Figure 5 and Table S2 (Supplementary Materials). The assessment of the microbiological contamination of the leachates from the dumping area is presented in Figure 5 and Table S2 (Supplementary Materials).
Many studies indicate that landfills are a significant epidemiological threat due to the possibility of the presence and multiplication of pathogenic bacteria in MSW. This is especially true of food waste, food packaging, used hygiene materials, as well as pet droppings. It has been shown that the number of microorganisms in the MSW can reach 10 8 -10 9 CFU g −1 . Therefore, they are a potential source of air, leachate and soil pollution in landfills.
It should be noted that there are currently no legal regulations in the world that would specify the limits of individual groups of microorganisms in the atmospheric air, leachate and soil in a landfill. Therefore, assessing this type of contamination is a difficult task and should be performed, taking into account the specifics of each place. There is little data in the literature on microbial contamination in illegal landfills. Thus far, such analyses have been performed for many MSW management facilities, such as sewage treatment plants [23,24], composting plants and other biomass processing facilities [25,26], waste sorting plants [27,28] and legal landfills [29,30]. Psychrophilic bacteria were detected in the largest number in these samples-from 8.53 × 10 5 to 1.28 × 10 6 CFU g −1 and mesophiles-from 5.49 × 10 5 to 1.74 × 10 6 CFU g −1 . A high number of hemolytic bacteria (3.0 × 10 4 -9.7 × 10 5 CFU g −1 ), mannitol-positive staphylococci (3.47 × 10 5-1.68 × 10 6 CFU g −1 ) and anaerobic bacteria (6.03 × 10 4 -4.59 × 10 6 CFU g −1 ) were also observed. Moreover, the presence of Actinomycetes ranging from 3.0 × 10 4 to 5.67 × 10 4 CFU g −1 was found. Enterobacteriaceae bacteria were also detected at the level of 1.47 × 10 3 -7.32 × 10 4 CFU g −1 ), and Pseudomonas fluorescens bacteria in 3 out of 5 examined samples (3.3 × 10 1 -2.67 × 10 2 CFU g −1 ). In the soil samples, a low number of fungi was detected, ranging from 1.67 × 10 2 to 1.87 × 10 2 CFU g −1 . The tested soil samples were slightly differentiated in terms of the number of microorganisms. No statistically significant differences (p = 0.05) were found in the number of fungi, Actinomycetes and sulphate-reducing anaerobes between individual samples.
Many studies indicate that landfills are a significant epidemiological threat due to the possibility of the presence and multiplication of pathogenic bacteria in MSW. This is especially true of food waste, food packaging, used hygiene materials, as well as pet droppings. It has been shown that the number of microorganisms in the MSW can reach 10 8 -10 9 CFU g −1 . Therefore, they are a potential source of air, leachate and soil pollution in landfills.
It should be noted that there are currently no legal regulations in the world that would specify the limits of individual groups of microorganisms in the atmospheric air, leachate and soil in a landfill. Therefore, assessing this type of contamination is a difficult task and should be performed, taking into account the specifics of each place. There is little data in the literature on microbial contamination in illegal landfills. Thus far, such Nair published a summary of microbial air pollution in landfills worldwide (Asia, Africa, Europe, North America, South America) [29]. In Poland, from municipal landfills areas (Sosnowiec and Janiki), the concentration of bacteria ranged from 7.0 × 10 1 to 4.0 × 10 4 CFU m −3 and fungal aerosol was from 2.0 × 10 2 to 1.2 × 10 4 CFUm −3 [31]. On the other hand, Breza-Boruta tested aerosol on the Municipal Waste Disposal Complex premises at Zolwin-Wypaleniska and Municipal Waste Treatment Plant in Inowrocław in Poland [4,32]. The author showed a large variation in the number of microorganisms in the air depending on the place of the study and the month in which the samples were taken. The number of bacteria in these studies was: 7.8 × 10 1 -5.4 × 10 5 CFU m −3 , Pseudomonas fluorescens: 0 to 4.3 × 10 2 CFU m −3 ; mannitol-positive staphylococci: 0 to 4.9 × 10 3 CFU m −3 ; Actinomycetes: 4 × 10 0 to 2.3 × 10 4 CFU m −3 , Enterobacteriaceae 0 to 2.8 × 10 2 CFU m −3 and fungi 0 to 2.2 × 10 4 CFU m −3 [4,32]. The results obtained in the present study are in line with the previously published ranges of the number of microorganisms isolated from legal landfills. The fact that the number of bacteria in the air around the MSW was statistically significantly higher than in the control air indicates that the tested landfill is the source of the emission of these microorganisms to the atmosphere.
There is also little data in the literature on the number of microorganisms in leachates and soils at illegal MSW disposal sites. Most of the work is focused on chemical pollutants, including heavy metals [6]. Lemanowicz et al. showed that in the soil (0-20 cm) at such places, there is 3.6 × 10 6 -8.6 × 10 6 CFU g −1 of bacteria, 6.7 × 10 4 -2.4× 10 6 CFU g −1 of Actinomycetes and 8.9 × 10 4 -1.3 × 10 5 CFU g −1 of fungi [14]. In turn, Hrenovic et al. studied total heterotrophic bacteria from 8 illegal dump-sites in Croatia [33]. The results obtained in the present study show a lower number of fungi than previously described in the literature [14]. This may be related to the specifics of the areas from which the samples were collected-Lemanowicz et al. studied landfills located in the forest, where the soil contains a lot of plant organic matter that favours the multiplication of fungi [14].
Mor et al. emphasised a high number of total coliform and faecal coliform in leachate and groundwater at a landfill site in India [34]. Likewise, Aderemi et al. analysed different locations adjacent to a MSW landfill [35]. They found that the total count of the Enterobacteriaceae in the leachate was 1.3 × 10 5 CFU mL −1 while the total viable count of Enterobacteriaceae in the groundwater samples ranged from 4.0 × 10 3 -1.0 × 10 6 CFU mL −1 . Further, in the present study, bacteria from the Enterobacteriaceae family were detected in the leachate but in a lower number at the level of 10 1 -10 3 CFU mL −1 . The presence of lower faecal contamination in illegal landfills compared to municipal facilities seems logical.
The presence of bacteria belonging to a total of 78 orders was recorded in the air in the landfill area; greater diversity of bacteria was found in the leachate-121 orders, and the highest in the soil-204 orders (Table S4).
The analysis of the biodiversity of bacteria based on the sequence analysis of the V3/V4 hypervariable region of the gene encoding the 16S rRNA showed that the samples collected at the landfill were dominated by the following genera of bacteria: Pseudomonas (from 1.3 in the leachate to 4.4% in the air) and an unknown genera from family Sinobacteraceae (from 2.6% in the air to 4.9% in the soil). In addition, the following genera occurred in the air with a large share: Escherichia (39.6%), Flavobacterium (6.3%), Bacillus (4.4%), Grevillea (3.6%), Stenotrophomonas (2.5%), unknown genera from of the order Streptophyta (18.4%), Burkholderiales (2%) and unknown genera from the family Micrococcaceae (2.9%).
The following bacteria had a significant share in the leachate: Aequorivita (4.8%), Perlucidibaca In the present study, we used high-throughput sequencing on Illumina MiSeq for assessing the diversity of microorganisms in the air, soil and leachate samples from illegal dumping sites.
The analysis of these samples allowed for a comprehensive study of the diversity of microorganisms in the dumpsite environment. Until now, only metagenomic analysis of soil and leachate from MSW management facilities has been published. It is worth emphasising that the conducted metagenomic analysis with high-throughput sequencing on Illumina MiSeq in the present study was performed for the first time for the air from illegal landfills.
This analysis showed a dominance of Proteobacteria, Actinobacteria, Bacteroidetes and Firmicutes in all types of tested samples. Additionally, in the air with high abundance was Cyanobacteria, and in the soil was Chloroflexi, TM7, TM6, and Verrucomicrobia.
Firmicutes, Proteobacteria, Bacteroidetes and Actinobacteria were the dominant phyla in landfills identified in previous studies in samples of cover soil, stored MSW, leachate and sediments [36][37][38].
Bacteria belonging to the Firmicutes phylum have cellulose-degrading abilities. These bacteria can play an important role in the anaerobic and methanogenic waste decomposition process. Simultaneously, Firmicutes have a leading share in the synthesis of fulvic-like substances and their humification [37,39]. Similarly, the main function of Bacteroidetes is a decomposition of starch polysaccharides and hydrolytic cellulose, while Proteobacteria are responsible for decomposing soluble sugars into monosaccharides and short-chain fatty acids [37,39]. Liu et al. indicate that Proteobacteria can promote aromatic compounds and alkoxy chain hydrocarbons with long branch chains and fever branches at both intermediate and old landfill stages [40].
Actinobacteria can play an essential role in the rapidly degrading simple lipids, proteins and carbohydrates [37].
Phylum TM7 (known as Saccharibacteria) is ubiquitous and has been detected in soils, sediments and wastewater sludge. It also can be associated with human diseases, including periodontitis and inflammatory bowel disease [42]. There were suspicions that TM7 was involved in the degradation of toluene and benzene, but Figueroa-Gonzalez et al. ruled it out [43].
Microbial taxonomic analysis at the genus level showed the highest abundance of genera Pseudomonas in all types of tested samples (air, soil, leachate). Pseudomonas sp. is widely detected in various environments, including landfills [31,39]. These bacteria utilise a wide range of polycyclic aromatic hydrocarbons and have roles in degrading organic matter and denitrification [44][45][46].
Bacillus sp. and Escherichia sp. were detected in high abundance in the air and soil. They are common bacteria present in landfill environments. Bacillus is known for cellulose degradation, polycyclic aromatic hydrocarbons and chromium oxidation [47,48]. Threedeach et al. pointed out that up to 87.5% of the Escherichia coli isolated from landfill environments can be resistant to antibiotics, especially doxycycline, cephalothin, tetracycline and minocycline [49].
However, also found with a large abundance were Alcanivorax, Candidatus Xiphinematobacter, Cellulosimicrobium and Tenacibaculum, which had not been previously reported for the test environment.
Alcanivorax is an isolated aquatic bacterium that plays an important role in the natural bioremediation of oil spills worldwide [52,53]. Likewise, bacteria of the genus Tenacibaculum constitute one of the dominant heterotrophic bacterial groups in aquatic habitats. Tenacibaculum can cause bacterial fish diseases such as enacibaculosis or flexibacteriosis [54].
Cellulosimicrobium is widely distributed in the environment and has been isolated mainly from the soil, water and grass cuttings. These bacteria have bioremediation capability for iron, zinc, copper, nickel and cadmium. Further, they have been reported for utilising benzo(a)pyrene, the sole carbon and energy source reducing toxic Cr(VI) to non-toxic Cr(III) and the remediation of thorium (IV) [56].
Notably, the literature describes the ability of the bacteria Pseudomonas and Tenacibaculum, which dominated in the landfill, to degrade polyester [58]. Moreover, Alcanivorax can form thick biofilms on low-density polyethene (LDPE) specifically and degrade petroleum-based plastic [59]. Additionally, Parvibaculum can degrade alkanes and alkynes. The domination of these bacteria in the landfill indicates that the microbiota has adapted to the environment where plastic MSW was deposited.
It is worth emphasising that of the identified bacteria, there may be potential human pathogens of the genera Bacillus, Clostridium, Cellulosimicrobium, Escherichia and Pseudomonas [60,61].
The percentage share of identified fungi at the Pylum level based on the analysis of the ITS region is summarised in Figure 8. Table S5 (Supplementary Materials) presents the correctly identified bacteria at the species level, along with their complete taxonomy and the percentage share in air, leachate and soil samples collected from the dumpsite.
In the air in the landfill, the presence of fungi belonging to 62 orders was noted, the lower diversity of fungi was demonstrated for the soil of 54 orders, and the lowest for the leachate-22 orders (Table S5). The analysis of the biodiversity of fungi based on the sequence analysis of the ITS region showed that the samples collected at the landfill were dominated by species belonging to the genera: Malassezia (from 1.0% in the soil to 8.5% in the air) and Bullera (from 12.9% in the air to 15.0% in leachate). In addition, there were types in the air with a high relative abundance: Blumeria (12.3%), Didymella (6.9%), Torula (3.9%), Alternaria (3.8%), Mycosphaerella (3.4%), Botrytis (2.5%), Peziza (1.8%), Aspergillus (1.6%), Cladosporium (1.4%), Strobilurus (1.2%) and Lycoperdon (1.1%).
Fungi of the genus Candida (1.2%) had a significant share in the leachate. In turn, the soil was dominated by fungi of the genera: Coprinellus (6.2%), Fusarium (5.4%), Microascus (4.6%), Mortierella (3.1%), Stachybotrys (2.0%), Chrysosporium (1.6%) and Aspergillus (1.4%). Unknown species from the order Agaricales (1.2%) and an unknown species from Phylum Ascomycota (9.6%) were also detected with a high percentage share. Other identified species in the tested samples accounted for <1%. Little is known about the diversity of fungi in landfills. The present work has detected the dominance of Basidiomycota, Ascomycota and Mortierellomycota.
In the air in the landfill, the presence of fungi belonging to 62 orders was noted, the lower diversity of fungi was demonstrated for the soil of 54 orders, and the lowest for the leachate-22 orders (Table S5). The analysis of the biodiversity of fungi based on the sequence analysis of the ITS region showed that the samples collected at the landfill were dominated by species belonging to the genera: Malassezia (from 1.0% in the soil to 8.5% in the air) and Bullera (from 12.9% in the air to 15.0% in leachate). In addition, there were types in the air with a high relative abundance: Blumeria (12.3%), Didymella (6.9%), Torula (3.9%), Alternaria (3.8%), Mycosphaerella (3.4%), Botrytis (2.5%), Peziza (1.8%), Aspergillus (1.6%), Cladosporium (1.4%), Strobilurus (1.2%) and Lycoperdon (1.1%).
Little is known about the diversity of fungi in landfills. The present work has detected the dominance of Basidiomycota, Ascomycota and Mortierellomycota.
Some of the identified species have previously been described as typical for environmental landfills: Aspergillus, Alternaria, Candida, Cladosporium, Chrysosporium, Didymella, Fusarium and Stachybotrys [29].
Coprinellus, Lycoperdon, Peziza and Strobilurus are saprobic cup fungi found throughout mainland Europe. Coprinellus is a wood-rotting fungus that grows on tree stumps and dead roots [67]. Lycoperdon occurs in all kinds of woodland, where they grow on the ground in leaf litter; also, less commonly, in permanent pasture. Peziza and Strobilurus occur mainly in the forest, often among gravel or shale [66]. The presence of these fungi is most likely related to the surroundings of the studied landfill and not the MSW.
Microascus (Scopulariopsis teleomorphs) are commonly isolated from soil, air, plant debris, paper, dung and moist indoor environments. These fungi can be opportunistic pathogens, primarily causing superficial tissue infections. Further, they are one of the principal causes of nondermatophytic onychomycoses [68].
Moreover, some other identified fungi are potentially hazardous to human health. Chrysosporium are common soil saprophytes; however, they may cause skin infections, and onychomycosis in humans has occasionally been isolated from systemic infections [69]. Likewise, few Candida species are human opportunistic pathogens and cause superficial or invasive infections in the case of weakened or reduced human immunity [70].
Malassezia fungi are associated with various conditions, including dandruff, atopic eczema dermatitis, pityriasis versicolor, seborrheic dermatitis and folliculitis. Malassezia can also cause systemic infections in immunocompromised patients [71].

Cytotoxicity of Soil Extracts and Waste Leachate Samples Assessment
The justification for selecting such cell lines for current research was as follows. The A-549 lung cell line was selected for its frequent use in assessing the cytotoxicity of dust and air pollutants [74]. The intestinal Caco-2 cell line because part of the inhaled dust can be swallowed with air, reach the gastrointestinal tract (GIT) and affect the intestinal epithelium, resulting in colitis [75]. The normal intestinal IEC-6 line was selected for the same reason. The Hep-G2 liver cell line was also applied, as different toxicants, odours and ingested dust can cause oxidative stress in the liver and lead to inflammation and DNA damage [76].
First, the cytotoxicity of the collected samples was screened against A-549 and Caco-2 cells (Figure 9). Leachate samples 2 and 3 demonstrated the strongest cytotoxicity against both A-549 and Caco-2 cells, while A-549 cells showed greater sensitivity towards the tested samples than Caco-2 cells. For sample 2, the IC 50 values were 159 (A-549) and 426 (Caco-2) mg/mL, while for sample 3 it was 182 (A-549) and 208 (Caco-2) mg/mL, respectively (Table 3).  After screening the cytotoxicity on A-549 and Caco-2 cells, the most cytotoxic samples were selected for further testing on Hep-G2 and then also for IEC-6 cells. These were MSW leachate samples 2, 3 and 5, and soil extract 5, respectively. Dose-dependent curves for these findings are not shown; only the IC50 values in Table 3 are presented. IEC-6 cells were more sensitive to the MSW leachate samples than Hep-G2 cells, while the opposite was for soil extract 5 in the case of IEC-6 cells, IC50 value could not be determined (Table  3). After analysing the results presented in Table 3, it can be concluded that leachate samples 2 and 3 were the most cytotoxic to cancerous Caco-2 cells, then Hep-G2, then A-549, while normal IEC-6 intestinal cells exhibited up to nine times greater sensitivity.   Both cell lines (i.e., A-549 and Caco-2) were less sensitive to the soil extracts than to the leachate samples. Caco-2 cells displayed the highest resistance to soil extracts to the extent that no IC 50 concentration could be determined for any ( Figure 10). For A-549 cells, IC 50 values were determined for samples 6, 9 and 10 (Table 3). However, only the stock extracts (1000 mg/mL) showed cytotoxicity up to a maximum approx. 78% ± 3.83%.  After screening the cytotoxicity on A-549 and Caco-2 cells, the most cytotoxic samples were selected for further testing on Hep-G2 and then also for IEC-6 cells. These were MSW leachate samples 2, 3 and 5, and soil extract 5, respectively. Dose-dependent curves for these findings are not shown; only the IC50 values in Table 3 are presented. IEC-6 cells were more sensitive to the MSW leachate samples than Hep-G2 cells, while the opposite was for soil extract 5 in the case of IEC-6 cells, IC50 value could not be determined (Table  3). After analysing the results presented in Table 3, it can be concluded that leachate samples 2 and 3 were the most cytotoxic to cancerous Caco-2 cells, then Hep-G2, then A-549, while normal IEC-6 intestinal cells exhibited up to nine times greater sensitivity.  After screening the cytotoxicity on A-549 and Caco-2 cells, the most cytotoxic samples were selected for further testing on Hep-G2 and then also for IEC-6 cells. These were MSW leachate samples 2, 3 and 5, and soil extract 5, respectively. Dose-dependent curves for these findings are not shown; only the IC 50 values in Table 3 are presented. IEC-6 cells were more sensitive to the MSW leachate samples than Hep-G2 cells, while the opposite was for soil extract 5 in the case of IEC-6 cells, IC 50 value could not be determined (Table 3). After analysing the results presented in Table 3, it can be concluded that leachate samples 2 and 3 were the most cytotoxic to cancerous Caco-2 cells, then Hep-G2, then A-549, while normal IEC-6 intestinal cells exhibited up to nine times greater sensitivity.

Genotoxicity of Seepage Leachate and Soil Samples
As a result of the cytotoxicity screening of leachate samples and soil extracts from the landfill, only those samples for which IC 50 values could be determined were selected for genotoxicity testing. Because the tested samples were weakly cytotoxic to Caco-2 cells, the genotoxicity analysis was performed only on A-549 cells. Concentrations close to and below the IC 50 values were selected. The results are summarised in Tables 4 and 5. The genotoxicity of the negative control (cells not exposed) was 5.4 ± 0.5%, while that of the positive control (cells exposed to 25 µM H 2 O 2 ) was 27.8 ± 3.6%. In general, the tested leachate samples were characterised as no genotoxic (concentration 125 mg/mL) or medium or weakly genotoxic (concentration 250 mg/mL) ( Table 4). Soil extracts showed even lower genotoxicity. The highest genotoxicity was obtained for sample 4, with a concentration of 1000 mg/mL (14.5 ± 3.5%) ( Table 5).

Targeted and Untargeted UPLC-HRMS Analysis
The results obtained in the process of the identification of toxic compounds in leachate and soil from illegal landfills are summarised in Table 6. The targeted identification of pesticides and other contaminants commonly found in environmental water was performed using the TargetScreener (Bruker Daltonics) method that is based on tight tolerances matching analyte retention time, precursor m/z, fragments m/z, precursor and fragments isotopic pattern and precursor-to-fragment signals intensity ratios. Detailed data on identifications are provided in Table S6 (Supplementary Materials).

Statistical Analysis of LC-MS Data
The comparative analysis of MS features further investigated the chemical composition of leachate samples and soil type. The much higher cytotoxicity of leachate and soil samples marked with numbers 2 and 3 compared to samples numbers 1, 4 and 5 was the reason behind the comparison of the mentioned two groups performed in MetaboScape 2021 and then in MetaboAnalyst 5.0. Heatmaps were created to visualise the differences in the MS features between the soil and leachate samples from different sampling sites (Figure 11). Heatmaps were created for the top 1000 m/z values selected based on their interquartile range (IQR) that differed significantly between the two groups of samples. Heatmaps clearly show that there is a high similarity between samples 2 and 3 (green) and 1, 4 and 5 (red) in both cases. In the case of leachates, samples 2 and 3 seem to be similar to sample number 1, which came from the same area of an illegal landfill.
Int. J. Environ. Res. Public Health 2022, 19, x FOR PEER REVIEW 22 of 32 ( Figure 11). Heatmaps were created for the top 1000 m/z values selected based on their interquartile range (IQR) that differed significantly between the two groups of samples. Heatmaps clearly show that there is a high similarity between samples 2 and 3 (green) and 1, 4 and 5 (red) in both cases. In the case of leachates, samples 2 and 3 seem to be similar to sample number 1, which came from the same area of an illegal landfill. Furthermore, much more MS features were identified in these groups than in samples 4 and 5. A similar trend can be observed in the soil samples. Comparisons of MS features in soil samples indicate that soil 5 appears associated first with both soil 1 and 4 and then with soil sample numbers 2 and 3. Univariate and multivariate statistical analyses of MS features were employed to examine the relationships between two groups of the soil and leachate samples from different areas of illegal landfills. MS features following normalisation by sum, log-transformation and auto-scaling were analysed with principal component analysis (PCA) to assess whether samples 2 and 3 versus samples 1, 4 and 5 could be separated based on distinct MS profiles. The resulting 2D and 3D PCA scores plots ( Figure 12A,B) showed that the MS features profiles of leachate samples 2 and 3 are, to a large extent, dissimilar from leachate samples 1, 4, 5, with PC1 and PC2 accounting for 52.9% and 19.7% of the variance, respectively.
Group separations were also inspected using OPLS-DA, which revealed a clear separation between these two groups of leachate samples ( Figure 12C). In order to evaluate the statistical robustness of this model, 2000 permutation tests were conducted. Decent discrimination was observed between the two groups of the samples (Q 2 = 0.987, R 2 Y = 0.998, p-value < 0.05 (0/2000)), revealing substantial differences in MS features profiles of leachates 2 and 3 and samples 1, 4 and 5 ( Figure S1, supplementary material).
An OPLS-DA statistical method referred to as S-plot was employed to identify MS features present in significantly different levels in both groups of samples. Higher values Furthermore, much more MS features were identified in these groups than in samples 4 and 5. A similar trend can be observed in the soil samples. Comparisons of MS features in soil samples indicate that soil 5 appears associated first with both soil 1 and 4 and then with soil sample numbers 2 and 3.
Univariate and multivariate statistical analyses of MS features were employed to examine the relationships between two groups of the soil and leachate samples from different areas of illegal landfills. MS features following normalisation by sum, log-transformation and auto-scaling were analysed with principal component analysis (PCA) to assess whether samples 2 and 3 versus samples 1, 4 and 5 could be separated based on distinct MS profiles. The resulting 2D and 3D PCA scores plots ( Figure 12A,B) showed that the MS features profiles of leachate samples 2 and 3 are, to a large extent, dissimilar from leachate samples 1, 4, 5, with PC1 and PC2 accounting for 52.9% and 19.7% of the variance, respectively.
we narrowed down this group to 1364 MS features with a fold change with FC ≥ 2 or ≤ 0.5 (|log2 FC| ≥ 1). In order to confirm the correlativity of each selected MS feature, we performed variable influence on projection (VIP) for the OPLS model. All selected m/z values have a VIP value above 1.0. The first 15 MS features with the highest VIP values are shown in Figure 12D. Out of the 1364 m/z values, which differ to the greatest extent between the two groups of leachate samples, compound name annotations were made for 784 MS features using MS and MS/MS data (  Chemical compounds abundances obtained from LC-MS experiments on soil samples were also subjected to statistical data analysis A 2D and 3D PCA scores plot to reveal good discrimination between soil samples with higher and lower cytotoxicity ( Figure  13A,B). Comparing soil samples 2 and 3 versus 1, 4 and 5 with OPLS-DA revealed excellent discrimination between these two groups ( Figure 12C). The quality factors for this model were: Q 2 : 0.982 and R 2 Y: 0.992, with p-values based on permutation tests lower than 0.05 ( Figure S2, Supplementary Materials). The analysis of S-plots revealed that 1014 MS features were positively correlated with group separation with a p(corr) score >0.5, and 397 features values were negatively correlated with a p(corr) score < −0.5. Fold change analysis indicated 1525 MS features with a fold change with FC ≥ 2 or ≤ 0.5 (|log2 FC| ≥ Group separations were also inspected using OPLS-DA, which revealed a clear separation between these two groups of leachate samples ( Figure 12C). In order to evaluate the statistical robustness of this model, 2000 permutation tests were conducted. Decent discrimination was observed between the two groups of the samples (Q 2 = 0.987, R 2 Y = 0.998, p-value < 0.05 (0/2000)), revealing substantial differences in MS features profiles of leachates 2 and 3 and samples 1, 4 and 5 ( Figure S1, Supplementary Material).
An OPLS-DA statistical method referred to as S-plot was employed to identify MS features present in significantly different levels in both groups of samples. Higher values of |p(corr)| indicate more important values to the classification. Variables with a |p(corr)| value greater than 0.5 were considered significant. A total of 406 variables were negatively correlated to group separation, showing a p(corr) score of <−0.5 and 1360 variables were positively correlated to group separation showing a p(corr) score of >0.5. A paired t-test was performed for all variables. An analysis of the S-score of the OPLS-DA model, combined with independent statistical t-test analysis (p values < 0.05), indicated that these 1766 MS features were responsible for the differences. Based on the fold change analysis (FC), we narrowed down this group to 1364 MS features with a fold change with FC ≥ 2 or ≤ 0.5 (|log 2 FC| ≥ 1). In order to confirm the correlativity of each selected MS feature, we performed variable influence on projection (VIP) for the OPLS model. All selected m/z values have a VIP value above 1.0. The first 15 MS features with the highest VIP values are shown in Figure 12D. Out of the 1364 m/z values, which differ to the greatest extent between the two groups of leachate samples, compound name annotations were made for 784 MS features using MS and MS/MS data (Table S7, Supplementary Materials).
Chemical compounds abundances obtained from LC-MS experiments on soil samples were also subjected to statistical data analysis A 2D and 3D PCA scores plot to reveal good discrimination between soil samples with higher and lower cytotoxicity ( Figure 13A,B). Comparing soil samples 2 and 3 versus 1, 4 and 5 with OPLS-DA revealed excellent discrimination between these two groups ( Figure 12C). The quality factors for this model were: Q 2 : 0.982 and R 2 Y: 0.992, with p-values based on permutation tests lower than 0.05 ( Figure S2, Supplementary Materials). The analysis of S-plots revealed that 1014 MS features were positively correlated with group separation with a p(corr) score >0.5, and 397 features values were negatively correlated with a p(corr) score < −0.5. Fold change analysis indicated 1525 MS features with a fold change with FC ≥ 2 or ≤ 0.5 (|log 2 FC| ≥ 1). The VIP plot for the OPLS model indicated that all selected m/z values have a VIP value above 1.0. The first 15 MS features with the highest VIP values are shown in Figure 13D. Furthermore, all data sets were further subjected to independent t-test analysis, which indicated that all selected MS features were statistically significant with p-values below 0.05. Out of the 1525 features that differ to the greatest extent between the two groups of soil samples, annotations were made for 611 MS features using MS and MS/MS data.  Figure 13D. Furthermore, all data sets were further subjected to independent t-test analysis, which indicated that all selected MS features were statistically significant with p-values below 0.05. Out of the 1525 features that differ to the greatest extent between the two groups of soil samples, annotations were made for 611 MS features using MS and MS/MS data.

Pathway Analysis
The most differentiating compounds identified in soil and leachate samples number 2 and 3 vs. 1, 4 and 5 were subjected to pathway analysis in MetaboAnalyst 5.0. In this study, we selected a library of twelve organisms derived from KEGG metabolic pathways, including all accessible microorganisms-yeast: Saccharomyces cerevisiae; bacteria: Escherichia coli, Bacillus subtilis, Pseudomonas putida, Staphylococcus aureus, Thermotoga maritima, Mesorhizobium japonicum, Klebsiella pneumoniae, Klebsiella variicola and Streptococcus pyogenes; cyanobacterium Synechococcus elongatus; and, additionally, parasite: Schistosoma mansoni; for pathway analysis.
For the KEGG pathway enrichment analysis, 471 compounds were assigned to

Pathway Analysis
The most differentiating compounds identified in soil and leachate samples number 2 and 3 vs. 1, 4 and 5 were subjected to pathway analysis in MetaboAnalyst 5.0. In this study, we selected a library of twelve organisms derived from KEGG metabolic pathways, including all accessible microorganisms-yeast: Saccharomyces cerevisiae; bacteria: Escherichia coli, Bacillus subtilis, Pseudomonas putida, Staphylococcus aureus, Thermotoga maritima, Mesorhizobium japonicum, Klebsiella pneumoniae, Klebsiella variicola and Streptococcus pyogenes; cyanobacterium Synechococcus elongatus; and, additionally, parasite: Schistosoma mansoni; for pathway analysis.
For the KEGG pathway enrichment analysis, 471 compounds were assigned to twentysix metabolic pathways, including butanoate metabolism, α-linolenic acid metabolism, styrene degradation, tryptophan metabolism, phenylalanine metabolism, atrazine degradation, sesquiterpenoid and triterpenoid biosynthesis, glyoxylate and dicarboxylate metabolism, pyruvate metabolism and aminobenzoate degradation. The most significant results for the selected microorganisms are shown in Figure 14. In the second step, we performed pathway enrichment analysis of leachate samples based on the main class compounds analysis. The most significantly enriched main class turned out to be amino acids and peptides ( Figure 15). Moreover, phenols, indoles and benzenes were also greatly differentiating classes. In the second step, we performed pathway enrichment analysis of leachate samples based on the main class compounds analysis. The most significantly enriched main class turned out to be amino acids and peptides ( Figure 15). Moreover, phenols, indoles and benzenes were also greatly differentiating classes.
In order to identify the pathways that are the most relevant in soil samples, all 614 annotated features were analysed. However, targeted pathway analysis was performed for 413 HMDB or KEGG ID compounds. Metabolic pathways with values > 0.1 were considered to be significant. Twenty-six metabolic pathways, including one carbon pool by folate, phenylalanine metabolism, sphingolipid metabolism, atrazine degradation, sesquiterpenoid and triterpenoid biosynthesis, styrene degradation and aminobenzoate degradation with the highest impact values were significantly related to the differentiation between the above-discussed soil samples. The best results for the selected microorganisms are shown in Figure 14.
Pathway enrichment analysis of soil samples was also performed, focusing on main class compounds. The most significantly enriched main classes were benzene, amino acids and peptides (Figure 16). A significant number of steroids and benzoic acid were also considered as significantly differentiating the two soil sample groups. In the second step, we performed pathway enrichment analysis of leachate samples based on the main class compounds analysis. The most significantly enriched main class turned out to be amino acids and peptides ( Figure 15). Moreover, phenols, indoles and benzenes were also greatly differentiating classes. Pathway analysis showed that the compounds that differentiate the most soil and leachate samples with different cytotoxicity were associated with 13 common metabolic pathways, including aminobenzoate degradation, arginine and proline metabolism, atrazine degradation, biotin metabolism, glutathione metabolism, lysine degradation, phenylalanine metabolism, sesquiterpenoid and triterpenoid biosynthesis, sphingolipid metabolism, styrene degradation, sulphur metabolism, tryptophan metabolism and tyrosine metabolism ( Figure 17). The results of metabolic pathway enrichment analysis showed that 21 main class compounds, common for soil and leachates, including fatty acids and conjugates, benzamides, steroids, amino acids and peptides, benzenes, sphingoid bases, benzoic acids, indoles, pyridines, phenylpropanoids, alcohols and polyols, amines, aromatic heteropolycyclic compounds, cinnamic acids, phenylacetic acids, alkaloids, piperazines, phosphate esters, organic carbonic acids, delta valerolactones and piperidinones were mainly involved in the toxicity of soil and leachate samples ( Figure 17). Many of these compounds were classified by the International Agency for Research on Cancer as "carcinogenic to humans" (Group 1), "probably carcinogenic to humans" (Group 2A) or "possibly carcinogenic to humans" (Group 2B) [85].
It is worth recalling that isolates from extreme environments may show different metabolic activity than strains of the same species grown under standard conditions. Present databases are still too poor to connect directly detected metabolites with microorganisms in the studied environment. Metabolome data deposited in KEGH and HMDB included only ten bacteria and fungi metabolites, while almost 900 species were identified. The strains from the landfill where plastics were collected primarily will likely have a modified metabolism in terms of polymer degradation, which is confirmed by the results of the metabolomic analysis obtained in the present work. radation with the highest impact values were significantly related to the differentiation between the above-discussed soil samples. The best results for the selected microorganisms are shown in Figure 14.
Pathway enrichment analysis of soil samples was also performed, focusing on main class compounds. The most significantly enriched main classes were benzene, amino acids and peptides (Figure 16). A significant number of steroids and benzoic acid were also considered as significantly differentiating the two soil sample groups. Pathway analysis showed that the compounds that differentiate the most soil and leachate samples with different cytotoxicity were associated with 13 common metabolic pathways, including aminobenzoate degradation, arginine and proline metabolism, atrazine degradation, biotin metabolism, glutathione metabolism, lysine degradation, phenylalanine metabolism, sesquiterpenoid and triterpenoid biosynthesis, sphingolipid metabolism, styrene degradation, sulphur metabolism, tryptophan metabolism and tyrosine metabolism ( Figure 17). The results of metabolic pathway enrichment analysis showed that 21 main class compounds, common for soil and leachates, including fatty acids and conjugates, benzamides, steroids, amino acids and peptides, benzenes, sphingoid bases, benzoic acids, indoles, pyridines, phenylpropanoids, alcohols and polyols, amines, aromatic heteropolycyclic compounds, cinnamic acids, phenylacetic acids, alkaloids, piperazines, phosphate esters, organic carbonic acids, delta valerolactones and piperidinones were mainly involved in the toxicity of soil and leachate samples ( Figure 17). Many of these compounds were classified by the International Agency for Research on Cancer as "carcinogenic to humans" (Group 1), "probably carcinogenic to humans" (Group 2A) or "possibly carcinogenic to humans" (Group 2B) [85]. It is worth recalling that isolates from extreme environments may show different metabolic activity than strains of the same species grown under standard conditions. Present databases are still too poor to connect directly detected metabolites with microorganisms in the studied environment. Metabolome data deposited in KEGH and HMDB included only ten bacteria and fungi metabolites, while almost 900 species were identified.

Conclusions
The tested landfill is a source of physicochemical, biological and toxicological hazards for humans. Dominance (99.89% and 99.99%) shows particles with dimensions below 1 µm (PM 1 ) in total airborne dust at the tested locations. The PM 1 concentration is almost twice as high as the environmental threshold (0.025 mg m −3 ).
Quantitative analysis of the microbial contamination of air, leachate and soil fills the data gap on microbiological hazards caused by illegal landfills. Thus far, such analyses have been conducted most often for legally operating landfills or other facilities related to waste management (wastewater treatment plant, composting plant, waste sorting plant).
It was shown that the number of microorganisms in the air of the examined landfill is in line with the previously published ranges for legal landfills. The fact that the number of bacteria in the air around the MSW was statistically significantly higher than in the control air indicates that the tested landfill is the source of the emission of these microorganisms to the atmosphere.
Cytotoxic activity was detected for 2 of 5 water samples and 3 of 5 soil samples from the landfill sites. Leachate samples 2 and 3 were the most cytotoxic to cancerous Caco-2 cells, then Hep-G2, then A-549, while normal IEC-6 intestinal cells exhibited up to nine times greater sensitivity. Soils 1, 4, and 5 were cytotoxic to A-549, while to Hep-G2, only soil sample 5 was cytotoxic. The toxic activity of the tested samples may be induced by the presence of pesticide and drug residues as well as their metabolites and/or metabolites of microorganisms detected in the tested samples. The cytotoxicity results of the leachate and soil samples are worrying because below the sampling site is a lake where people fish. Compounds responsible for cytotoxicity may also adversely affect fish, accumulate in them, and go further to other elements of the trophic chain (animals, humans). However, tested leachates and soils were characterised as no genotoxic or weakly genotoxic, except for the most cytotoxic leachate no. 2, the genotoxicity of which could be defined as "medium".
Common pesticides (carbamazepine, prometryn, terbutryn, permethrin, carbanilide, pyrethrin, carbaryl and prallethrin), quaternary ammonium compounds (benzalkonium chlorides), chemicals and/or polymer degradation products (melamine, triphenylphosphate, diphenylphtalate, insect repellent diethyltoluamide), as well as a common drug (ketoprofen) were found in soil and leachate samples. Compounds from soil (471) and leachates (413) belonged to twenty-six metabolic pathways, including metabolisms of pesticides, herbicides and polymers. However, present databases are still too data poor for connecting directly detected metabolites with microorganisms present in the studied environment.
This paper shows that the dangers posed by illegal landfills should not be ignored. It is recommended to perform comprehensive analyses on this type of site, including detailed microbiological and toxicological analyses that would allow for a better understanding of the mechanisms and scale of the impact of illegal landfills on the environment. Moreover, currently, there are no clear legal regulations regarding the assessment of risks caused by illegal landfills in Poland. In most cases, experts decide on the scope of analysis, considering the ratio of their costs to the expected risk. The presented study may constitute the basis for the development of guidelines for conducting this type of research, considering the need to perform microbiological (including air) and toxicological analyses, which are often omitted. Future work should seek to develop a critical workflow for assessing the risks posed by illegal landfills in cooperation with state administration authorities.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijerph19084826/s1, Table S1: Microbiological contamination of the air in the tested landfill; Table S2: Microbiological contamination of the leachate in the tested landfill; Table S3: Microbiological contamination of the soil in the tested landfill; Table S4: Diversity of bacteria in landfill environment; Table S5: Diversity of fungi in landfill environment; Table S6: Identification parameters of toxic compounds found in leachate and soil samples; Table S7: Analysis of metabolites detected in leachate samples; Figure