Temporal Analysis of the Microbial Community from the Crystallizer Ponds in Cabo Rojo, Puerto Rico, Using Metagenomics

Abstract

The Cabo Rojo solar salterns are a hypersaline environment located in a tropical climate, where conditions remain stable throughout the year. These conditions can favor the establishment of steady microbial communities. Little is known about the microbial composition that thrives in hypersaline environments in the tropics. The main goal of this study was to assess the microbial diversity present in the crystallizer ponds of Cabo Rojo, in terms of structure and metabolic processes across time using metagenomic techniques. Three samplings (December 2014, March and July 2016) were carried out, where water samples (50 L each) were filtered through a Millipore pressurized filtering system. DNA was subsequently extracted using physical-chemical methods and sequenced using paired end Illumina technologies. The sequencing effort produced three paired end libraries with a total of 111,816,040 reads, that were subsequently assembled into three metagenomes. Out of the phyla detected, the microbial diversity was dominated in all three samples by Euryarchaeota, followed by Bacteroidetes and Proteobacteria. However, sample MFF1 (for Muestreo Final Fraternidad) exhibited a higher diversity, with 12 prokaryotic phyla detected at 34% NaCl (w/v), when compared to samples MFF2 and MFF3, which only exhibited three phyla. Precipitation events might be one of the contributing factors to the change in the microbial community composition through time. Diversity at genus level revealed a more stable community structure, with an overwhelming dominance of the square archaeon Haloquadratum in the three metagenomes. Furthermore, functional annotation was carried out in order to detect genes related to metabolic processes, such as carbon, nitrogen, and sulfur cycles. The presence of gene sequences related to nitrogen fixation, ammonia oxidation, sulfate reduction, sulfur oxidation, and phosphate solubilization were detected. Through binning methods, four putative novel genomes were obtained, including a possible novel genus belonging to the Bacteroidetes and possible new species for the genera Natronomonas, Halomicrobium, and Haloquadratum. Using a metagenomic approach, a 3-year study has been performed in a Caribbean hypersaline environment. When compared to other salterns around the world, the Cabo Rojo salterns harbor a similar community composition, which is stable through time. Moreover, an analysis of gene composition highlights the importance of the microbial community in the biogeochemical cycles at hypersaline environments.

Keywords: Caribbean; Puerto Rico; halophilic archaea; hypersaline; metagenomics.

Figure 1

Aerial map of the Solar Salterns of Cabo Rojo (17°57′25.2″ N, 67°11′58.0″ W). Water samples (5 L) were obtained from ten crystallizers (circled in black) and pooled together into one sample (50 L). Three samplings were performed on the same crystallizers in rainy (December 2014), and dry (March and July 2016) seasons. Photo taken by “Puerto Rico desde el Aire” reproduced with permission.

Figure 2

Taxonomic hits by phylum. Each slice indicates the number of reads with predicted 16SrRNA genes annotated to the indicated phylum. Samples were named MFF1 for the first sampling, MFF2 for the second sampling and MFF3 for the third sampling (MFF stands for “Muestra Final Fraternidad” which means Fraternidad Final Sample). Phylum Euryarchaeota were shown to be dominant in the three samples (73.69% for MFF1, 77.92% for MFF2, and 78.75% for MFF3), followed by Bacteroidetes (16.59, 22.03, 21.18%, respectively), and Proteobacteria (3.03, 0.05, 0.07%, respectively). Other groups include Acidobacteria, Chlamydiae, Cyanobacteria, Deinococcus-Thermus, Fusobacteria, Planctomycetes, and Verrucomicrobia, with less than 1% in MFF1.

Figure 3

KEGG Orthology (KO) of functional genes obtained from samples MFF1, MFF2, and MFF3. Genes related to metabolism dominate in about 60% of the sequences in all three samples, followed by genetic information processing (23%), environmental information processing (12%), and cellular processes (3%). Other funtions are less than 1%. Samples were named MFF1 for the first sampling, MFF2 for the second sampling, and MFF3 for the third sampling (MFF stands for “Muestra Final Fraternidad” which means Fraternidad Final Sample).

Figure 4

(a) Carbon metabolism pathways detected in the three metagenomes. Colored enzymes indicate the presence of the enzyme, light blue indicates presence in MFF1, red indicates presence in MFF2, and orange indicates presence in MFF3 (number of hits for each enzyme can be seen at Table S1 of supplementary figures). (b) Reaction catalyzed by the enzyme Ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO). The enzyme is present in all three metagenomes. Pathways obtained from KEGG pathways (https://www.genome.jp/kegg/pathway.html) with permission.

Figure 5

Nitrogen metabolism pathways present in the three metagenomes. Colored enzymes (Light blue and green for MFF1, red for MFF2, orange for MFF3) indicate the presence of hits related to the enzyme in the metagenome (number of hits for each enzyme can be seen at Table S1 of supplementary figures). Enzymes related to reductive pathways in the nitrogen cycle were encountered, including nitrate reductase (Enzyme Nomenclature database numbers (EC) 1.7.1.1, 1.7.1.3, 1.7.99.4), nitrite reductases (EC 1.7.2.1, 1.7.7.1, 1.7.1.4), and the nitrogenase needed for nitrogen fixation (EC 1.18.6.1). Pathways obtained from KEGG pathways (https://www.genome.jp/kegg/pathway.html) with permission.

Figure 6

Sulfur metabolism pathways present in the metagenomes. Colored enzymes indicate the presence of hits related to the enzyme in the metagenome (number of hits for each enzyme can be seen at Table S1 of supplementary figures). Enzymes related to sulfate reduction (2.7.7.4, 1.8.99.2, 1.8.1.2, 1.8.7.1) were encountered. Pathways obtained from KEGG pathways (https://www.genome.jp/kegg/pathway.html) with permission.

Figure 7

(a) Phylogeny for RC33 using amino acid identity. The scale represents change of amino acid substitution over time. (b) Subsystem category distribution for RC33. The graph represents the number of proteins that were grouped into a specific subsystems; 883 from a total 1570 coding sequences were identified to fit into subsystems. This chart was generated using Rapid Annotation System Technology (RAST).

Figure 7

(a) Phylogeny for RC33 using amino acid identity. The scale represents change of amino acid substitution over time. (b) Subsystem category distribution for RC33. The graph represents the number of proteins that were grouped into a specific subsystems; 883 from a total 1570 coding sequences were identified to fit into subsystems. This chart was generated using Rapid Annotation System Technology (RAST).

Figure 8

(a) Phylogeny for RC24 using AAI. The scale represents the number of amino acid substitutions over time. (b) Subsystem category distribution for RC24. The graph represents the number of proteins that were grouped into a specific subsystem; 723 of 1561 coding sequences were identified to fit into subsystems. This chart was generated using Rapid Annotation System Technology (RAST).

Figure 9

(a) Phylogeny for RC39 using AAI. The scale represents the number of amino acid substitutions over time. (b) Subsystem category distribution for RC39. The graph represents the number of proteins that were grouped into a specific subsystems; 1366 of 2439 coding sequences were identified to fit into subsystems. This chart was generated using Rapid Annotation System Technology (RAST).

Figure 10

(a) Phylogeny for RC20 using AAI. The scale represents the amount of amino acid identity change over time. (b) Subsystem category distribution BIN20. The graph represents the number of proteins that were grouped into a specific subsystem; 1579 of 4980 coding sequences were identified to fit into subsystems. This chart was generated using Rapid Annotation System Technology (RAST).

Figure 10

(a) Phylogeny for RC20 using AAI. The scale represents the amount of amino acid identity change over time. (b) Subsystem category distribution BIN20. The graph represents the number of proteins that were grouped into a specific subsystem; 1579 of 4980 coding sequences were identified to fit into subsystems. This chart was generated using Rapid Annotation System Technology (RAST).