Maria W. Smith

Contrasting genomic properties of free-living and particle-attached microbial assemblages within a coastal ecosystem

The Columbia River (CR) is a powerful economic and environmental driver in the US Pacific Northwest. Microbial communities in the water column were analyzed from four diverse habitats: (1) an estuarine turbidity maximum (ETM), (2) a chlorophyll maximum of the river plume, (3) an upwelling-associated hypoxic zone, and (4) the deep ocean bottom. Three size fractions, 0.1–0.8, 0.8–3, and 3–200 μm were collected for each habitat in August 2007, and used for DNA isolation and 454 sequencing, resulting in 12 metagenomes of >5 million reads (>1.6 Gbp). To characterize the dominant microorganisms and metabolisms contributing to coastal biogeochemistry, we used predicted peptide and rRNA data. The 3- and 0.8-μm metagenomes, representing particulate fractions, were taxonomically diverse across habitats. The 3-μm size fractions contained a high abundance of eukaryota with diatoms dominating the hypoxic water and plume, while cryptophytes were more abundant in the ETM. The 0.1-μm metagenomes represented mainly free-living bacteria and archaea. The most abundant archaeal hits were observed in the deep ocean and hypoxic water (19% of prokaryotic peptides in the 0.1-μm metagenomes), and were homologous to Nitrosopumilus maritimus (ammonia-oxidizing Thaumarchaeota). Bacteria dominated metagenomes of all samples. In the euphotic zone (estuary, plume and hypoxic ocean), the most abundant bacterial taxa (≥40% of prokaryotic peptides) represented aerobic photoheterotrophs. In contrast, the low-oxygen, deep water metagenome was enriched with sequences for strict and facultative anaerobes. Interestingly, many of the same anaerobic bacterial families were enriched in the 3-μm size fraction of the ETM (2–10X more abundant relative to the 0.1-μm metagenome), indicating possible formation of anoxic microniches within particles. Results from this study provide a metagenome perspective on ecosystem-scale metabolism in an upwelling-influenced river-dominated coastal margin.

Metagenomic insights into particles and their associated microbiota in a coastal margin ecosystem

Our previously published research was one of the pioneering studies on the use of metagenomics to directly compare taxonomic and metabolic properties of aquatic microorganisms from different filter size-fractions. We compared size-fractionated water samples representing free-living and particle-attached communities from four diverse habitats in the Columbia River coastal margin, analyzing 12 metagenomes consisting of >5 million sequence reads (>1.6 Gbp). With predicted peptide and rRNA data we evaluated eukaryotic, bacterial and archaeal populations across size fractions and related their properties to attached and free-living lifestyles, and their potential roles in carbon and nutrient cycling. In this focused review, we expand our discussion on the use of high-throughput sequence data to relate microbial community structure and function to the origin, fate and transport of particulate organic matter (POM) in coastal margins. We additionally discuss the potential impact of the priming effect on organic matter cycling at the land-ocean interface, and build a case for the importance, in particle-rich estuaries and coastal margin waters, of microbial activities in low-oxygen microzones within particle interiors.

Seasonal changes in bacterial and archaeal gene expression patterns across salinity gradients in the Columbia River coastal margin

Through their metabolic activities, microbial populations mediate the impact of high gradient regions on ecological function and productivity of the highly dynamic Columbia River coastal margin (CRCM). A 2226-probe oligonucleotide DNA microarray was developed to investigate expression patterns for microbial genes involved in nitrogen and carbon metabolism in the CRCM. Initial experiments with the environmental microarrays were directed toward validation of the platform and yielded high reproducibility in multiple tests. Bioinformatic and experimental validation also indicated that >85% of the microarray probes were specific for their corresponding target genes and for a few homologs within the same microbial family. The validated probe set was used to query gene expression responses by microbial assemblages to environmental variability. Sixty-four samples from the river, estuary, plume, and adjacent ocean were collected in different seasons and analyzed to correlate the measured variability in chemical, physical and biological water parameters to differences in global gene expression profiles. The method produced robust seasonal profiles corresponding to pre-freshet spring (April) and late summer (August). Overall relative gene expression was high in both seasons and was consistent with high microbial abundance measured by total RNA, heterotrophic bacterial production, and chlorophyll a. Both seasonal patterns involved large numbers of genes that were highly expressed relative to background, yet each produced very different gene expression profiles. April patterns revealed high differential gene expression in the coastal margin samples (estuary, plume and adjacent ocean) relative to freshwater, while little differential gene expression was observed along the river-to-ocean transition in August. Microbial gene expression profiles appeared to relate, in part, to seasonal differences in nutrient availability and potential resource competition. Furthermore, our results suggest that highly-active particle-attached microbiota in the Columbia River water column may perform dissimilatory nitrate reduction (both dentrification and DNRA) within anoxic particle microniches.

Genomic and phenotypic differentiation among Methanosarcina mazei populations from Columbia River sediment

Methanogenic archaea are genotypically and phenotypically diverse organisms that are integral to carbon cycling in anaerobic environments. Owing to their genetic tractability and ability to be readily cultivated, Methanosarcina spp. have become a powerful model system for understanding methanogen biology at the cellular systems level. However, relatively little is known of how genotypic and phenotypic variation is partitioned in Methanosarcina populations inhabiting natural environments and the possible ecological and evolutionary implications of such variation. Here, we have identified how genomic and phenotypic diversity is partitioned within and between Methanosarcina mazei populations obtained from two different sediment environments in the Columbia River Estuary (Oregon, USA). Population genomic analysis of 56 M. mazei isolates averaging <1% nucleotide divergence revealed two distinct clades, which we refer to as ‘mazei-T’ and ‘mazei-WC’. Genomic analyses showed that these clades differed in gene content and fixation of allelic variants, which point to potential differences in primary metabolism and also interactions with foreign genetic elements. This hypothesis of niche partitioning was supported by laboratory growth experiments that revealed significant differences in trimethylamine utilization. These findings improve our understanding of the ecologically relevant scales of genomic variation in natural systems and demonstrate interactions between genetic and ecological diversity in these easily cultivable and genetically tractable model methanogens.

Development of real-time assays for impedance-based detection of microbial double-stranded DNA targets: Optimization and data analysis

A real-time, label free assay was developed for microbial detection, utilizing double-stranded DNA targets and employing the next generation of an impedimetric sensor array platform designed by Sharp Laboratories of America (SLA). Real-time curves of the impedimetric signal response were obtained at fixed frequency and voltage for target binding to oligonucleotide probes attached to the sensor array surface. Kinetic parameters of these curves were analyzed by the integrated data analysis package for signal quantification. Non-specific binding presented a major challenge for assay development, and required assay optimization. For this, differences were maximized between binding curve kinetic parameters for probes binding to complementary targets versus non-target controls. Variables manipulated for assay optimization included target concentration, hybridization temperature, buffer concentration, and the use of surfactants. Our results showed that (i) different target-probe combinations required optimization of specific sets of variables; (ii) for each assay condition, the optimum range was relatively narrow, and had to be determined empirically; and (iii) outside of the optimum range, the assay could not distinguish between specific and non-specific binding. For each target-probe combination evaluated, conditions resulting in good separation between specific and non-specific binding signals were established, generating high confidence in the SLA impedimetric dsDNA assay results.

Metagenomic evidence for reciprocal particle exchange between the mainstem estuary and lateral bay sediments of the lower Columbia River

Lateral bays of the lower Columbia River estuary are areas of enhanced water retention that influence net ecosystem metabolism through activities of their diverse microbial communities. Metagenomic characterization of sediment microbiota from three disparate sites in two brackish lateral bays (Baker and Youngs) produced ∼100 Gbp of DNA sequence data analyzed subsequently for predicted SSU rRNA and peptide-coding genes. The metagenomes were dominated by Bacteria. A large component of Eukaryota was present in Youngs Bay samples, i.e., the inner bay sediment was enriched with the invasive New Zealand mudsnail, Potamopyrgus antipodarum, known for high ammonia production. The metagenome was also highly enriched with an archaeal ammonia oxidizer closely related to Nitrosoarchaeum limnia. Combined analysis of sequences and continuous, high-resolution time series of biogeochemical data from fixed and mobile platforms revealed the importance of large-scale reciprocal particle exchanges between the mainstem estuarine water column and lateral bay sediments. Deposition of marine diatom particles in sediments near Youngs Bay mouth was associated with a dramatic enrichment of Bacteroidetes (58% of total Bacteria) and corresponding genes involved in phytoplankton polysaccharide degradation. The Baker Bay sediment metagenome contained abundant Archaea, including diverse methanogens, as well as functional genes for methylotrophy and taxonomic markers for syntrophic bacteria, suggesting that active methane cycling occurs at this location. Our previous work showed enrichments of similar anaerobic taxa in particulate matter of the mainstem estuarine water column. In total, our results identify the lateral bays as both sources and sinks of biogenic particles significantly impacting microbial community composition and biogeochemical activities in the estuary.

Real-Time Biosensor Platform: Fully Integrated Device for Impedimetric Assays

An impedimetric biosensor platform for bioaffinity assays has been developed that is based on real-time, label-free electrochemical detection performed via a direct interface to electronic digital data processing. The sensor array consists of 15 gold microelectrode pairs (Fig. 1) that are enclosed in three reaction chambers and biofunctionalized with specific probes. The impedance change caused by specific capture of target analyte molecules on the functionalized electrode surface is recorded in real time. The measuring instrument is capable of continuous and simultaneous stimulation and recording of all electrodes on the array. A corresponding mathematical algorithm and a software package for data analysis have been developed. The software performs (i) filtering of the instrument noise, and (ii) extraction of the exponential component of the impedance signal. Thus, the algorithm can quantify both rate of target to probe binding, and target to probe affinity. The described fully integrated platform can be used as a basic research tool for development of various bio-affinity impedimetric assays. To facilitate such applications, we have developed a streamlined manufacturing technology, and a set of assay protocols for detection of microbes based on nucleic acid hybridization. The assay was shown to detect and distinguish between two closely related but different Escherichia coli strains. The assay sensitivity was sufficient for reliable measurements of specific PCR products amplified from microbial genomic DNA. The sensor array platform is adaptable for detection of a wide range of analytes of practical significance, and it has potential for further integration with amplification (i.e. PCR) and sample preparation modules.

Microbial players and processes involved in phytoplankton bloom utilization in the water column of a fast-flowing, river-dominated estuary

Fueled by seasonal phytoplankton blooms, the Columbia River estuary is a natural bioreactor for organic matter transformations. Prior metagenome analyses indicated high abundances of diverse Bacteroidetes taxa in estuarine samples containing phytoplankton. To examine the hypothesis that Bacteroidetes taxa have important roles in phytoplankton turnover, we further analyzed metagenomes from water collected along a salinity gradient at 0, 5, 15, 25, and 33 PSU during bloom events. Size fractions were obtained by using a 3‐μm prefilter and 0.2‐μm collection filter. Although this approach targeted bacteria by removing comparatively large eukaryotic cells, the metagenome from the ES‐5 sample (5 PSU) nevertheless contained an abundance of diatom DNA. Biogeochemical measurements and prior studies indicated that this finding resulted from the leakage of cellular material due to freshwater diatom lysis at low salinity. Relative to the other metagenomes, the bacterial fraction of ES‐5 was dramatically depleted of genes annotated as Bacteroidetes and lysogenic bacteriophages, but was overrepresented in DNA of protists and Myxococcales bacterivores. We suggest the following equally plausible scenarios for the microbial response to phytoplankton lysis: (1) Bacteroidetes depletion in the free‐living fraction may at least in part be caused by their attachment to fluvial diatoms as the latter are lysed upon contact with low‐salinity estuarine waters; (2) diatom particle colonization is likely followed by rapid bacterial growth and lytic phage infection, resulting in depletion of lysogenic bacteriophages and host bacteria; and (3) the subsequent availability of labile organic matter attracted both grazers and predators to feed in this estuarine biogeochemical “hotspot,” which may have additionally depleted Bacteroidetes populations. These results represent the first detailed molecular analysis of the microbial response to phytoplankton lysis at the freshwater–brackish water interface in the fast‐flowing Columbia River estuary.

Restriction Cascade Exponential Amplification (RCEA) assay with an attomolar detection limit: a novel, highly specific, isothermal alternative to qPCR

An alternative to qPCR was developed for nucleic acid assays, involving signal rather than target amplification. The new technology, Restriction Cascade Exponential Amplification (RCEA), relies on specific cleavage of probe-target hybrids by restriction endonucleases (REase). Two mutant REases for amplification (Ramp), S17C BamHI and K249C EcoRI, were conjugated to oligonucleotides, and immobilized on a solid surface. The signal generation was based on: (i) hybridization of a target DNA to a Ramp-oligonucleotide probe conjugate, followed by (ii) specific cleavage of the probe-target hybrid using a non-immobilized recognition REase. The amount of Ramp released into solution upon cleavage was proportionate to the DNA target amount. Signal amplification was achieved through catalysis, by the free Ramp, of a restriction cascade containing additional oligonucleotide-conjugated Ramp and horseradish peroxidase (HRP). Colorimetric quantification of free HRP indicated that the RCEA achieved a detection limit of 10 aM (10−17 M) target concentration, or approximately 200 molecules, comparable to the sensitivity of qPCR-based assays. The RCEA assay had high specificity, it was insensitive to non-specific binding, and detected target sequences in the presence of foreign DNA. RCEA is an inexpensive isothermal assay that allows coupling of the restriction cascade signal amplification with any DNA target of interest.

A New Restriction Endonuclease-Based Method for Highly-Specific Detection of DNA Targets from Methicillin-Resistant Staphylococcus aureus

PCR multiplexing has proven to be challenging, and thus has provided limited means for pathogen genotyping. We developed a new approach for analysis of PCR amplicons based on restriction endonuclease digestion. The first stage of the restriction enzyme assay is hybridization of a target DNA to immobilized complementary oligonucleotide probes that carry a molecular marker, horseradish peroxidase (HRP). At the second stage, a target-specific restriction enzyme is added, cleaving the target-probe duplex at the corresponding restriction site and releasing the HRP marker into solution, where it is quantified colorimetrically. The assay was tested for detection of the methicillin-resistant Staphylococcus aureus (MRSA) pathogen, using the mecA gene as a target. Calibration curves indicated that the limit of detection for both target oligonucleotide and PCR amplicon was approximately 1 nM. Sequences of target oligonucleotides were altered to demonstrate that (i) any mutation of the restriction site reduced the signal to zero; (ii) double and triple point mutations of sequences flanking the restriction site reduced restriction to 50–80% of the positive control; and (iii) a minimum of a 16-bp target-probe dsDNA hybrid was required for significant cleavage. Further experiments showed that the assay could detect the mecA amplicon from an unpurified PCR mixture with detection limits similar to those with standard fluorescence-based qPCR. Furthermore, addition of a large excess of heterologous genomic DNA did not affect amplicon detection. Specificity of the assay is very high because it involves two biorecognition steps. The proposed assay is low-cost and can be completed in less than 1 hour. Thus, we have demonstrated an efficient new approach for pathogen detection and amplicon genotyping in conjunction with various end-point and qPCR applications. The restriction enzyme assay may also be used for parallel analysis of multiple different amplicons from the same unpurified mixture in broad-range PCR applications.