Functional responses of key marine bacteria to environmental change - toward genetic counselling for coastal waters

A benchmark analysis of feature selection and machine learning methods for environmental metabarcoding datasets

Next-Generation Sequencing methods like DNA metabarcoding enable the generation of large community composition datasets and have grown instrumental in many branches of ecology in recent years. However, the sparsity, compositionality, and high dimensionality of metabarcoding datasets pose challenges in data analysis. In theory, feature selection methods improve the analyzability of eDNA metabarcoding datasets by identifying a subset of informative taxa that are relevant for a certain task and discarding those that are redundant or irrelevant. However, general guidelines on selecting a feature selection method for application to a given setting are lacking. Here, we report a comparison of feature selection methods in a supervised machine learning setup across 13 environmental metabarcoding datasets with differing characteristics. We evaluate workflows that consist of data preprocessing, feature selection and a machine learning model by their ability to capture the ecological relationship between the microbial community composition and environmental parameters. Our results demonstrate that, while the optimal feature selection approach depends on dataset characteristics, feature selection is more likely to impair model performance than to improve it for tree ensemble models like Random Forests. Furthermore, our results show that calculating relative counts impairs model performance, which suggests that novel methods to combat the compositionality of metabarcoding data are required.

Discovery of a phylogenetically novel tropical marine Gammaproteobacteria elucidated from assembled genomes and the proposed transfer of the genus Umboniibacter from the family Cellvibrionaceae to Umboniibacteraceae fam. nov

Marine heterotrophic bacteria in coastal waters respond to the influx of carbon from natural and anthropogenic sources. We identified two nearly identical, (99.9% average nucleotide identity; 100% amino acid identity; same DNA G + C content of 52.3 mol%) high-quality (≥99% CheckM completeness and ≤ 1.3% contamination) draft metagenome-assembled genomes (MAGs; SJ0813 and SJ0972) from seawater microbiomes of a southern island of Singapore that is in a protected marine park. The MAGs were only assigned to the Cellvibrionaceae family according to Genome Taxonomy Database. Overall genome related indices to Pseudomaricurvus alkylphenolicus KU41GT as the closest phylogenetic relative revealed no more than 70.45% average nucleotide identity (ANIcutoff < 95%), below the 50% percentage of conserved proteins (POCPcutoff = 43.54%) for genera cutoff and low digital DNA-DNA hybridization values (DDH = 20.6 and 20.8%). The major respiratory quinone is predicted to be ubiquinone-9 from the annotation of 3-demethylubiquinone-9 3-methyltransferase (ubiG, K00568) involved in the last step of the ubiquinone biosynthesis pathway (M00117), which differed from the ubiquinone-8 utilized by known members of Cellvibrionaceae. Both MAGs contained a complete pathway for dissimilatory nitrate reduction to ammonia, which increases bioavailability of nitrogen in seawater. An identical choline dehydrogenase found in both MAGs have a low amino-acid identity (≤64.47%) compared to existing GMC family oxidoreductases, expanding on the diversity of this family of enzymes. The MAGs meet nearly all the minimum requirements but lack a 16S rRNA gene of sufficient length required for the proposed novel genus and species under SeqCode. Nevertheless, phylogenetic trees based on core-genome and RpoB as an alternative phylogenetic marker are congruent with the taxon standing as a monophyletic clade to other taxa of the order Cellvibrionales. Taken together, the MAGs (SJ0813 and SJ0972) represent an uncultured, undescribed genus and species in which we tentatively propose the name Candidatus Pelagadaptatus aseana gen. nov., sp. nov. and strain SJ0813TS (=BAABNI000000000.1TS) as type sequence. Phylogenetic inference from core-genome and RpoB phylogenetic trees placed Umboniibacter marinipuniceus KMM 3891T outside Cellvibrionaceae. We, therefore, propose the transfer of the genus Umboniibacter from the family Cellvibrionaceae to a new family Umboniibacteraceae according to the International Code of Nomenclature of Prokaryotes.

Climate change will amplify the impacts of harmful microorganisms in aquatic ecosystems

More than 70% of the human population lives within five kilometres of a natural water feature. These aquatic ecosystems are heavily used for resource provision and recreation, and represent the interface between human populations and aquatic microbiomes, which can sometimes negatively impact human health. Diverse species of endemic aquatic microorganisms, including toxic microalgae and pathogenic bacteria, can be harmful to humans. Aquatic ecosystems are also subject to intrusions of allochthonous pathogenic microorganisms through pollution and runoff. Notably, environmental processes that amplify the abundance and impact of harmful aquatic microorganisms are occurring with increasing frequency owing to climate change. For instance, increases in water temperature stimulate outbreaks of pathogenic and toxic species, whereas more intense precipitation events escalate microbial contamination from stormwater discharge. In this Perspective we discuss the influence of aquatic microbiomes on the health and economies of human populations and examine how climate change is increasing these impacts.

Metagenomic Analysis of Surface Waters and Wastewater in the Colombian Andean Highlands: Implications for Health and Disease

Urban water bodies serve as critical reservoirs of microbial diversity, with major implications for public health and environmental quality. This study aimed to characterize the microbial diversity of surface waters and wastewater from the Pasto River in the Colombian Andean Highlands, offering insights that may support water quality monitoring efforts. Sampling was conducted at three river sites and one wastewater location. Standard physicochemical and microbiological analyses were performed, including real-time PCR to detect protozoan pathogens Giardia spp. and Cryptosporidium spp. Metagenomic sequencing provided an in-depth taxonomic and functional profile of microbial communities through two complementary approaches: (i) read-based analysis to identify abundant families and species, both pathogenic and beneficial, and (ii) detection of health-related molecular markers, including antimicrobial resistance markers and virulence factors. Physicochemical analyses showed distinct profiles between wastewater and surface water, with wastewater exhibiting elevated levels of suspended solids (113.6 mg/L), biochemical oxygen demand (BOD, 311.2 mg/L), and chemical oxygen demand (COD, 426.7 mg/L). Real-time PCR detected Giardia spp. DNA in 75% (76/102) of the samples and Cryptosporidium spp. DNA in 94% (96/102) of samples. The metagenomic read-based profiling identified Aeromonas media as a prevalent pathogen and Polaromonas naphthalenivorans as a potential biodegradative agent. The metagenomic assembly produced 270 high-quality genomes, revealing 16 bacterial species (e. g., Acinetobacter johnsonii and Megamonas funiformis) that provided insights into fecal contaminants and native aquatic microbes. Functional profiling further revealed a high prevalence of antimicrobial resistance markers, particularly for tetracyclines, aminoglycosides, and macrolides, with the highest abundance found in wastewater samples. Additionally, virulence factors were notably present in Zoogloea ramigera. The findings underscore the value of metagenomic profiling as a comprehensive tool for water quality monitoring, facilitating the detection of pathogens, beneficial species, and molecular markers indicative of potential health risks. This approach supports continuous monitoring efforts, offering actionable data for water management strategies to safeguard public health and maintain ecological integrity.

Thalassotalea aquiviva sp. nov., and Thalassotalea maritima sp. nov., Isolated from Seawater of the Coast in South Korea

Two novel bacterial strains, 273M-4T and Sam97T, were isolated from seawater in the Yellow Sea, Muan-gun, South Korea, and identified as members of the genus Thalassotalea. Both strains were Gram-stain-negative, aerobic, rod-shaped, non-motile, non-flagellated, and oxidase- and catalase-positive. Phylogenetic analysis based on 16S rRNA gene sequences showed that strains 273M-4T and Sam97T were most closely related to Thalassotalea ponticola KCTC 42155T, with sequence similarities of 97.5% and 98.3%, respectively. Optimal growth for strain 273M-4T occurred at 25-30 °C, pH 7.0, and 2% NaCl, while strain Sam97T grew optimally at 30 °C, pH 8.0, and 2% NaCl. Genome sizes of strains 273M-4T and Sam97T were 3.37 and 3.31 Mb, with DNA G + C contents of 41.0 mol% and 42.9 mol%, respectively. The orthologous average nucleotide identity (OrthoANI) and digital DNA-DNA hybridization (dDDH) values between the two strains were 71.6% and 24.4%, respectively, indicating that they are distinct species. Further genomic analyses of these two strains revealed OrthoANI values of < 73.5% and dDDH values of < 26.7% within the genus Thalassotalea, suggesting their distinctiveness from other Thalassotalea species. The predominate fatty acids of strains 273M-4T and Sam97T were summed feature 3 (consisting of C16:1 ω7c/C16:1 ω6c) and C16:0. All strains contained phosphatidylethanolamine and phosphatidylglycerol as the major polar lipids and ubiquinone-8 (Q-8) as the primary respiratory quinone. Based on phenotypic, phylogenetic, genotypic, and chemotaxonomic data, strains 273M-4T (= KCTC 8644T = LMG 33695T) and Sam97T (= KCTC 8645T = LMG 33694T) represent novel species of the genus Thalassotalea, named Thalassotalea aquiviva sp. nov. and Thalassotalea maritima sp. nov..

Effect of Light Regime on Candidatus Puniceispirillum marinum IMCC1322 in Nutrient-Replete Conditions

Previous studies showed no improvement in bacterial biomass for Candidatus Puniceispirillum marinum IMCC1322 under light regimes. Nevertheless, in nutrient-replete cultures with higher inoculating cell densities, strain IMCC1322 exhibited proteorhodopsin photoheterotrophy. Increasing both inoculum size and the amino acid pool can eliminate quorum sensing and starvation responses in strain IMCC1322. Light regimes affected IMCC1322 cultures in stationary/death phases, where cellular ATP levels ranged from 0.0331 to 1.74 mM, with ATP/cell ranging from 13.9 to 367 zeptomoles. In nutrient-depleted conditions, strain IMCC1322 may suffer from excessive protons generated by proteorhodopsin under light conditions. IMCC1322 may tolerate excessive periplasmic protons through ATP-dependent proton pumping and protonation of augmented amino acids. Meanwhile, acid stress could also be mitigated by refining membrane permeability through unsaturation and cyclopropanation of phospholipids. Oceanic bacteria such as IMCC1322 and SAR11 preferred anaplerotic TCA cycles over glycolysis and rely on the Entner-Doudoroff (ED) pathway for growth. Although ATP generation is less efficient in the ED pathway, it offers advantages during rapid growth owing to its strong thermodynamic driving force. The metabolism of IMCC1322 favors gluconeogenesis over glycolysis, aligning with the metabolism of SAR11 reported in previous studies. However, the additional light-driven, PR-dependent ATP synthesis in IMCC1322 is expected to be insufficient to support protein turnover after the log phase, as well as in nutrient-limited conditions. Stable isotope measurements showed no significant differences in the inorganic carbon assimilation between constant light and constant dark cultures in late log phase.

Microbial communities as indicators of marine ecosystem health: Insights from coastal sediments in the eastern Adriatic Sea

Considering the adaptability and responsiveness of microorganisms to environmental changes, their indicator potential is still not acknowledged in European directives. This comprehensive study examined the changes of microbial communities in sediments and a range of geochemical parameters from pristine and anthropogenically impacted coastal areas in the eastern Adriatic Sea. Various analytical methods found evidence of sediment contamination (high toxicity level, enrichments of metals, tributyltin) in certain areas, leading to the categorization of sediments based on the level of anthropogenic disturbance. Prokaryotes were identified as the most promising group of microbes for further research, with specific bacterial families (Rhodobacteraceae, Ectothiorhodospiraceae, Cyclobacteriaceae) and genera (Boseongicola, B2M28, Subgroup 23, Sva0485, Thiogranum) proposed as potential indicators of environmental status. Finally, predictive models were developed to identify key indicator variables for assessing anthropogenic impact in sediments. This research represents an essential step toward incorporating microbial communities into assessments of benthic environmental health.

Genome-resolved metagenomics of Venice Lagoon surface sediment bacteria reveals high biosynthetic potential and metabolic plasticity as successful strategies in an impacted environment

Bacteria living in sediments play essential roles in marine ecosystems and deeper insights into the ecology and biogeochemistry of these largely unexplored organisms can be obtained from 'omics' approaches. Here, we characterized metagenome-assembled-genomes (MAGs) from the surface sediment microbes of the Venice Lagoon (northern Adriatic Sea) in distinct sub-basins exposed to various natural and anthropogenic pressures. MAGs were explored for biodiversity, major marine metabolic processes, anthropogenic activity-related functions, adaptations at the microscale, and biosynthetic gene clusters. Starting from 126 MAGs, a non-redundant dataset of 58 was compiled, the majority of which (35) belonged to (Alpha- and Gamma-) Proteobacteria. Within the broad microbial metabolic repertoire (including C, N, and S metabolisms) the potential to live without oxygen emerged as one of the most important features. Mixotrophy was also found as a successful lifestyle. Cluster analysis showed that different MAGs encoded the same metabolic patterns (e.g., C fixation, sulfate oxidation) thus suggesting metabolic redundancy. Antibiotic and toxic compounds resistance genes were coupled, a condition that could promote the spreading of these genetic traits. MAGs showed a high biosynthetic potential related to antimicrobial and biotechnological classes and to organism defense and interactions as well as adaptive strategies for micronutrient uptake and cellular detoxification. Our results highlighted that bacteria living in an impacted environment, such as the surface sediments of the Venice Lagoon, may benefit from metabolic plasticity as well as from the synthesis of a wide array of secondary metabolites, promoting ecosystem resilience and stability toward environmental pressures.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42995-023-00192-z.