Biology subjects

Gordon, V.

Publications and source records attributed to Gordon, V..

8 recordsLinked to original sources

bioRxiv2025

Predicting characterization of microbiome taxonomy from imaging using machine learning approaches

AO_SCPLOWBSTRACTC_SCPLOWFor this study, a total of 47 mock human skin microbiome communities were created using microorganisms collected from human donors and grown in vitro for between eight and 32 days. Each mock community sample was split. Ten mL of each sample was used to determine the taxonomy of the community, using metatranscriptomics and Kraken2 to provide population-level taxonomic information; five mL of each sample was used for imaging. The resulting micrographs served as the basis for establishing a new analysis pipeline that sequentially used two different methods for machine learning and one statistical technique: (1) confocal microscopy images were segmented into individual cells using the generalist, deep learning, publicly available machine learning model Cellpose; (2) continuous probability density functions describing the joint distribution of the cell area and eccentricity were found using algorithms expressing the statistical technique of kernel density estimation; (3) these probability density functions were used as input for convolutional neural networks, that were trained to predict both the taxonomic diversity and the most common bacterial class, independently of metatranscriptomics. Specifically, models were made to predict the Shannon index (a quantitative measure of taxonomic diversity) and to predict the most common bacterial class, for each micrograph. Measured Shannon indices (based on metatranscriptomics) ranged from nearly 0 to 1.4. The model predictions of Shannon indices had a mean squared error of 0.0321 +/- 0.0035. The model predictions of the most common taxonomic class of bacteria had an accuracy of 94.0% +/- 0.7%. IO_SCPLOWMPORTANCEC_SCPLOWTaxonomic diversity is a useful metric for describing microbial communities and can be used as a measure of ecosystems health, resilience, and biological interactions. Characterization of microbial community diversity also has diagnostic applications. For the human skin microbiome in particular, microbial diversity directly impacts skin health, including resilience against pathogens and regulation of immune responses. Currently, microbial diversity can be determined either using traditional staining methods that are limited to pure cultures or using sequencing methods that require high investment in cost, time, and expertise. In this study, we demonstrate an innovative method that employs microscopy images of bacterial communities and machine learning to predict taxonomic diversity and the dominant bacterial classes of bacterial communities. The underlying framework of the pipeline for taxonomy prediction has the potential to be adapted and extended to other organisms and microbiomes and to make taxonomic analyses less expensive and more feasible in low-resource settings.

microbiology↗

bioRxiv2024

Geometric changes in the nucleoids of Deinococcus radiodurans reveal involvement of new proteins in recovery from ionizing radiation.

The extremophile Deinococcus radiodurans maintains a highly-organized and condensed nucleoid as its default state, possibly contributing to high tolerance of ionizing radiation (IR). Previous studies of the D. radiodurans nucleoid were limited by reliance on manual image annotation and qualitative metrics. Here, we introduce a high-throughput approach to quantify the geometric properties of cells and nucleoids, using confocal microscopy, digital reconstructions of cells, and computational modeling. We utilize this novel approach to investigate the dynamic process of nucleoid condensation in response to IR stress. Our quantitative analysis reveals that at the population level, exposure to IR induced nucleoid compaction and decreased size of D. radiodurans cells. Morphological analysis and clustering identified six distinct sub-populations across all tested experimental conditions. Results indicate that exposure to IR induces fractional redistributions of cells across sub-populations to exhibit morphologies that associate with greater nucleoid condensation, and decreased abundance of sub-populations associated with cell division. Nucleoid associated proteins (NAPs) may link nucleoid compaction and stress tolerance, but their roles in regulating compaction in D. radiodurans is unknown. Imaging of genomic mutants of known and suspected NAPs that contribute to nucleoid condensation found that deletion of nucleic acid binding proteins, not previously described as NAPs, can remodel the nucleoid by driving condensation or decondensation in the absence of stress and that IR increases the abundance of these morphological states. Thus, our integrated analysis introduces a new methodology for studying environmental influences on bacterial nucleoids and provides an opportunity to further investigate potential regulators of nucleoid condensation. ImportanceD. radiodurans, an extremophile known for its stress tolerance, constitutively maintains a highly-condensed nucleoid. Qualitative studies have described nucleoid behavior under a variety of conditions. However, a lack of quantitative data regarding nucleoid organization and dynamics have limited our understanding of regulatory mechanisms controlling nucleoid organization in D. radiodurans. Here, we introduce a quantitative approach that enables high-throughput quantitative measurements of subcellular spatial characteristics in bacterial cells. Applying this to wild-type or single-protein-deficient populations of D. radiodurans subjected to ionizing radiation, we identified significant stress-responsive changes in cell shape, nucleoid organization, and morphology. These findings highlight this methodologys adaptability and capacity for quantitatively analyzing the cellular response to stressors for screening cellular proteins involved in bacterial nucleoid organization.

molecular biology↗

bioRxiv2023

Bacterial mechanosensing of surface stiffness promotes signaling and growth leading to biofilm formation by Pseudomonas aeruginosa

The attachment of bacteria onto a surface, consequent signaling, and the accumulation and growth of the surface-bound bacterial population are key initial steps in the formation of pathogenic biofilms. While recent reports have hinted that the stiffness of a surface may affect the accumulation of bacteria on that surface, the processes that underlie bacterial perception of and response to surface stiffness are unknown. Furthermore, whether, and how, the surface stiffness impacts biofilm development, after initial accumulation, is not known. We use thin and thick hydrogels to create stiff and soft composite materials, respectively, with the same surface chemistry. Using quantitative microscopy, we find that the accumulation, motility, and growth of the opportunistic human pathogen Pseudomonas aeruginosa respond to surface stiffness, and that these are linked through cyclic-di-GMP signaling that depends on surface stiffness. The mechanical cue stemming from surface stiffness is elucidated using finite-element modeling combined with experiments - adhesion to stiffer surfaces results in greater changes in mechanical stress and strain in the bacterial envelope than does adhesion to softer surfaces with identical surface chemistry. The cell-surface-exposed protein PilY1 acts as a mechanosensor, that upon surface engagement, results in higher cyclic-di-GMP levels, lower motility, and greater accumulation on stiffer surfaces. PilY1 impacts the biofilm lag phase, which is extended for bacteria attaching to stiffer surfaces. This study shows clear evidence that bacteria actively respond to different stiffness of surfaces where they adhere via perceiving varied mechanical stress and strain upon surface engagement. ImportanceBacteria colonize many types of biological and medical surfaces with a large range of stiffnesses. Colonization leads to the formation of biofilms, which cause costly and life-impairing chronic infections. However, whether and how bacteria can sense and respond to the mechanical cue provided by surface stiffness has remained unknown. We find that bacteria do indeed respond to surface stiffness in a way that is both consistent with expectations based on equilibrium continuum mechanics and that quantitatively impacts multiple aspects of early biofilm formation. This is a new understanding for the nascent field of bacterial mechanobiology. Furthermore, this finding suggests the possibility of a new category of approaches to hindering biofilm development by tuning the mechanical properties of biomedical surfaces.

microbiology↗

bioRxiv2019

Lfc/Arhgef2 regulates mitotic spindle orientation in hematopoietic stem and progenitor cells and is essential for productive hematopoiesis

How hematopoietic stem cells (HSCs) coordinate their divisional axis relative to supportive niche cells and whether or not their divisional orientation is important for stem cell-driven hematopoiesis is poorly understood. Single cell RNA sequencing data from patients with the inherited bone marrow failure Shwachman-Diamond syndrome (SDS) show that ARHGEF2, a RhoA-specific guanine nucleotide exchange factor (GEF) and determinant of mitotic spindle orientation, is one of a restricted group of genes specifically downregulated in SDS HSCs and multipotent progenitors. Here, we describe Lfc/Arhgef2 as an important regulator of hematopoiesis in vivo. Transplanted Lfc/Arhgef2-/- bone marrow shows impaired hematopoietic recovery and a production deficit of long-term HSCs. These phenotypes cannot be explained by differences in numbers of transplanted HSCs, their cell cycle status, level of apoptosis, progenitor output or homing ability. Using live imaging of dividing hematopoietic stem and progenitor cells (HSPCs), we show an increased frequency of misoriented divisions in the absence of Lfc/Arhgef2. Functional ARHGEF2 knockdown in human HSCs also impairs their ability to regenerate hematopoiesis, culminating in significantly smaller hematopoietic xenografts. Together, these data provide evidence demonstrating a conserved role for Lfc/Arhgef2 in orienting HSPC division and suggest that HSCs divide in certain orientations to establish hematopoiesis, the loss of which leads to their exhaustion in a mechanism that may underlie certain bone marrow failure syndromes.

cell biology↗

bioRxiv2019

Specific Disruption of Established P. aeruginosa Biofilms Using Polymer-Attacking Enzymes

Biofilms are communities of bacteria embedded in an extracellular matrix of self-produced polymeric substances. This polymer matrix lends the bacteria protection against a wide array of chemical and mechanical stresses that they may experience in their environment, which might be a location in the human body in the case of a biofilm infection, or a surface immersed in fluid in an industrial setting. Breaking down the matrix network renders biofilms more susceptible to physical disruption and to treatments. Different species of bacteria, and different strains within the same species, produce different types of matrix polymers - this suggests that targeting specific polymers for disruption may be more effective than non-specific approaches to disrupting biofilm matrices. In this study, we treated Pseudomonas aeruginosa biofilms with enzymes that are specific to different matrix polymers. We used bulk rheology to measure the resulting alteration in biofilm mechanics, and scanning electron microscopy to visualize the alteration in the matrix network upon treatment. Different lab strains of P. aeruginosa form biofilms that can be dominated by one of three main extracellular polysaccharides: Psl, alginate, and Pel, which binds electrostatically to extracellular DNA in the matrix. We applied enzymes to biofilms dominated by different extracellular polysaccharides and found that, for biofilms grown in vitro, the effect of enzymatic treatment is maximized when the enzyme is specific to a dominant matrix polymer - for such a case, specifically-matched enzymatic treatment tends to: reduce yield strain and yield stress; reduce or eliminate long-range structure and shorten or eliminate connecting network fibers in the biofilm as seen under scanning electron microscopy; and increase the rate of biofilm drying, most likely due to increased diffusivity as a result of network compromise. However, for ex vivo biofilms grown in murine wounds, we find that generic glycoside hydrolases have more profound disruptive effects than specifically-matched enzymes, even though they had no measurable effect for biofilms grown in vitro. This highlights the importance of the environment in which the biofilms are grown, the need to take this into account when developing treatments for biofilms, and the possibility that effective approaches to eradicating biofilms in environmental or industrial settings may need to be very different from effective treatments of infection.

microbiology↗

bioRxiv2019

Assaying how the success of phagocytosis depends on the mechanics of a large viscoelastic target

The state of the art does not provide a method for determining how the success of phagocytosis depends on the mechanics of a target that is much larger than the phagocytosing cell. We have developed such a method. We vary the elastic moduli of millimeter-sized abiotic gels that contain fluorescent beads to act as tracers for phagocytosis. We isolate human neutrophils, expose them to gels for one hour, and then measure what percentage of neutrophils contain beads - this is our metric for successful phagocytosis. Both increased polymer concentration in agarose gels and increased crosslinking density in alginate gels are associated with decreased success of phagocytosis. When we plot the percentage of neutrophils containing beads as a function of the gel elastic modulus, we find that data from both alginate and agarose gels collapse onto the same curve. This demonstrates the utility of our method as a way of measuring how the viscoelastic mechanics of a large target impact the success of phagocytosis.

biophysics↗

bioRxiv2018

Endogenous BioID elucidates TCF7L1 interactome modulation upon GSK-3 inhibition in mouse ESCs

Modulation of Wnt target gene expression via the TCF/LEFs remains poorly understood. We employ proximity-based biotin labeling (BioID) to examine GSK-3 inhibitor effects on the TCF7L1 interactome in mouse ESCs. We generated ESC lines with biotin ligase BirA* fused to TCF7L1 by knocking it into the endogenous TCF7L1 locus or by inserting a doxinducible BirA*-TCF7L1 transgene into the Rosa26 locus. Induction yielded BirA*-TCF7L1 levels 3-fold higher than in the endogenous system, but substantial overlap in biotinylated proteins with high peptide counts were detected by each method. Known TCF7L1 interactors TLE3/4 and {beta}-catenin, and numerous proteins not previously associated with TCF7L1, were identified in both systems. Despite reduced BirA*-TCF7L1 levels, the number of hits identified with both BioID approaches increased after GSK-3 inhibition. We elucidate the network of TCF7L1 proximal proteins regulated by GSK-3 inhibition, validate the utility of endogenous BioID, and provide mechanistic insights into TCF7L1 target gene regulation.\n\nHighlightsO_LIBirA*-TCF7L1 at single-copy physiological levels generates robust BioID data\nC_LIO_LICHIR99021 reduces TCF7L1 levels but increases detectable TCF7L1-proximal proteins\nC_LIO_LIThe TCF7L1 interactome of largely epigenetic/transcription factors fluctuates with GSK-3 inhibition\nC_LIO_LIJMJD1C, SALL4 and BRG1/SMARCA4 are validated as TCF7L-interacting proteins\nC_LI

cell biology↗

bioRxiv2018

Temporal and spatial factors that influence magnetotaxis in C. elegans

Many animals can orient using the earths magnetic field. In a recent study, we performed three distinct behavioral assays providing evidence that the nematode Caenorhabditis elegans orients to earth-strength magnetic fields (Vidal-Gadea et al., 2015). In addition to these behavioral assays, we found that magnetic orientation in C. elegans depends on the AFD sensory neurons and conducted subsequent physiological experiments showing that AFD neurons respond to earth-strength magnetic fields. A new behavioral study by Landler et al. (2017) suggested that C. elegans does not orient to magnetic fields and raises issues that cast doubt on our study. Here we reanalyze Lander et al.s data to show how they appear to have missed observing positive results, and we highlight differences in experimental methods and interpretations that may explain our different results and conclusions. Moreover, we present new data from our labs together with replication by an independent lab to show how temporal and spatial factors influence the unique spatiotemporal trajectory that worms make during magnetotaxis. Together, these findings provide guidance on how to achieve robust magnetotaxis and reinforce our original finding that C. elegans is a suitable model system to study magnetoreception.

neuroscience↗