Microbial Diversity Impacts Non-Protein Amino Acid Production in Cyanobacterial Bloom Cultures Collected from Lake Winnipeg.

Lake Winnipeg in Manitoba, Canada is heavily impacted by harmful algal blooms that contain non-protein amino acids (NPAAs) produced by cyanobacteria: N-(2-aminoethyl)glycine (AEG), ß-aminomethyl-L-alanine (BAMA), ß-N-methylamino-L-alanine (BMAA), and 2,4-diaminobutyric acid (DAB). Our objective was to investigate the impact of microbial diversity on NPAA production by cyanobacteria using semi-purified crude cyanobacterial cultures established from field samples collected by the Lake Winnipeg Research Consortium between 2016 and 2021. NPAAs were detected and quantified by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) using validated analytical methods, while Shannon and Simpson alpha diversity scores were determined from 16S rRNA metagenomic sequences. Alpha diversity in isolate cultures was significantly decreased compared to crude cyanobacterial cultures (p < 0.001), indicating successful semi-purification. BMAA and AEG concentrations were higher in crude compared to isolate cultures (p < 0.0001), and AEG concentrations were correlated to the alpha diversity in cultures (r = 0.554; p < 0.0001). BAMA concentrations were increased in isolate cultures (p < 0.05), while DAB concentrations were similar in crude and isolate cultures. These results demonstrate that microbial community complexity impacts NPAA production by cyanobacteria and related organisms.

SCALiR: A Web Application for Automating Absolute Quantification of Mass Spectrometry-Based Metabolomics Data.

Quantitative liquid chromatography-mass spectrometry (LC-MS)-based metabolomics is becoming an important approach for studying complex biological systems but presents several technical challenges that limit its widespread use. Computing metabolite concentrations using standard curves generated from standard mixtures of known concentrations is a labor-intensive process that is often performed manually. Currently, there are few options for open-source software tools that can automatically calculate metabolite concentrations. Herein, we introduce SCALiR (standard curve application for determining linear ranges), a new web-based software tool specifically built for this task, which allows users to automatically transform LC-MS signals into absolute quantitative data (https://www.lewisresearchgroup.org/software). SCALiR uses an algorithm that automatically finds the equation of the line of best fit for each standard curve and uses this equation to calculate compound concentrations from the LC-MS signal. Using a standard mix containing 77 metabolites, we show a close correlation between the concentrations calculated by SCALiR and the expected concentrations of each compound (R2 = 0.99 for a y = x curve fitting). Moreover, we demonstrate that SCALiR reproducibly calculates concentrations of midrange standards across ten analytical batches (average coefficient of variation 0.091). SCALiR can be used to calculate metabolite concentrations either using external calibration curves or by using internal standards to correct for matrix effects. This open-source and vendor agnostic software offers users several advantages in that (1) it requires only 10 s of analysis time to compute concentrations of >75 compounds, (2) it facilitates automation of quantitative workflows, and (3) it performs deterministic evaluations of compound quantification limits. SCALiR therefore provides the metabolomics community with a simple and rapid tool that enables rigorous and reproducible quantitative metabolomics studies.

MINNO: An Open Source Software for Refining Metabolic Networks and Investigating Complex Network Activity Using Empirical Metabolomics Data.

Metabolomics is a powerful tool for uncovering biochemical diversity in a wide range of organisms. Metabolic network modeling is commonly used to frame metabolomics data in the context of a broader biological system. However, network modeling of poorly characterized nonmodel organisms remains challenging due to gene homology mismatches which lead to network architecture errors. To address this, we developed the Metabolic Interactive Nodular Network for Omics (MINNO), a web-based mapping tool that uses empirical metabolomics data to refine metabolic networks. MINNO allows users to create, modify, and interact with metabolic pathway visualizations for thousands of organisms, in both individual and multispecies contexts. Herein, we illustrate the use of MINNO in elucidating the metabolic networks of understudied species, such as those of the Borrelia genus, which cause Lyme and relapsing fever diseases. Using a hybrid genomics-metabolomics modeling approach, we constructed species-specific metabolic networks for threeBorrelia species. Using these empirically refined networks, we were able to metabolically differentiate these species via their nucleotide metabolism, which cannot be predicted from genomic networks. Additionally, using MINNO, we identified 18 missing reactions from the KEGG database, of which nine were supported by the primary literature. These examples illustrate the use of metabolomics for the empirical refining of genetically constructed networks and show how MINNO can be used to study nonmodel organisms.

SCALiR: a web application for automating absolute quantification of mass spectrometry-based metabolomics data.

Metabolomics is an important approach for studying complex biological systems. Quantitative liquid chromatography-mass spectrometry (LC-MS)-based metabolomics is becoming a mainstream strategy but presents several technical challenges that limit its widespread use. Computing metabolite concentrations using standard curves generated from standard mixtures of known concentrations is a labor-intensive process which is often performed manually. Currently, there are few options for open-source software tools that can automatically calculate metabolite concentrations. Herein, we introduce SCALiR (Standard Curve Application for determining Linear Ranges), a new web-based software tool specifically built for this task, which allows users to automatically transform LC-MS signal data into absolute quantitative data (https://www.lewisresearchgroup.org/software). The algorithm used in SCALiR automatically finds the equation of the line of best fit for each standard curve and uses this equation to calculate compound concentrations from their LC-MS signal. Using a standard mix containing 77 metabolites, we found excellent correlation between the concentrations calculated by SCALiR and the expected concentrations of each compound (R2 = 0.99) and that SCALiR reproducibly calculated concentrations of mid-range standards across ten analytical batches (average coefficient of variation 0.091). SCALiR offers users several advantages, including that it (1) is open-source and vendor agnostic; (2) requires only 10 seconds of analysis time to compute concentrations of >75 compounds; (3) facilitates automation of quantitative workflows; and (4) performs deterministic evaluation of compound quantification limits. SCALiR provides the metabolomics community with a simple and rapid tool that enables rigorous and reproducible quantitative metabolomics studies.

Metabolic Interactive Nodular Network for Omics (MINNO): Refining and investigating metabolic networks based on empirical metabolomics data.

Metabolomics is a powerful tool for uncovering biochemical diversity in a wide range of organisms, and metabolic network modeling is commonly used to frame results in the context of a broader homeostatic system. However, network modeling of poorly characterized, non-model organisms remains challenging due to gene homology mismatches. To address this challenge, we developed Metabolic Interactive Nodular Network for Omics (MINNO), a web-based mapping tool that takes in empirical metabolomics data to refine metabolic networks for both model and unusual organisms. MINNO allows users to create and modify interactive metabolic pathway visualizations for thousands of organisms, in both individual and multi-species contexts. Herein, we demonstrate an important application of MINNO in elucidating the metabolic networks of understudied species, such as those of the Borrelia genus, which cause Lyme disease and relapsing fever. Using a hybrid genomics-metabolomics modeling approach, we constructed species-specific metabolic networks for three Borrelia species. Using these empirically refined networks, we were able to metabolically differentiate these genetically similar species via their nucleotide and nicotinate metabolic pathways that cannot be predicted from genomic networks. These examples illustrate the use of metabolomics for the empirical refining of genetically constructed networks and show how MINNO can be used to study non-model organisms.

Microbial containment device: A platform for comprehensive analysis of microbial metabolism without sample preparation.

Metabolomics is a mainstream strategy for investigating microbial metabolism. One emerging application of metabolomics is the systematic quantification of metabolic boundary fluxes - the rates at which metabolites flow into and out of cultured cells. Metabolic boundary fluxes can capture complex metabolic phenotypes in a rapid assay, allow computational models to be built that predict the behavior of cultured organisms, and are an emerging strategy for clinical diagnostics. One advantage of quantifying metabolic boundary fluxes rather than intracellular metabolite levels is that it requires minimal sample processing. Whereas traditional intracellular analyses require a multi-step process involving extraction, centrifugation, and solvent exchange, boundary fluxes can be measured by simply analyzing the soluble components of the culture medium. To further simplify boundary flux analyses, we developed a custom 96-well sampling system-the Microbial Containment Device (MCD)-that allows water-soluble metabolites to diffuse from a microbial culture well into a bacteria-free analytical well via a semi-permeable membrane. The MCD was designed to be compatible with the autosamplers present in commercial liquid chromatography-mass spectrometry systems, allowing metabolic fluxes to be analyzed with minimal sample handling. Herein, we describe the design, evaluation, and performance testing of the MCD relative to traditional culture methods. We illustrate the utility of this platform, by quantifying the unique boundary fluxes of four bacterial species and demonstrate antibiotic-induced perturbations in their metabolic activity. We propose the use of the MCD for enabling single-step metabolomics sample preparation for microbial identification, antimicrobial susceptibility testing, and other metabolic boundary flux applications where traditional sample preparation methods are impractical.

Moving beyond descriptive studies: harnessing metabolomics to elucidate the molecular mechanisms underpinning host-microbiome phenotypes.

Advances in technology and software have radically expanded the scope of metabolomics studies and allow us to monitor a broad transect of central carbon metabolism in routine studies. These increasingly sophisticated tools have shown that many human diseases are modulated by microbial metabolism. Despite this, it remains surprisingly difficult to move beyond these statistical associations and identify the specific molecular mechanisms that link dysbiosis to the progression of human disease. This difficulty stems from both the biological intricacies of host-microbiome dynamics as well as the analytical complexities inherent to microbiome metabolism research. The primary objective of this review is to examine the experimental and computational tools that can provide insights into the molecular mechanisms at work in host-microbiome interactions and to highlight the undeveloped frontiers that are currently holding back microbiome research from fully leveraging the benefits of modern metabolomics.

Methods for Quantifying the Metabolic Boundary Fluxes of Cell Cultures in Large Cohorts by High-Resolution Hydrophilic Liquid Chromatography Mass Spectrometry.

Metabolomics is a mainstream approach for investigating the metabolic underpinnings of complex biological phenomena and is increasingly being applied to large-scale studies involving hundreds or thousands of samples. Although metabolomics methods are robust in smaller-scale studies, they can be challenging to apply to larger cohorts due to the inherent variability of liquid chromatography mass spectrometry (LC-MS). Much of this difficulty results from the time-dependent changes in the LC-MS system, which affects both the qualitative and quantitative performances of the instrument. Herein, we introduce an analytical strategy for addressing this problem in large-scale microbial studies. Our approach quantifies microbial boundary fluxes using two zwitterionic hydrophilic interaction liquid chromatography (ZIC-HILIC) columns that are plumbed to enable offline column equilibration. Using this strategy, we show that over 397 common metabolites can be resolved in 4.5 min per sample and that metabolites can be quantified with a median coefficient of variation of 0.127 across 1100 technical replicates. We illustrate the utility of this strategy via an analysis of 960 strains of Staphylococcus aureus isolated from bloodstream infections. These data capture the diversity of metabolic phenotypes observed in clinical isolates and provide an example of how large-scale investigations can leverage our novel analytical strategy.

Metabolomics-Guided Hypothesis Generation for Mechanisms of Intestinal Protection by Live Biotherapeutic Products.

The use of live biotherapeutic products (LBPs), including single strains of beneficial probiotic bacteria or consortiums, is gaining traction as a viable option to treat inflammatory-mediated diseases like inflammatory bowel disease (IBD). However, LBPs' persistence in the intestine is heterogeneous since many beneficial bacteria lack mechanisms to tolerate the inflammation and the oxidative stress associated with IBD. We rationalized that optimizing LBPs with enhanced colonization and persistence in the inflamed intestine would help beneficial bacteria increase their bioavailability and sustain their beneficial responses. Our lab developed two bioengineered LBPs (SBT001/BioPersist and SBT002/BioColoniz) modified to enhance colonization or persistence in the inflamed intestine. In this study, we examined colon-derived metabolites via ultra-high performance liquid chromatography-mass spectrometry in colitic mice treated with either BioPersist or BioColoniz as compared to their unmodified parent strains (Escherichia coli Nissle 1917 [EcN] and Lactobacillus reuteri, respectively) or to each other. BioPersist administration resulted in lowered concentrations of inflammatory prostaglandins, decreased stress hormones such as adrenaline and corticosterone, increased serotonin, and decreased bile acid in comparison to EcN. In comparison to BioColoniz, BioPersist increased serotonin and antioxidant production, limited bile acid accumulation, and enhanced tissue restoration via activated purine and pyrimidine metabolism. These data generated several novel hypotheses for the beneficial roles that LBPs may play during colitis.

A Single Laboratory Validation for the Analysis of Underivatized ß-N-Methylamino-L-Alanine (BMAA).

ß-N-Methylamino-L-alanine (BMAA) is a non-protein amino acid produced by cyanobacteria that can accumulate in ecosystems and food webs. Human exposure to cyanobacterial and algal blooms may be a risk factor for neurodegenerative diseases such as Alzheimer's disease and amyotrophic lateral sclerosis. Analytical chemists have struggled to find reliable methods for BMAA analysis in complex sample matrices. Analysis of BMAA is complicated by at least 3 naturally occurring isomers: N-(2-aminoethyl)glycine (AEG), 2,4-diaminobutyric acid (DAB), and ß-aminomethyl-L-alanine (BAMA). More than 350 publications have reported detection and quantification of BMAA and its isomers, but varying results have led to controversy in the literature. The objective of this study was to perform a single laboratory validation (SLV) of a frequently published method for BMAA analysis using a ZIC-HILIC column. We investigated the selectivity, linearity, accuracy, precision, and sensitivity of the method and our data show that this HILIC method fails many of the criteria for a validated method. The method fails the criterion for selectivity as the chromatography does not separate BMAA from its isomer BAMA. Sensitivity of the method greatly decreased over the experimental period and it demonstrated a higher limit of detection (LOD) (7.5 pg on column) and a higher lower limit of quantification (LLOQ) (30 pg on column) than other published validated methods. The method demonstrated poor precision of repeated injections of standards of BMAA with % relative standard deviation (%RSD) values that ranged from 37 to 107% while HorRat values for BMAA had a fail rate of 80% and BAMA had a fail rate of 73%. No HorRat values between 0.5 and 2 were found for repeated injections of standards of AEG and DAB. Recovery of 13C3,15N2-BMAA in a cyanobacterial matrix was < 10% in experiments and we were also unable to accurately detect other protein amino acids including methionine, cysteine, or alanine, indicating matrix effects. The results of this study demonstrate that the ZIC-HILIC column is not fit for purpose for the analysis of BMAA in cyanobacterial matrices and further provides explanations for the high level of negative results reported by researchers using this method.

A systematic review of analytical methods for the detection and quantification of ß-N-methylamino-l-alanine (BMAA).

Neurodegenerative diseases are influenced by environmental factors such as exposure to toxins including the cyanotoxin ß-N-methylamino-l-alanine (BMAA) that can bioaccumulate in common food sources such as fish, mussels and crabs. Accurate and precise analytical methods are needed to detect and quantify BMAA to minimize human health risks. The objective of this review is to provide a comprehensive overview of the methods used for BMAA analysis from 2003 to 2019 and to evaluate the reported performance characteristics for each method to determine the consensus data for each analytical approach and different sample matrices. Detailed searches of the database Web of Science™ (WoS) were performed between August 21st, 2018 and April 5th, 2019. Eligible studies included analytical methods for the detection and quantification of BMAA in cyanobacteria and bioaccumulated BMAA in higher trophic levels, in phytoplankton and zooplankton and in human tissues and fluids. This systematic review has limitations in that only the English language literature is included and it did not include standard operating protocols nor any method validation data that have not been made public. We identified 148 eligible studies, of which a positive result for BMAA in one or more samples analyzed was reported in 84% (125 out of 148) of total studies, 57% of HILIC studies, 92% of RPLC studies and 71% of other studies. The largest discrepancy between different methods arose from the analysis of cyanobacteria samples, where BMAA was detected in 95% of RPLC studies but only in 25% of HILIC studies. Without sufficient published validation of each method's performance characteristics, it is difficult to establish each method as fit for purpose for each sample matrix. The importance of establishing methods as appropriate for their intended use is evidenced by the inconsistent reporting of BMAA across environmental samples, despite its prevalence in diverse ecosystems and food webs.

N-ß-Methylamino-L-Alanine and Its Naturally Occurring Isomers in Cyanobacterial Blooms in Lake Winnipeg.

Cyanobacterial blooms have affected Lake Winnipeg since the mid-1990s due to an increased phosphorus loading into the lake, which has been exacerbated by stressors such as climate change and eutrophication. Aquatic ecosystems involving cyanobacteria have been found to contain N-ß-methylamino-L-alanine (BMAA) and 2,4-diaminobutyric acid (DAB), non-protein amino acids that are associated with neurodegenerative disease, as well as two of the naturally occurring isomers, N-2(amino)ethylglycine (AEG) and ß-amino-N-methylalanine (BAMA). We hypothesized that the cyanobacterial bloom in Lake Winnipeg produces BMAA and/or its naturally occurring isomers. Samples of cyanobacteria were collected by the Lake Winnipeg Research Consortium from standard sampling stations and blooms in July and September of 2016 and were analyzed for BMAA, DAB, AEG, and BAMA using previously published validated analytical methods. BMAA and BAMA were found in the highest concentration in the center of the north basin, the deepest and lowest-nitrogen zone of the lake, at an average concentration of 4 µg/g (collected in July and September 2016) and 1.5 mg/g (collected in July 2016), respectively. AEG and DAB were found in the highest concentration in cyanobacterial blooms from the nearshore region of the north basin, the slightly shallower and more nitrogen-rich zone of the lake, at 2.1 mg/g (collected in July 2016) and 0.2 mg/g (collected in July and September 2016), respectively. These findings indicate that the production of non-protein amino acids varies with the depth and nutrient contents of the bloom. It is important to note that we did not measure food or water samples directly and further study of the Lake Winnipeg food web is required to determine whether BMAA bioaccumulation represents an increased risk factor for neurodegenerative disease in the region.