The microbiome of modern microbialites in Bacalar Lagoon, Mexico

Microbialites are highly diverse microbial communities that represent modern examples of the oldest life forms, stromatolites (dated >3.7 Ga). Bacalar Lagoon, in Mexico, harbors the largest freshwater microbialite occurrences of the world; yet diverse anthropogenic activities are changing the oligotrophic conditions of the lagoon. The objective of this work was to perform a spatial exploration of the microbialites of Bacalar Lagoon, analyze their prokaryote diversity, following a high throughput sequencing approach of the V4 region of the 16S rDNA, and correlate to the environmental parameters that influence the structure of these communities. The results indicate the presence of microbialites throughout the periphery of the lagoon. The microbiome of the microbialites is composed primarily of Proteobacteria (40–80%), Cyanobacteria (1–11%), Bacteroidetes (7–8%), Chloroflexi (8–14%), Firmicutes (1–23%), Planctomycetes (1–8%), and Verrucomicrobia (1–4%). Phylogenetic distance analyses suggests two distinct groups of microbialites associated with regions in the lagoon that have differences in their environmental parameters, including soluble reactive silicate (in the north), bicarbonates and available forms of nitrogen (ammonium, nitrates and nitrites) (in the south). These microbialite groups had differences in their microbiome composition associated to strong anthropogenic pressure on water quality (agriculture, landfill leachate, lack of water treatment infrastructure and intensive tourism), which were related to a loss of microbial diversity.


Introduction
Bacteria and Archaea (prokaryotes) represent the most diverse and abundant organisms on the planet [1]. They are involved in maintaining and controlling biogeochemical cycling of the fundamental elements of life (H, C, N, O, S and P) [2]. Understanding the multiple ecological and evolutionary processes that are related to the distribution and structure of prokaryote diversity at the local and global scales is a main interest of microbial ecology [3][4]. The formation of biogeographic distribution patterns in prokaryotes is determined by environmental heterogeneity (ecological factor) and dispersion (historical factor) [5]. At the local scale, factors that include pH, habitat heterogeneity, system productivity, and more recently, human a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 bacteria [38][39]. In this study we wanted to answer if habitat transformation of Bacalar Lagoon influences microbialite community structure and composition. We characterized the microbiome of Bacalar microbialites throughout the lagoon and analyzed if there are structuring effects on their prokaryote composition related to environmental variables, following a next-gen sequencing approach of the V4 hypervariable region of the 16S rDNA gene.

Study site
Bacalar is a karstic and freshwater lagoon located in the southeast of Quintana Roo, Mexico in the Yucatan peninsula (Fig 1, S1 Table). The lagoon is a geological fault due to its orientation and shape. Bacalar Lagoon, has been considered an oligotrophic system due to the low concentration of nitrogen (N) and phosphorus (P), and is part of the Transverse Coastal Corridor, a complex water system, where a series of karst freshwater lakes, lagoons and estuaries are connected through underground water flows

Sample collection
Microbialites were collected in 15 sites along the western shore of Bacalar Lagoon along a north-south gradient. Cores of approximately 2.5 cm in diameter were sampled in duplicates from individual microbialite heads, and three to five individuals per site were sampled (Fig 1, S1 Table). Samples were taken with gloves and sterile material to avoid cross-site contamination. Collection was carried out during the spring of 2018. Samples were stored at 4˚C during transport to the laboratory where they remained frozen at -70˚C until processed. All microbialite samples were carried under collector permit PPF/DGOPA-113/14 awarded by SEMAR-NAT, Mexico. Field studies did not involve endangered or protected species. Three water samples (500 mL) were taken at each sampling point using Nalgene bottles, previously washed with 15% HCl, and were filtered (0.22 μm Millipore membrane) in situ and stored at 4˚C for dissolved nutrients analysis. In situ conductivity, pH and temperature were measured using a YSI Professional handheld (YSI model Pro 30) and pH-meter (Hanna HI 9146).
The degree of tourist visitation per site was assessed during the sampling with interviews to locals. A high level represents sites that have tourism throughout the year; medium represents sites that only have tourists during holiday seasons; low represent sites that are seldom visited by tourists.
Principal Component Analysis (PCA) was used to describe the relationship between the chemical variables measured in the water with each sampling location. The compiled data set representing the environmental variables analyzed in this study was transformed into a "site x variable" matrix. Euclidean distance and ordinations were plotted with FactoMineR and factoextra in Rstudio [44].

Biogeochemical analysis
The biogeochemical analyzes of the microbialite fabric were carried out with different methodologies. For total carbon and nitrogen we used a soil analyzer (Thermo Scientific Flash 2000). Barnard

X-Ray Diffraction (XRD)
For XRD analyses samples were cold dried (10˚C), homogenized with a pestle and agate mortar and sieved through a mesh < 75μm. The measurement was made in the angular interval 2θ from 5˚to 80˚in step scanner with a "step scan" of 0.003˚and an integration time of 40 sec per step, using double-side aluminum holders (unoriented fractions). Each diffractogram was

PLOS ONE
The microbiome of modern microbialites in Bacalar Lagoon, Mexico obtained in a diffractometer (Empyrean) equipped with a Ni filter, a monochromator, a thin tube focus copper and PIXcel3D detector. The diffraction patterns were analyzed with the HighScore software (version 4.5) with reference patterns from the ICDDPDF-2 and ICSD databases. All determinations were done in the X-Ray Diffraction Laboratory, Institute of Geology, UNAM.
Total DNA extraction and 16S rDNA amplification DNA extractions of microbialite samples (0.25 g) were done in triplicate using the DNeasy PowerSoil1 Kit (Qiagen) following the manufacturer's instructions. Amplifications of the 16S rDNA V4 region were done following an established protocol [48]. Each sample was amplified in three independent PCR reactions. PCR conditions were: 98˚C for 30 s followed by 35 cycles of 95˚C for 30 s, 52˚C for 40 s, and 72˚C for 90 s, and a final elongation step of 12 min at 72˚C, then kept at 4˚C. PCR products were pooled and purified with Ampliclean carboxyl-coated magnetic beads (NimaGen, NDL). The purified amplicon library was quantified with a QUBIT fluorometer (Promega, USA). The amplicon library with 20 ng/μl sample was sequenced on an Illumina MiSeq 2 x 300 platform (Yale Center for Genome Analysis, CT, USA).

Analysis of Illumina 16S rDNA V4 sequences
The 16S rDNA V4 sequences of 90 samples of microbialites collected throughout Bacalar Lagoon, were deposited in the GenBank under BioProject PRJNA 550210. In addition, the data used during the analyses are available in the Open Science Framework: https://osf.io/ zme9y/. Sequences were denoised, chimera and singletons were removed, then sequences were assigned into ASVs (Amplicon Sequence Variants) in QIIME2 (v.2018.6) [49] and truncated at position 200 with DADA2 [50] using the plugin qiime dada2 denoised-paired. ASVs representing less than 0.01% of the sequences across the dataset were eliminated. Taxonomy was resolved using the SILVA database (release 132-99% OTUs, 515-806 region), with the featureclassifier classify-consensus-vsearch (v2.9.0) plugin [51]. Mitochondrial and chloroplast sequences were filtered out from the feature table before rarefaction. Rarefaction was done at 10,000 ASVs per sample, resulting in the removal of 14 samples that had less than 9,000 sequences. The total dataset includes 90 samples for 15 sites.
After QIIME analyses, all sequence data were analyzed using multivariate correlational and ordination methods in the R statistical environment (version 3.6.2), for this, we used Phyloseq R [52]. We considered using the R markdown document that contains the complete commands for the analysis which is available here: https://github.com/YanezAlfredo/Themicrobiome-microbialites-in-Bacalar-Lagoon-Mexico.git. The weighted Unifrac matrix was used to calculate the dissimilarity between the groups (D). The associations between environment and prokaryote community structure from different sites are shown using a constrained multidimensional scaling by Canonical Analysis of Principal Coordinates (CAP) based on weighted unifrac distance dissimilarity [53]. The differences between regions in the lagoon was analyzed using the PERMANOVA approach [54], implemented in "vegan" as the ADO-NIS function using R package.
The ASV table was used to construct the biological matrix of genetic diversity based on 16S rDNA taxonomy. The alpha diversity indices such as species and Shannon index were calculated with the R package "vegan" [https://cran.r-project.org, https://github.com/vegandevs/ vegan]. Wilcoxon tests were used to test for group differences in microbial diversity. A Venn diagram was created to compare the North-Center and South-Center regions obtained by unifrac weighted analysis, using the DrawVenn tool available online (http://bioinformatics.psb. ugent.be/webtools/Venn/).The total sum of-squares of the community composition matrix was partitioned into additive components of species (ASVs) to obtain their contributions to beta diversity (SCBD) and the local contributions of individual sampling units to beta diversity (LCBD) [55]. Following Legendre and De Cáceres [56], we first transformed (Hellinger) the species abundance per site matrix and then we calculated multiple-site β-diversity indices (betapart) [57]; LCBD and SCBD indices were ran in adespatial [58], ade4 [59] and with beta. div functions in "vegan" [56].

The physicochemical environment surrounding microbialites in Bacalar Lagoon
The survey conducted in Bacalar Lagoon suggested an overall north-south gradient defined by higher conductivity and SRSi in the north, while the southern region had higher values of bicarbonate and available forms of nitrogen (ammonium, NOx), with similar values of sulfate and calcium throughout the lagoon (S1 Table). In the PCA, two general gradients were observed in Bacalar Lagoon. A north-south gradient based on PC1, where the following correlations were made: SRSi to  (Fig 2). Nonetheless, certain variables separated these regions, including bicarbonates, NOx (NO 3 -+ NO -2 ) and ammonium in the center-southern sampling sites (B 12-15) which increase near urban areas (Fig 2). Likewise, an analysis of previous research in the lagoon showed that concentrations of nitrates and ammonium increased two orders of magnitude between 2008 and 2018 in the southern sampling sites [8, 30, 60, this study] (Table 1).

Microbialite genetic composition (16S rDNA V4)
A total of 4,167,392 reads were obtained for the 16S rDNA V4 hypervariable region. The mean number of sequences per site was 40,071. To include samples from all sites we defined a rarefaction at 10,000 sequences per subsample per site. All the microbialites were fully characterized at this sampling coverage.
The UniFrac weighted distance matrix separated Bacalar microbialites in two phylogenetically differentiated microbial communities. This result allowed us to classify the 15 sampling sites into two regions (S2 Fig). The first region was defined as North-Center and included sites

PLOS ONE
The microbiome of modern microbialites in Bacalar Lagoon, Mexico B 1-3 and B 6-9, which represented 80% of the global microbial diversity and were very similar between them (D = 0.82). The second region was defined as South-Center with sites B 4-5 and B 10-15. Overall, the CAP of the genetic diversity matrix and environmental dataset, suggested that the factors that correlate in the South-Center region of Bacalar Lagoon with microbialite diversity are the concentrations of available forms of N (NO 3 and NH 4 + , respectively) (Fig 4). PERMANOVA analysis also indicated that the differences between regions in Bacalar Lagoon were significant (p<0.05).
A Mann-Whitney-Wilcoxon test was conducted to compare the richness and diversity indices between the North-Center and South-Center regions. Several diversity indices demonstrated that the microbiome diversity of the North-Center was significantly greater than that observed in the South-Center region (p < 0.01) (Fig 5). The Shannon index indicated that the sampling sites North-Center of the town of Bacalar had a greater bacterial diversity (H´= 5.7), and the sites to the South-Center had 42% less diversity (H' = 3.3).

PLOS ONE
The microbiome of modern microbialites in Bacalar Lagoon, Mexico microbialites, respectively. Six phyla (Altiarchaeota, Asgardaeota, Diapherotrites, Euryarchaeota, Nanoarchaeaeota and Thaumarchaeota) were identified in this study. Again, the microbialites in the North-Center had the greatest diversity, where Heimdallarchaeia, Woesearchaeia and Nitrososphaeria were the most abundant (Fig 6J).
To elucidate why these significant changes in community structure were occurring and which taxa were associated with variations at each site, we used the LCBD and SCBD metrics, as proposed by Legendre and De Cáceres [56]. The highest and most significant differences in LCBDs were found at sites B 5, B 12 and B 13 (p<0.05). The SCBD showed that Pseudomonas, Aeromonas, Stenotrophomonas, Acinetobacter, Bacillus, Chryseobacterium, Achromobacter, Brevundimonas and Bacillus were bacterial genera that contributed mostly to community structure substitution.

PLOS ONE
The microbiome of modern microbialites in Bacalar Lagoon, Mexico

Discussion
The Yucatan peninsula is an emerged carbonated continental platform. The northern region emerged during the Paleogene and Neogene, while the south began to rise in the Oligocene   [66] reported that southern Quintana Roo has high rates of infiltration of nutrients-such as nitrates-and there is a high risk of contamination of the aquifer by human activities, such as agriculture.
Economic development and population growth are direct threats to freshwater ecosystems

Biogeochemical and mineralogical and characterization of Bacalar Lagoon microbialites
Microbialite are spatially distributed throughout Bacalar Lagoon. Bacalar Lagoon's hydrogeochemical dynamics make it different from other sites with microbialites in seawater and continental environments. All the microbialites analyzed in Bacalar Lagoon share mineral composition (CaCO 3 ,~97%). Valdespino et al., [23] reported a similar mineralogy for the microbialites of Bacalar Lagoon and Cuatro Cienegas Basin, which are water bodies of karstic origin. Bacalar Lagoon, which is located in the evaporative hydrogeochemical region [62], presents carbonate dissolution processes of the subterranean water tunnels that reach the lagoon and the walls (carbonate rock) of the cenotes within the lagoon, favoring the saturation of bicarbonates [37, 72]. The development of larger microbialites in the south of Bacalar might be associated to bicarbonate saturation. Chagas et al., [27] also report for lacustrine systems with microbialites such as Lake Pavilion, Lake Van, Cuatro Cienegas Basin, Alchichica and Clifton, that calcite minerals and aragonite are the main minerals in microbialites.
Depending on the chemistry of the water and the bacterial community, microbialites present a diverse range of minerals, although generally they have been reported in greater percentage aragonite, hydromagnesite, gypsum and calcite [23, 73]. Cyanobacteria such as Pleurocapsales and Chroococcales and Alphaproteobacteria are associated with the formation of aragonite in microbialites from Lake Alchichica (Mexico), a Mg-rich hyperalkaline crater lake (pH 8.9), while in Cuatro Cienegas and Bacalar, a S-rich karstic system, filamentous cyanobacteria and Sulfate Reducing Bacteria (SRB) favor calcite precipitation [39]. The hydrogeochemical conditions in Bacalar Lagoon favor the presence of bacterial groups (including cyanobacteria and S-cycling bacteria) that are involved in carbonate precipitation processes [20, 22, 74] and SRB reduce sulfates to sulfides with a consequent oxidation of organic carbon to bicarbonates. They contribute to a state of saturation, which occurs within the EPS matrix (associated mainly to cyanobacterial activity), precipitated by cyanobacteria in an alkaline pH, where calcium ions finally precipitate as CaCO 3 [33, 75]. We report the same groups of SRB (Desulfovibrionales, Desulfobacteraceae, Syntrophobacteraceae, Desulfobulbaceae and Desulfomonile), distributed in all Bacalar Lagoon sites and reported by Johnson et al., [39]. All of these SRB have larger abundances in microbialites of the North-Center region.
In addition, Bacalar Lagoon microbialites have been described for their interactions with organisms such as gastropods (Pomacea flagellata), bivalves (Mytilopsis sallei), nematodes and mangroves [38, 76]. Johnson et al., [39] reported the presence of Cyanobacteria and Rhizobiales, a nitrogen-fixing Alphaproteobacteria, in the microbialites associated with mangroves in their study sites in southern Bacalar Lagoon.

Bacterial community structure of the microbialites of Bacalar Lagoon
This study proposes the presence of two phylogenetically differentiated communities in Bacalar Lagoon microbialites. Generally, studies mention that population differences occur in biogeographic patterns with the differentiation of niches at large geographic scales [77]. However, within ecosystems, biogeographic regionalization is possible due to the presence of gradients that induce changes in biological communities [3]. Currently, anthropological activities can be considered a selective force, either physically (implementation of infrastructure) or by chemical alteration, which includes eutrophication of water bodies related to nutrient availability [12].
The microbialite sites that represent the South-Center region of this study are located near the city of Bacalar, and to the south of the lagoon. These sites are associated to urban development in the shoreline of the lagoon which lack infrastructure for domestic water treatment, have leaking septic systems, agriculture and intense tourist activity, that are causing trophic affectation in the system [68, [78][79][80]. Alterations in water quality was related to changes in the structure of the microbiome of microbialites between the North-Center and the South-Center regions Bacalar Lagoon. Recently, the work of Lindsay et al., [13] reported that in Great Salt Lake (GSL), USA, the bacterial community of microbialites responded to anthropogenic perturbation of the system related to construction of a railroad causeway. These authors demonstrated that microbialites in less disturbed areas of GSL have a greater abundance of cyanobacteria and diatoms compared to the almost total absence of these organisms in the microbialites where disturbance exists. Therefore, the monitoring of the community diversity of the microbialites, could be a strategy to know how bacterial groups react to the processes of alteration of the environment [81], either before a physical affectation or through the chemical changes of the water as in the case of Bacalar Lagoon.
Microbialites in the world maintain, regardless of their geographical region, a similar composition at the phylum level [8, 23]. Actinobacteria, Bacteroidetes, Cyanobacteria and Proteobacteria [8, 37-39], are common components of microbialites. Bacalar Lagoon microbialites in the North-Center region have a high diversity (H´= 5.7) (Fig 5), which contributes to understand that oligotrophy is not a limiting factor in the development of complex communities [81]. The decrease of almost half of the bacterial diversity in the South-Center region is associated to dominance of specific microbes of the Gammaproteobacteria and Firmicutes groups.
Cyanobacteria were most abundant in the North-Center region. Considering that the South-Center region of Bacalar Lagoon is suffering an increase of nutrients due to anthropogenic activities, our results coincide with other works where it is reported that cyanobacteria are more diverse in oligotrophic waters than in eutrophic waters [82]. Cyanobacteria, a phylum that is relevant in EPS formation and has been considered to form nucleation sites for carbonate precipitation [83][84], showed a greater abundance and diversity in the North-Center microbialites of Bacalar Lagoon. The North-Center region also presented a higher diversity of Planctomycetes and Verrucomicrobia (~8.4% and~3.8%, respectively), both forming part of a taxonomic super phylum called PVC [85-86], described with a relative abundance between 7-12% in different microbialites of the world [8,86]. Recently, the presence of these bacteria was correlated in places where calcite crystals predominated [23]. Chloroflexi, an anoxygenic phototrophic phylum, which participates in the "alkaline machinery" which in combination with oxygenic photosynthesis by cyanobacteria and sulfate reduction, promote the precipitation of carbonated minerals [20, 36], was also more abundant in the North-Center region. This would suggest that loss of cyanobacterial, PVC and chloroflexi diversity could affect microbialite growth and maintenance in the South-Center region of Bacalar.
Further, the microbiome of microbialites in the South-Center region presented a high abundance of Firmicutes (~23.3%). This group occupies between 0-2% of relative abundance in other microbialites of the world

Changes in the bacterial community of microbialites in Bacalar Lagoon
It is important to define the factors that are causing the environmental disturbance of a system, especially if it is due to human activities [90]. We used the LCBD-SCBD metrics and a CAP to associate the environmental variants of home site (niche) and the association with their bacterial community (dispersion) [91]. Legendre and De Cáceres [56] mention that high values of LCBD indicate the degree of ecological singularity of each sampling site. From this perspective, sites with high values of LCBD may contain unusual species or are sites that respond to human disturbance [92]. In both cases, the use of beta-diversity metrics can be a starting point for decision-making in conservation or ecological restoration scenarios [93]. In this study, the highest values of LCBD were related to sites in the South-Center region. The sites B5, B 12 and 13, obtained the highest LCBD values. Site B5 is a particular case of microbialite growth that has a strong correlation in the CAP to ammonium. Further research is needed to identify the sources of ammonium to this specific area in Bacalar Lagoon that shows an increase in domestic and tourist developments.
The Mexican Caribbean is an area whose economy depends mainly on tourism related to its natural resources [60,94]. Particularly within Quintana Roo, places like Cancun and Playa del Carmen that have intense tourist activity, show affectations to the water quality of the underground aquifer systems and cenotes [95][96]. Currently, Bacalar presents an increase in tourist occupation. According to the Mexican Government office of statistics (INEGI-SecTur, 2019) Bacalar Lagoon received approximately 90,000 tourists in 2018, which was twice the amount of tourist visitation in 2017 (45,000). This phenomenon is likely to continue and the infrastructure to accommodate these visitors is not available. Tobón-Velázquez et al., [60] mentioned that the lack of infrastructure regulation from the government could result in the degradation of the water quality of Bacalar Lagoon, hence affecting the microbialites.
A direct correlation is reported between the most visited sites for tourists and the lowest prokaryote diversity. In addition, the sites with the lowest diversity that are located at the South-Center region of the lagoon, are the same sites that have been historically used for tourism. These results postulate that the changes in the microbiome of microbialites along Bacalar Lagoon are probably associated to a greater extent, with poor water quality due to high concentrations of ammonium and NOx  (Table 1). After an environmental disturbance, the possibility of a community of returning to its previous state will depend on its genetic and physiological diversity [105], yet so far, no research has demonstrated that microbialite communities can recover in the short term.
A hypothesis that rises from this study is that microbialites in Bacalar Lagoon have the same phylogenetic origin, yet disturbances in water quality detected in the South-Center region are causing loss of biodiversity. Another possible explanation is that high concentrations of carbonate present in the South-Center region, promote larger and faster microbialite growth, which is associated to a different community structure, differing from their North-Center counterparts. So far, we do not have elements to prove any of these open questions, but we do know that microbialites have fundamental biological constituents, where Cyanobacteria and bacteria associated to S-cycling are the main contributors to microbialite formation and growth. We still need to understand the dynamics of the communities that form microbialites, while trying to document their transformations in fragile habitats, like the tropical lagoon that is represented in this study. The increase in available forms of nitrogen is preoccupying to say the least since our research shows this is associated to lack of water treatment and planned agriculture in the region. How much can the native communities, represented in this study by microbialites, deal with the rate of change that human activities cause in the environment?
Supporting information S1