Assessment of Common Cyanotoxins in Cyanobacteria of Biological Loess Crusts

Cyanotoxins are a diverse group of bioactive compounds produced by cyanobacteria that have adverse effects on human and animal health. While the phenomenon of cyanotoxin production in aquatic environments is well studied, research on cyanotoxins in terrestrial environments, where cyanobacteria abundantly occur in biocrusts, is still in its infancy. Here, we investigated the potential cyanotoxin production in cyanobacteria-dominated biological loess crusts (BLCs) from three different regions (China, Iran, and Serbia) and in cyanobacterial cultures isolated from the BLCs. The presence of cyanotoxins microcystins, cylindrospermopsin, saxitoxins, and β-N-methylamino-L-alanine was analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) method, while the presence of cyanotoxin-encoding genes (mcyE, cyrJ, sxtA, sxtG, sxtS, and anaC) was investigated by polymerase chain reaction (PCR) method. We could not detect any of the targeted cyanotoxins in the biocrusts or the cyanobacterial cultures, nor could we amplify any cyanotoxin-encoding genes in the cyanobacterial strains. The results are discussed in terms of the biological role of cyanotoxins, the application of cyanobacteria in land restoration programs, and the use of cyanotoxins as biosignatures of cyanobacterial populations in loess research. The article highlights the need to extend the field of research on cyanobacteria and cyanotoxin production to terrestrial environments.


Introduction
Cyanobacteria produce various bioactive metabolites, some of which exert toxic effects on human and animal cells and cause chronic and acute diseases. Cyanotoxins can be surfaces to establish soil particle stability, nutrient uptake, and circulation, and promote the establishment of higher vegetation [28][29][30]. It is, therefore, important that the cyanobacterial strains used for inoculation do not express toxicity to humans and animals, because particles of the crust containing potentially toxic cyanobacteria could become airborne and affect human and animal health by inhalation [1,8].
The toxicity of cyanobacterial taxa within BLCs has only recently been addressed for the first time. In the two studies, one BLC sample [31] and one cyanobacterial strain isolated from a BLC [27] showed low toxicity to the Artemia salina larvae, indicating the necessity for further studies. This study aims to investigate cyanotoxin production in cyanobacteriadominated BLCs and in cyanobacterial strains isolated from BLCs to address the following: 1. the extent of cyanotoxin production in cyanobacteria of BLC; 2. to avoid the application of potentially toxic strains in the restoration of deteriorated loess surfaces; 3. potential of cyanotoxins as a biosignature of cyanobacterial communities in paleoenvironmental research of loess.
Loess deposits, because of their porous structure and extensive exploitation and degradation, are a significant source of airborne dust affecting people living in loess-rich areas. With a similar idea to the BLOCDUST hypothesis on the interaction between cyanobacteria and loess particles, artificial formation of BLC cover over the degraded loess surfaces by inoculation of cyanobacteria is proposed as a promising mechanism for their restoration [26,27]. The technology is based on the inoculation of laboratory-grown cyanobacteria on disturbed soil surfaces to establish soil particle stability, nutrient uptake, and circulation, and promote the establishment of higher vegetation [28][29][30]. It is, therefore, important that the cyanobacterial strains used for inoculation do not express toxicity to humans and animals, because particles of the crust containing potentially toxic cyanobacteria could become airborne and affect human and animal health by inhalation [1,8].
The toxicity of cyanobacterial taxa within BLCs has only recently been addressed for the first time. In the two studies, one BLC sample [31] and one cyanobacterial strain isolated from a BLC [27] showed low toxicity to the Artemia salina larvae, indicating the necessity for further studies. This study aims to investigate cyanotoxin production in cyanobacteria-dominated BLCs and in cyanobacterial strains isolated from BLCs to address the following: 1. the extent of cyanotoxin production in cyanobacteria of BLC; 2. to avoid the application of potentially toxic strains in the restoration of deteriorated loess surfaces; 3. potential of cyanotoxins as a biosignature of cyanobacterial communities in paleoenvironmental research of loess.

Results
Cyanobacteria isolated from BLC samples consisted of either mixed-strain or monocyanobacterial cultures from the following genera: Leptolyngbya, Nostoc, cf. Mojavia, Trichormus, Chlorogloeopsis, Calothrix, Tolypothrix, Symploca, Scytonema, Chroococcus, Chroococcidiopsis, Desmonostoc, Symplocastrum, Oculatella, Pseudanabaena, and Pseudophormidium, Scytonematopsis. The underlined genera involve previously described cyanotoxin producing strains. The highest number of isolated strains was determined for the genera Nostoc and Leptolyngbya ( Figure 1). The distribution of genera within the cultures is shown in Supplementary Materials File S1. The distribution of identified genera. The most frequently isolated genera were Nostoc and Leptolyngbya, followed by Chlorogloeopsis, Scytonema, and Tolypothrix. Figure 1. The distribution of identified genera. The most frequently isolated genera were Nostoc and Leptolyngbya, followed by Chlorogloeopsis, Scytonema, and Tolypothrix.

Chromatographic Analyses
We could not detect any of the eight MCs, CYN, and BMAA in analyzed BLCs. Similarly, we could not detect any of the eight MCs, CYN, STX, and GTX2/3 in mixedstrain cyanobacterial cultures derived from BLCs. Chromatograms and fragmentation spectra/MRM chromatograms for all groups of reference material are provided in Supplementary Materials File S2.

PCR Analyses
PCR amplification of the 16S RNA genes was successful in 55 out of 60 cyanobacterial strains. We could not detect PCR amplification of the genes coding the synthesis of MCs (mcyE), NOD (nodF), CYN (cyrJ), STX (sxtA, sxtG, and sxtS), and ANTX (anaC) in any of the 60 cyanobacterial strains isolated from BLCs. Images showing amplification of targeted genes in reference material are presented in Figure 2 (the raw gel images are provided in Supplementary Material S3).

Chromatographic Analyses
We could not detect any of the eight MCs, CYN, and BMAA in analyzed B Similarly, we could not detect any of the eight MCs, CYN, STX, and GTX2/3 in m strain cyanobacterial cultures derived from BLCs. Chromatograms and fragment spectra/MRM chromatograms for all groups of reference material are provide Supplementary Materials File S2.

PCR Analyses
PCR amplification of the 16S RNA genes was successful in 55 out of 60 cyanobac strains. We could not detect PCR amplification of the genes coding the synthesis of (mcyE), NOD (nodF), CYN (cyrJ), STX (sxtA, sxtG, and sxtS), and ANTX (anaC) in a the 60 cyanobacterial strains isolated from BLCs. Images showing amplificatio targeted genes in reference material are presented in Figure 2 (the raw gel image provided in Supplementary Material S3).

Discussion
During this study, standard analytical and molecular biological methods developed for cyanobacterial samples from the aquatic environment were used. It is unlikely that the type of matrix (biomass and BLCs) affected the extraction and analysis of cyanotoxins and DNA, e.g., the extraction and instrumental analyses were based on widely-accepted protocols [32][33][34]. Interestingly, we could not detect any cyanotoxins, nor any cyanotoxin synthetase encoding genes in our study. This noteworthy contrast in toxin production between cyanobacteria inhabiting a specific terrestrial substrate, loess sediment, compared to aquatic environments is very intriguing and opens many questions about cyanotoxin production in cyanobacteria.
As there are cyanobacterial strains producing toxins and toxin-deficient strains, most cyanobacterial toxins have been regarded as secondary metabolites, i.e., compounds lacking essential functions in the growth of the organism under optimal conditions. Such compounds may still confer a certain selective advantage on the organism producing them and promote the long-term survival of the population. General information on the cause of toxin production in cyanobacteria and their functional biological/ecological role is still very limited and requires further research. However, several hypotheses have emerged over the years, supported by the results of various studies. Some possible biochemical and ecophysiological functions of cyanotoxins are shown in Table 1. Table 1. Possible biochemical and ecophysiological roles of cyanotoxins that have been reported in the literature.
Autoregulation and other biological phenomena have also been recognized as an ecological/survival strategy. There is increasing evidence of the release of microcystin as an essential part of the life cycle strategy of populations of Microcystis and other bloom-forming cyanobacteria [58,65,70,71]. Hu and Rzymski [70] examined apoptosis, i.e., programmed cell death, as an ecological/survival strategy of Microcystis and provided a conceptual model coupling apoptosis and MC release. Furthermore, the role of MCs and other non-ribosomal peptides in cell-to-cell communication during lytic events has been recognized. Schatz et al. [58] showed that lysis of Microcystis cells or exposure to the cyanobacterial metabolites MC, micropeptin, or microginin induced an enhanced production of MCs in the remaining Microcystis population. The authors interpreted this elevation of MC production and increased toxicity as an attempt to raise the fitness of the population in its ecological niche. MC involvement in the regulation of Microcystis morphotype (with influence on cell-to-cell contacts) was indicated by Zilliges et al. [65] who showed that the expression pattern of the extracellular glycoprotein MrpC was different in microcystincontaining and microcystin-deficient strains. Makower et al. [71] showed that the relative expression of genes related to the central intermediary metabolism, photosynthesis, and energy metabolism were different in a wild-type toxin-producing strain of Microcystis aeruginosa and a microcystin-deficient mutant. Interestingly, Makower et al. [71] showed that the signaling effect of extracellular, added microcystin was limited to the regulation of genes related to secondary metabolism only.
Successful cell-to-cell and especially colony-to-colony communication require a medium for the diffusion of autoinducers and other signaling molecules. It is evident that such communication cannot be equally successful or meaningful in areas with strongly limited water availability. Considering communication as an important biological function of MCs, it would be a waste of resources and an adaptive disadvantage for the cell to produce MCs in an environment where they cannot be utilized.
Could water availability actually explain the difference between our results and those of studies reporting cyanotoxins in biocrust cyanobacteria? Water availability varies widely in arid and semiarid regions. The BLCs and isolated cyanobacteria we studied originate predominantly from exposed vertical loess deposit profiles and hillslopes where water retention is very unlikely. However, some biocrusts develop in depressions where intense dew formation occurs or where rainwater can accumulate for some time. Assuming that sufficient water is available for a period of time, cyanobacteria would have a medium through which cyanotoxin-mediated signaling would be possible.
On the other hand, research on cyanotoxin production in biocrust cyanobacteria is insufficient, and we do not have a clear picture of the extent of cyanotoxin production in environments where biocrusts are present. It could simply be that the cyanobacterial strains studied here do not produce common cyanotoxins, as some cyanobacterial strains/species do in aquatic environments. As highlighted by Gärtner et al. [72] and Huang et al. [7], research on cyanotoxins in terrestrial environments should be more precise in terms of sampling site characteristics and conditions and species diversity if we are to fully understand the factors driving cyanotoxin production in terrestrial environments and the biochemical/ecophysiological role of cyanotoxins in general.
We assessed common cyanotoxins in strains of BLC cyanobacteria to evaluate their potential in restoration of degraded loess environments. Our results suggest that studied BLC cyanobacteria do not produce common cyanotoxins and imply their safe use in restoration. However, with recent developments in analytical techniques and in vitro and in vivo bioassays, bioactive compounds are being discovered that have toxic effects on human and animal cells but are structurally different from the known cyanotoxins, so we should not rule out the presence of some other toxic compounds [1,73]. For example, strains isolated from terrestrial environments have recently been reported to produce an uncommon cyanotoxin apratoxin, the production of which was previously associated exclusively with marine environments [7]. Bioassays could be a great tool to examine the general toxicity of inoculants [27,73], which is especially important for the rehabilitation of degraded and desertified land by cyanobacterial inoculation, since the application of toxic cyanobacterial strains would be hazardous.
One of the objectives of this study was to examine the biosignature capacity of cyanotoxins for use in paleoenvironmental studies of loess, in terms of genus/species specificity and abundance. Cyanotoxins are organism-specific compounds, and MCs and CYN have been reported to maintain stable structure under a variety of environmental condi-tions [35,[74][75][76], which is a prerequisite for sedimentary biosignature. Cyanotoxins MCs and CYN have been used as a paleolimnological tool to detect past harmful algal blooms (HABs), track population fluctuations of cyanotoxin producers, and understand under which environmental conditions cyanotoxin production was triggered [77][78][79][80][81][82]. Although there are some concerns about the biomarker potential of MCs regarding stability (e.g., resistance of MCs to bacterial degradation under anoxic conditions) and extraction methods, they are considered to have great potential for use in paleolimnological studies along with CYN [83].
The notable absence of cyanotoxin production by the BLC cyanobacteria studied here suggests that cyanotoxin production is not as common in BLC and therefore may not be reliable biosignature candidates to represent terrestrial cyanobacterial populations in loess research. Further research is needed to understand the extent of cyanotoxin production in BLCs and terrestrial environments in general, and to link a particular terrestrial environment (e.g., loess) or environmental conditions to cyanotoxin production.

Conclusions
We did not detect targeted common cyanotoxins or cyanotoxin-encoding genes in the studied BLCs and cyanobacterial strains. This is an indication of either lack of common cyanotoxin production in studied BLC cyanobacteria or production of novel cyanotoxins/secondary metabolites. The environment of biocrust cyanobacteria is completely different from that of aquatic cyanobacteria. Therefore, different metabolic profiles might be expected, as metabolites may not have the same adaptive value in different environments. To date, little research has been conducted on the production of cyanotoxins by terrestrial cyanobacteria, and we hope that this article will stimulate further studies on this topic. This is of great importance if we are to define the biochemical or ecophysiological roles of cyanotoxins.
Furthermore, our results suggest that analyzed cyanobacterial strains can potentially be used in land restoration. We suggest that cyanobacterial strains should be tested with e.g., bioassays prior to inoculation in the field to investigate their potential toxicity, as they may produce non-common toxic compounds or toxic compounds not yet described.
The observed absence of cyanotoxin production in studied BLC cyanobacteria does not favor the use of cyanotoxins as a biosignature of cyanobacterial communities in the reconstruction of paleoclimate and paleoenvironment in loess. Further studies are needed to link specific environments to cyanotoxin occurrence in terrestrial sediment records.
Prior to isolation of cyanobacterial strains, biocrust samples were visually examined using binocular stereomicroscope (Leica MZ) to confirm the presence of cyanobacterial colonies. To isolate cyanobacteria, approximately two cm 2 of a BLC was soaked in 50 mL of BG11-N and BG11-N 0 (nitrate-free) medium [84] and cultured under a 14/10 (light/dark) illumination period (8000 K, 60 lux, 750 lm; Power-GLO, Rolf C. Hagen Inc., Mansfield, MA, USA) at room temperature ( ∼ =22 • C) for 30 days. During this period, colonies of cyanobacteria formed over and around the biocrust sample. The colonies were separated from the biocrust surface and flask walls/bottom and transferred to 5 mL of fresh BG11-N/N 0 . After 30 days of cultivation under the conditions mentioned above, 40 mL of fresh BG11-N/N 0 medium was added to the culture. After 30 days of cultivation, 30 mL of culture medium was removed, and another 30 mL of BG11-N/N 0 medium was added to the culture. The substitution of the culture media was repeated monthly until further analysis. A total of 98 mixed-strain cultures were established.
In further steps, a total of 60 cyanobacterial strains were isolated to form the monocyanobacterial cultures. Isolation and purification were conducted through a series of alternate cultivation in liquid and solid (1.5% agar) BG-N 0 media [85]. Finally, the isolated strains were cultured in 50 mL of BG-N 0 medium in an incubator under a 14/10 (light/dark) illumination period (8000 K, 60 lux, 750 lm; Power-GLO, Rolf C. Hagen Inc., Mansfield, MA, USA) at 27 • C during four weeks. Approximately 10 mL of each strain culture was freeze-dried before extraction.

Extraction of Cyanotoxins
Extraction of microcystins was performed on ten biocrust samples collected in Serbia and 98 mixed-strain cultures. Extraction of cylindrospermopsin was performed on ten biocrust samples collected in Serbia and 35 mixed-strain cultures. Extractions were performed according to SOP_TOXIC_AAU_04F [89] with minor modifications. For the extraction of cyanotoxins from the biocrusts, approximately 1 g of the biocrust surface from paired samples containing cyanobacteria was scraped and soaked with 5 mL of 75% MeOH (J.T. Baker HPLC Gradient Grade, The Netherlands). Extraction continued with 15 min sonication in an ultrasonic bath, followed by 1 min sonication with a probe (3 mm microtip, 30% pulse, 30% energy; Bandelin Sonopuls HD2070). To avoid contamination, the probe was washed with 75% MeOH after each sample. Samples were then placed in a water bath at 50 • C for 30 min with shaking. The centrifugation at 2000× g for 15 min followed, and 3 mL of the supernatant was separated and evaporated to dryness under a stream of nitrogen at 50 • C. For analysis of cylindrospermopsin from the same vial, the extracts were dissolved with 300 µL of 50% MeOH (LC-MS Chromasolv, Riedel-de Haën™, Germany) and vortex shaken for 30 s. Finally, the extracts were filtered through a GHP ACRODISC 13 mm syringe filter with a 0.2 µm GHP membrane (Pall Corporation, New York, NY, USA).
For extraction of the same cyanotoxins from cyanobacteria in mixed cultures, approximately 100 mL of the culture was filtered through Whatman GF/C 47 mm glass microfiber filters (GE Healthcare UK Limited, Hatfield, UK). The biomass retained on the filters (50.2-141 mg) was freeze-dried and then dissolved in 3 mL of 75% MeOH (J.T. Baker HPLC Gradient Grade, Phillipsburg, NJ, USA). Samples were further sonicated for 15 min in an ultrasonic bath, followed by 1 min sonication with a probe (3 mm microtip, 30% pulse, 30% energy; Bandelin Sonopuls HD2070, Sigma-Aldrich, Burlington, MA, USA). The probe was washed with 75% MeOH after each sample. A volume of 1.5 mL from each sample was transferred to Eppendorf tubes and centrifuged at 10,000× g for 10 min. A volume of 1 mL of the supernatant was evaporated to dryness under a nitrogen stream at 50 • C. Samples were further processed in the same manner as biocrust samples.
Extraction of saxitoxin and gonyautoxin 2-, and 3 hydrochloride from 10 mixed-strain cultures was performed in 80% acetonitrile (ACN) containing 0.1% formic acid. First, the samples were sonicated in a bath for 15 min, after which each sample was sonicated separately with a probe (3 mm microtip, 30% pulse, 30% energy; Bandelin Sonopuls HD2070). Samples were then centrifuged at 10,000× g for 10 min. The supernatant was separated and filtered through a GHP Acrodisc Ø13 mm syringe filter with a 0.2 µm GHP membrane. The filtered extracts were stored at −20 • C until two layers were separated (upper-ACN; lower-H 2 O). The lower layer was separated and submitted to HPLC analysis.
BMAA extraction was based on [13]. First, 600 µL of 0.1 M trichloroacetic acid (TCA) was added to 12 mg of cyanobacterial BLC sample. After vortexing and standing at room temperature for 10 min, the sample was vortexed and centrifuged. The supernatant was transferred to a spin filter tube and centrifuged again for 5 min. The filtrate was transferred to a plastic tube. Then, 600 µL of TCA was again added to the pellet of the sample, and the extraction and centrifugation steps were repeated. The pooled filtrate was split for the extraction of free and soluble bound BMAA. For free BMAA extraction, 20 µL of a 2 mg/L D 3 BMAA solution in 20 mM HCl was added to 500 µL extract. These samples were dried under vacuum, and then reconstituted in 500 µL 67% acetonitrile/33% water/0.1% formic acid and transferred to a vial for analysis. For soluble bound BMAA, 450 µL of the pooled filtrate was transferred to a small glass vial. Next, 20 µL of D 3 BMAA solution was added and the samples were dried under vacuum. Then, 30 µL 6 M HCl was added, after which the samples were flushed with nitrogen and hydrolyzed at 105 • C for 20 h. After hydrolysis, the samples were again dried under vacuum and reconstituted in two times 250 µL 67% acetonitrile/33% water/0.1% formic acid. After centrifugation in a spin filter tube, the samples were transferred to a vial for analysis. Total BMAA was extracted by adding 20 µL D 3 BMAA solution to 0.5 mg of sample. The samples were dried, and then hydrolyzed in the same way as the extracts for soluble bound BMAA. The analyses of GTX 2/3 and STX in ten mixed-strain cultures were carried out using the ion-pair HPLC with post-column oxidation and fluorescence detection (HPLC-FLD) according to [92] on a Merck Hitachi LaChrom HPLC-system (Tokyo, Japan) coupled to a Hewlett-Packard Series 1100 Fluorescence detector. Extracts were loaded (10 µL injection volume) onto Waters Xbridge C18 150 mm × 3 mm I.D. column with 3.5 µm particles (Milford, MA, USA) at 40 • C. Solvent A was (6 mM octanesulfonic acid, 6 mM heptanesulfonic acid and 40 mM ammonium phosphate, 20% phosphoric acid, pH 7)-tetrahydrofuran, 99.25:0.75 (v/v). Solvent B was (7 mM octanesulfonic acid, 7 mM heptane sulfonic acid, 48 mM ammonium phosphate, 20% phosphoric acid, pH 7)-tetrahydrofuran-acetonitrile, 89:1:10 (v/v/v). The following pump gradient program was used: 0 min 100% A, 6 min 100% A (GTX2/3 elute), 7.5 min 100% B, 32 min 100% B (STX elue), 33 min 100% A, 45 min 100% A, and flow rate 0.55 mL/min. The oxidized derivatives of STX and GTX2/3 were detected at λ Ex 330 nm and λ Em 395 nm, following the post-column oxidation: 60 • C in a PTFE reaction coil (15 m × 0.3 mm I.D.) with 5 mM periodic acid and 275 mM ammonia in water at 0.3 mL/min. 0.38 M nitric acid at 0.4 mL min −1 , was used to lower the pH to acidic.

Chromatographic Analyses
In-house prepared microcystin reference materials containing MC-LR, dmMC-LR, MC-RR, dmMC-RR, MC-LY, MC-LW, MC-YR, and MC-LF obtained from PCC7820 Microcystis and NIES-107 Microcystis strains were used for identification of MCs [93]. Certified reference materials for CYN, STX and gonyautoxin 2-, and 3 hydrochloride (GTX2/3) were acquired from NRC-IMB (Institute for Marine Biosciences, Halifax, Canada). Table 2 shows the lowest analyte levels in the diluted reference materials which gave a signal-to-noise ratio (S/N) higher than three. BMAA analysis was performed as in [13]. Samples were analyzed without derivatization by LC-MS/MS. Quantification was performed against a calibration curve in solvent, and each sample was corrected for the recovery of the internal standard D 3 BMAA. Table 3 shows limit of detection (LOD) and limit of quantification (LOQ) values for BMAA.

DNA Extraction
Screening for cyanotoxin-coding genes was performed on 60 monocyanobacterial cultures (20:20:20 Serbia:Iran:China, Supplementary Materials File S1). Approximately 10 mg of freeze-dried biomass of both reference strains and culture samples was used for DNA extraction. Genomic DNA was extracted using the DNeasy Plant Mini Kit (QIAGEN, Hilden, Germany) according to the manufacturer's instructions. The quality and quantity of the extracts were assessed spectrophotometrically (NanoDrop ND-1000, Thermo Scientific, Waltham, MA, USA) through the A 260 /A 280 ratio.

Polymerase Chain Reaction (PCR)
Qualitative PCR was run to analyze samples for the presence of MC (mcyE), CYN (cyrJ), STX (sxtA, sxtG, and sxtS), and ANTX (anaC) synthetase genes in 60 cyanobacterial strains. The presence of cyanobacterial DNA in the samples was confirmed by PCR amplification of 16S rRNA gene. PCR reaction mixtures were prepared in a total volume of 20 µL containing 1 × Phire Reaction Buffer, 0.4 µL Phire II HotStart polymerase (Thermo Scientific, Waltham, MA, USA), 0.2 mM dNTPs (Thermo Scientific), 0.5 µL forward and reverse primers (Table 4), 2 µL of the template containing 0.35 to 0.5 ng/µL of genomic DNA, and sterile deionized water. The PCR protocol for the amplification of the 16S RNA gene [94] included: initial denaturation for 5 min at 94 • C; 30 cycles of denaturation for 45 s at 94 • C, primer annealing for 45 s at 57 • C, and strand elongation for 2 min at 68 • C; final elongation for 7 min at 68 • C. The PCR protocol for genes coding cyanotoxin production included: initial denaturation for 30 s at 98 • C; 40 cycles of denaturation for 5 s at 98 • C, primer annealing for 5 s at 52 • C or 61 • C or 62 • C (Table 4), strand elongation for 10 s at 72 • C; final elongation of 1 min at 72 • C (C1000 Touch Thermal Cycler (Bio-Rad, Hercules, CA, USA)). Potential inhibition of PCRs was assessed through an exogenous amplification control template containing 1 µL: 1 µL (reference: sample). Following strains were used as a reference in the control template: PCC7820 for mcyE, CS-506 for cyrJ, CS-537/13 for sxtA, sxtG, sxtS, and ANA123 for anaC. Visualization of PCR products was performed on a 1.5% Top Vision agarose gel (Thermo Scientific) dyed with SYBR ® Safe DNA gel stain. The bands were documented on Gel Doc™ XR (Bio-Rad, Hercules, CA, USA) using Quantity One software (v. 4.6.9) (Bio-Rad, Hercules, CA, USA).

Reference Strains for PCR Analysis
The reference strains for PCR analysis were obtained from Finnish Environment Institute (