BranchSquare.jpg (621611 bytes)
banner.JPG (30152 bytes)
  MI-ASM Home  
     
  National ASM site  
Abstracts - Spring 2018
Oral Presentations

Assembling Viral Genomes in Metagenomic Data using Incomplete or Remote Reference Genomes

Jiao Chen* and Yanni Sun

Computer Science and Engineering, Michigan State UniversityEast Lansing, MI 48824, USA.

Characterizing viral genomes is an important step for understanding viral evolution, fitness, and their interactions with the host. An increasing number of metagenomic data sets containing viral communities become available and bioinformatic analysis has revealed important new knowledge about viruses. The performance of existing pipelines for characterizing viruses in metagenomic data heavily depends on both de novo assembly and the construction of the reference databases. Conducting strain-level de novo assembly is computationally difficult due to the high similarity between strains, heterogeneous coverage of viruses, and the sheer data size. Constructing quality references is not trivial either because of the high diversity of viruses and the high mutation rates of RNA viruses. Thus, there is still a need of new methods for strain-level RNA virus assembly in metagenomic data.In this work, we developed a hybrid method named OL_PEHaplo that conducts strain-level de novo assembly using only partial or remotely related reference genomes. It applies a BWT-based overlap extension method to enrich reads from incomplete aligned reads. Subsequently, with a paired-end guided path finding algorithm, it assembles viral strains at the strain-level from the enriched reads constructed overlap graph. We demonstrated its usage on identifying fast-mutating RNA viruses such as HCV, HBV from a clinical metagenomic data sampled from human plasma. It competes favourably with benchmarked tools. Furthermore, when the reference sequences are functional sites, such as protein coding genes, OL_PEHaplo can output the whole virus genome containing the target genes. We applied OL_PEHaplo on reads aligned to a partial CDS of HGV genome (1,073 bp), it successfully assembled the near complete HGV genome (9,392). OL_PEHaplo can be used by itself for assembling targeted viruses in metagenomic data. In addition, it also provides better inputs for existing virus functional and taxonomical analysis pipelines.Availability: The source code and the documentation of OL_PEHaplo is available at https://github.com/chjiao/OL_PEHaplo.

BUCCILLI WINS BEST GRADUATE STUDENT TALK!

Copine A Interacts with Actin Filaments and Plays a Role in Regulation of Actin Filaments in Dictyostelium discoideum

Matthew J. Buccilli1*, Mingxi Han1, Andrew A. Banas1, Hanqian Mao1, April N. Ilacqua1, David Loiselle2, Timothy A.J. Haystead2, and Cynthia K. Damer1

1Department of Biology, Central Michigan University, Mount Pleasant, MI 48859
2Department of Pharmacology and Cancer, Duke University Medical Center, Durham, NC 27708

Copines make up a family of calcium-dependent, phospholipid-binding proteins. Copine proteins consist of two C2 domains at the N-terminus followed by an A domain similar to the von Willebrand A domain found in integrins. We are studying copine protein function in the model organism, Dictyostelium discoideum, which has six copine genes, cpnA- cpnF. Previous research showed that cells lacking the cpnA gene exhibit a cytokinesis defect, a contractile vacuole function defect, and developmental defects. To provide insight into the role of CpnA in these cellular processes, we used column chromatography and immunoprecipitation to isolate proteins that interact with CpnA. These proteins were identified by mass spectrometry. One of the proteins identified was actin. To determine if CpnA associates with the actin cytoskeleton, we treated cells expressing GFP-CpnA or GFP-Ado (containing only the A domain of CpnA) with Triton X-100 and spun down the insoluble cytoskeletal fraction. GFP-CpnA was found in the cytoskeletal pellet only in the presence of calcium, while GFP-Ado was found in the cytoskeletal pellet in the presence and absence of calcium. To determine if CpnA directly binds to actin, we performed F-actin binding assays with purified GST-CpnA and found that GST-CpnA bound to actin filaments in a calcium-dependent manner. In addition, we performed immunoprecipitations with cells expressing GFP, GFP-CpnA, and GFP-Ado. We then incubated these precipitated proteins with F-actin or G-actin in the presence or absence of calcium. We found that F-actin, but not G-actin, co-precipitated with GFP-CpnA in the presence of calcium. Cells lacking cpnA exhibited aberrant actin filament polymerization/depolarization in response to cAMP stimulation. These results indicate CpnA plays a role in actin-based processes and may do so by directly interacting and regulating actin filament dynamics.

LARNER WINS BEST UNDERGRADUATE STUDENT TALK!

Identifying Fecal Contaminants using Microbial Source Tracking at Luna Pier Beach
Larner, Paige V*, Nshimyimana, Jean Pierre1, Rose, Joan B.1
1Michigan State University, East Lansing, Michigan, 48823

Fecal excretion and associated bacteria due to various animals contaminate public beaches and influence public health. Luna Pier Beach (Monroe County, MI) experienced 72 days of closure during summer months due to an increase in microbial pollution. This study was conducted to determine the source of the high Escherichia coli (E. coli) concentrations. This study was conducted to determine the sources using Bacteroidales-associated microbial source tracking markers. Water samples ranging from 65 to 250 milliliters were collected from five different locations along the beach and filtered on site. DNA was then extracted and processed using droplet digital polymerase chain reaction (ddPCR). Gull (qGull), dog (BacCan) and human-specific (B. theta) Bacteroidales-associated markers based on published primers and probes were used. Thirty water samples were collected between July and August 2017. E. coli counts ranged from 7 colony forming units per 100 milliliters (CFU/100mL) to 517 CFU/100mL. The B. theta human marker accounted for the highest count of positive samples, with 27 of the 30 total samples. Fifteen and zero of the samples tested positive for qGull and BacCan, respectively. The North and Central locations of the beach resulted in the greatest incidences of fecal contamination for both gull and human markers. B. theta ranged from 97 to 1100 gene copies (GC)/100mL and qGull resulted in 97 to 743 GC/100mL. The results are being used to remediate septic tanks and sewage spills in the area. Storage of environmental samples is essential in the process of analysis. Environmental samples of manure and raw wastewater of Lansing, MI were collected, combined and diluted, then filtered as 10mL volumes and extracted for DNA. Quantifiable differences of DNA stability were measured over a 60-day period between samples- half of which are stored as filters and half stored as DNA extracts. Samples were assayed with B. theta and CowM2 markers using ddPCR. Our current results do not indicate a significant change in DNA concentration over this time period. Results from this study can be used to advance suggested methods of storage for environmental samples.

Computational Evidence of Colon Cancer Promoting Mechanisms from Streptococcal Infections
School, Kuana M.* and Harris, Laura K. Davenport University Lansing, MI. 48933

Enterococcus faecalis (E. faecalis) and Streptococcus gallolyticus (S. gallolyticus) infections cause colorectal cancer (CRC) through increased inflammation, mitochondrial DNA damage, and decreased cell cycle regulation mechanisms. Although Streptococcus macedonicus (S. macedonicus) is a relative to these species, no studies have examined if S. macedonicus causes CRC by the same mechanisms. We defined 3 signatures (T-statistic ranked list of differentially expressed genes) between uninfected CRC cells and E. faecalis, S. gallolyticus, or S. macedonicus infected cells. We used Gene Set Enrichment Analysis (GSEA) to compare those signatures, between species and to biologically relevant oncology and immunological gene sets from a Molecular Signatures database (Broad Institute), to predict if S. macedonicus contributes to CRC development. We used the Database for Annotation, Visualization and Integrated Discovery to identify Gene Ontology categories associated with similarities. We found significant (p<0.001) similarities among signatures, suggesting streptococcal infection regardless of species promote CRC mechanisms. We see down-regulation of mitochondrial processes (p<0.05). Comparing signatures to biologically relevant gene sets reveal that S. macedonicus infection increases cancer-promoting cytokines while decreasing cell cycle regulation in CRC cells, similar to its streptococcal relatives. These findings suggest S. macedonicus may cause CRC in humans. Our computational results suggest S. macedonicus promote CRC similarly to its streptococcal relatives. This work is the first evidence supporting that prediction. We also show GSEA is a helpful tool for examining bacterial effects on cancer. Presented at the annual West Michigan Regional Undergraduate Science (WMRUGS) Research Conference on November 4, 2017, at the Van Andel Research Institute.

Poster Presentation

SPITZER WINS BEST UNDERGRADUATE POSTER!

Genome-wide Computational Analyses of Antibiotic Response in Escherichia coli
Kristine Spitzer* and Laura Harris, Davenport University, Lansing, MI 48933

BACKGROUND: Escherichia coli causes a wide variety of diseases including urinary, gastrointestinal, and blood infections, leading to approximately 73,480 illnesses annually in the United States. This is complicated by growing antibiotic resistance, making E. coli infections a serious public health threat. This study compares how E. coli responds to sub-lethal antibiotic treatments to explore selective pressures leading to developing antibiotic resistances. METHOD: This study applied Gene Set Enrichment Analysis, an algorithm commonly used for examining human cancers, to microarray data from a sensitive E. coli strain treated with various antibiotics to examine how these therapies affect cellular processes. RESULTS: We observed similarities in the cellular processes altered by ampicillin, kanamycin, and gentamycin treatment. These processes included decreased ferredoxin (i.e. iron-sulphur binding) gene expression. Norfloxacin did not induce such changes. Rather, response to norfloxacin is characterized by increased DNA repair proteins and decreased flagellum expression. Decreased flagellum expression was also noted in protein synthesis inhibitors, but not cell wall synthesis inhibitors. CONCLUSION: Our findings concur with previous reports that ferredoxin plays a role in antibiotic resistance. We further confirmed previous literature describing the up-regulation of DNA repair proteins for norfloxacin and down-regulation of flagella in protein synthesis inhibitors. We discovered the same applies to DNA synthesis but not cell wall synthesis inhibitors. This knowledge provides insight into the selective pressures E. coli must overcome for the development of antibiotic resistance and these molecular changes have the potential to be developed into reverse antibiotic therapy.

Multi Antibiotic Resistance Mechanisms in E. coli Exposed from Genome-wide Analyses
Ronna Rummer* and Laura Harris, Davenport University, Lansing, MI 48933

BACKGROUND: Antibiotics are the front line defense against bacterial infections. Antibiotic resistant strains are developing at alarming rates and causing life-threatening illnesses. Among them, E. coli is a frequent cause of human infections and untreatable E.coli strains are a growing public health concern. Our goal is to compare molecular changes associated with antibiotic resistance in E. coli to identify similarities that represent global antibiotic resistance mechanisms. METHOD: We used Gene Set Enrichment Analysis on microarray data that compared a sensitive E. coli strain to 10 E.coli strains, each resistant to one of 10 different antibiotics to look for similarities in cellular processes associated with antibiotic resistance. RESULTS: We detected similarities in differentially expressed genes across 10 antibiotic resistant strains, regardless of mechanism of action. In these similarities, we found thiamine production suppressed in strains resistant to cell wall and DNA synthesis inhibitors. We further found suppression of aminoacyl-tRNA biosynthesis in all strains resistant to folic acid, protein synthesis, and DNA synthesis inhibitors. While up-regulated cellular processes could not be identified, we determined which genes were most commonly over-expressed across antibiotic resistances. We recommend these genes and their associated processes be further characterized as they may be future therapeutic targets. CONCLUSION: Uncovering similarities in antibiotic resistance mechanisms can lead to the identification of reverse antibiotic therapies that can be co-administered with current antibiotics to restore their efficacy against antibiotic resistant infections. If further developed, the targets identified here could help thousands of patients with life-threatening infections.

Improve the sensitivity of detecting long read overlaps using grouped short k-mer matches
Nan Du*, Michigan State University, East Lansing, MI 48824
Jiao Chen, Michigan State University, East Lansing, MI 48824
Yanni Sun, Michigan State University, East Lansing, MI 48824

Single-molecule, real-time sequencing (SMRT) developed by Pacific BioSciences produces longer reads than secondary generation sequencing technologies such as Illumina. The long read length enables PacBio sequencing to close gaps in genome assembly, reveal structural variations, and characterize the intra-species variations for microbes. It also holds the promise to decipher the community structure in complex microbial communities because long read length helps metagenomic assembly. One key step in genome assembly using long reads is to quickly determine the overlap regions between reads and use that information to create an overlap graph. Because PacBio data has higher sequencing error rate and lower coverage than popular short read sequencing technologies (such as Illumina), efficient detection of true overlaps requires specially designed algorithms. In particular, there is still a need for improving the sensitivity of detecting small overlaps or overlaps with low sequence identity. Addressing this need will enable better assembly for metagenomic data produced by the third-generation sequencing technologies. In this work, we designed and implemented an efficient overlap detection program named GroupK, for PacBio reads based on k-mer hits. While using k-mer hits for detecting read overlap has been adopted by several existing programs, our method uses a group of short k-mer hits to increase the sensitivity in detecting small overlaps between long reads. Group k-mer hit was originally designed for remote homology search. We are the first to apply group hit to long read overlap detection. The experimental results of applying our pipeline to both simulated and real PacBio data showed that GroupK enables more sensitive overlap detection, especially for PacBio datasets of low sequencing coverage.

Investigating Frequency of Copy Number Variations (CNV) at Fragile Sites in S. cerevisiae
Presenters: Constance Vernham* & Samantha Bartolet
Eastern Michigan University, Ypsilanti, MI 48197

Copy number variations (CNVs) are deletions or duplications of large genomic regions, which have been linked to genetic diseases including tumor development and autism. CNVs are hypothesized to form by replication fork stalling and polymerase template switching, events that frequently happen in DNA at human common fragile sites. Using yeast containing human fragile site DNA, we investigated deletions and duplications of a reporter cassette at three locations that differ by their proximity to an AT-rich area of the fragile site, a sequence that can stall replication. Our data show that CNV frequency is affected by which side of the AT-rich area the reporter cassette is integrated.

Nutritional Shifts Affect Coaggregation Patterns and Cell Surface Properties in Lactobacillus rhamnosus GG
Authors: Rielinger, Amanda L*, Clemans, Daniel, PhD, Eastern Michigan University, Ypsilanti, MI 48197

Abstract: Coaggregation is thought to play a predominant role in biofilm development. Understanding how intestinal microorganisms undergo coaggregation interactions is therefore essential to understanding how these species form biofilms in the human gastrointestinal tract. This research attempted to characterize coaggregation interactions between Bacteroides spp. and Parabacteroides spp. and probiotic Lactobacillus rhamnosus GG (LGG) cultured in tryptone, yeast extract, and glucose (TYG), tryptone, yeast extract, and fructose (TYF), soytone, yeast extract, and glucose (SYG), and beef extract, yeast extract, and glucose (BYG) media. Visual in vitro coaggregation assays indicated that nutritional variation resulted in changes in coaggregation patterns between LGG and Bacteroides spp. and Parabacteroides spp. Visual in vitro carbohydrate and amino acid coaggregation inhibition assays were conducted on those pairs that consistently scored a 2 or above in the coaggregation assays. Coaggregation between Bacteroides fragilis ATCC 25285 and LGG cultured in TYG, TYF, SYG, and BYG was inhibited by L-histidine and L-lysine. Coaggregation between Bacteroides vulgatus ATCC 8481 and LGG cultured in TYF was inhibited by L-histidine, L-lysine, and L-serine. Hydrophobicity assays conducted on LGG cultured in different media types suggested that cell surface properties are also affected by culture conditions. As the evidence presented here suggests, dietary nutritional changes lead to metabolic shifts that may affect the cell surface and subsequent coaggregation interactions in LGG.

The Cellular Localization and Lipid Binding Properties of CpnF in Dictyostelium Cells
Graham, Bria N.* and Damer, Cynthia K.
Biology Department, Central Michigan University, Mount Pleasant, MI 48859

Copines are made up of a family of calcium dependent, phospholipid binding proteins. Copine proteins contain two C2 domains and A domain. The C2 domains are in the N-terminal half and are followed by an A domain in the C-terminal half. Although the exact function of copines is unknown, copines are thought to be involved in cell signaling. In the model organism Dictyostelium discoideum, six copines have been discovered, cpnA-F. We are studying one of the copine proteins, cpnF. To study the properties of CpnF, we have created two cell lines that express a GFP (Green Florescent protein)-tagged version of CpnF: one with one with GFP at the N-terminus and one with GFP at the C-terminus. The lipid binding characteristics of CpnF were studied by immunoprecipitation and lipid dot blots, and the intracellular localization of CpnF was investigated using confocal microscopy. GFP-tagged CpnF was shown to bind to phosphorylated forms of phosphatidylinositol in calcium-dependent manner. When using a PIP array lipid dot blot, the strongest binding of CpnF was to phosphatidylinositol 4-phosphate (PI(4)P) and phosphatidylinositol 3-phopshate (PI(3)P), with less binding to phosphatidylinositol 4,5-phosphate (PI(4,5)) and phosphatidylinositol 3,5-phosphate (PI(3,5)). In fixed cells, GFP-tagged CpnF was found on intracellular vacuoles and the plasma membrane. To identify the GFP-CpnF labeled intracellular structures, we used DAPI to label the nucleus and mitochondria, fluorescent-tagged wheat germ agglutin (WGA) to label the Golgi apparatus, and small fluorescent beads (0.04?m) to label endosomes/lysosomes. GFP-tagged CpnF was found to colocalize the nucleus, the Golgi apparatus, and endosomes/lysosomes. No colocalization was observed for the mitochondria. GFP alone was shown to localize to the nucleus and Golgi apparatus. Given that CpnF bound to phosphorylated forms of PI, which are important in endocytic processes, and associated with endosomes and lysosomes, our data suggests that CpnF may play a role in endocytosis. We are currently working on creating a cpnF knockout mutant in Dictyostelium to study this further.

Copine A is Involved in Membrane Trafficking in Dictyostelium discoideum
Amber D. Ide, Elise M. Wight and Cynthia K. Damer
Biology Department, Central Michigan University, Mount Pleasant, MI 48859

Copines are highly conserved, calcium-dependent binding proteins that have a unique combination of two C2 domains at the N-terminus and an A domain at the C-terminus. Copines are found in many eukaryotic species and are thought to be involved in many cellular processes, such as membrane trafficking, calcium-dependent signaling pathways, and growth and development. The goal of this research is to understand what role Copine A (CpnA) may have in membrane trafficking processes and cell adhesion properties. Using flow cytometry and fluorescence microscopy, we have found that cpnA- cells exhibited a faster rate of phagocytosis of both fluorescent-labeled beads and GFP-labeled bacteria compared to wild-type Dictyostelium cells. In addition, we found that the wild-type strain NC4A2 exhibited a faster rate of phagocytosis than another wild-type stain, AX4. Exocytosis rates of beads and digestion rates of bacteria were similar in cpnA- and wild-type cells. To further understand CpnA’s role in phagocytosis, we wanted to know if adhesion properties were contributing to the higher rates of phagocytosis observed with the cpnA- cells. Therefore, we used sodium azide to rid the cells of bacteria and latex beads that may be adhered to the outside of the cell. These studies showed that more beads and bacteria were attached to the outside of cpnA- cells than wild-type cells and that washing cells with sodium azide did release the bacteria and beads from the outside of the cells. Lastly, we wanted to test the adherent properties of wild-type versus cpnA- cells to surfaces. By shaking cells on small dishes we observed that cpnA- cells have increased adhesion to the surface of the dish. These results indicate that CpnA cells have enhanced adhesion properties compared to wild-type Dictyostelium cells and these adhesion properties allow the cells to phagocytosis at faster rates.

Salmonella enterica serovar Typhimurium is a facultative, foodborne, enteric pathogen responsible for gastric enteritis in humans and typhoid in mice
Rebecca Gordon and Aaron Baxter, Grand Valley State University

A common reservoir is poultry, thus undercooked eggs and contaminated food can lead to an infection. After ingestion of this pathogen, it adheres and invades the epithelium of the distal small intestine using a type three secretion apparatus. Once invasion of the epithelium occurs, the bacteria can invade surrounding epithelial tissue through the basolateral surface or invade mucosal macrophages. Should the bacteria enter the bloodstream, a systemic infection can occur with mortal consequences. Salmonella uses complex regulatory mechanisms to determine where to adhere, invade, and colonize the host. The region of the genome responsible for regulation of mucosal epithelial invasion is Salmonella Pathogenicity Island-1 (SPI-1) and colonization of macrophages utilizes SPI-2. Multiple different SPIs exist that regulate other aspects of pathogenesis, and the focus of our research is a putative pathogenicity island that has yet to be characterized. Within this putative island, individual mutations were made in the open reading frames to discern whether the deletion impacted SPI-1 expression (via a hilA-lacZ gene fusion), motility, and invasion.

Salmonella enterica serovar Typhimurium is a facultative, foodborne, enteric pathogen responsible for gastric enteritis in humans and typhoid in mice
Rebecca Gordon and Aaron Baxter, Grand Valley State University

A common reservoir is poultry, thus undercooked eggs and contaminated food can lead to an infection. After ingestion of this pathogen, it adheres and invades the epithelium of the distal small intestine using a type three secretion apparatus. Once invasion of the epithelium occurs, the bacteria can invade surrounding epithelial tissue through the basolateral surface or invade mucosal macrophages. Should the bacteria enter the bloodstream, a systemic infection can occur with mortal consequences. Salmonella uses complex regulatory mechanisms to determine where to adhere, invade, and colonize the host. The region of the genome responsible for regulation of mucosal epithelial invasion is Salmonella Pathogenicity Island-1 (SPI-1) and colonization of macrophages utilizes SPI-2. Multiple different SPIs exist that regulate other aspects of pathogenesis, and the focus of our research is a putative pathogenicity island that has yet to be characterized. Within this putative island, individual mutations were made in the open reading frames to discern whether the deletion impacted SPI-1 expression (via a hilA-lacZ gene fusion), motility, and invasion.

Copine A Regulates cAMP-mediated Actin Filament Dynamics and Chemotaxis in Dictyostelium
April N. Ilacqua, Kory J. Hobbs*, Samantha P. Perry and Cynthia K. Damer
Biology Department, Central Michigan University, Mount Pleasant, MI 48859

Copines are calcium-dependent phospholipid binding proteins found in many eukaryotic organisms. Copines have been suggested to be involved in calcium-dependent cell signaling pathways that regulate a wide variety of cellular processes. We have chosen the model organism Dictyostelium discoideum to investigate the function of copine proteins. Copines bind to membranes in a calcium-dependent manner, but their exact function is unknown. In previous studies, Dictyostelium cells lacking the cpnA gene were unable to develop properly and demonstrated cytokinesis and contractile vacuole functional defects. Furthermore, previous studies also indicated that CpnA is able to bind actin filaments and cpnA- cells exhibited atypical actin polymerization in response to cAMP. To further investigate CpnA’s role in actin filament regulation, aggregate competent cells were stimulated with cAMP and subsequently fixed and stained with rhodamine-phalloidin at 0, 5, 10, 20, 40, 60 second time-points. Cells were analyzed using ImageJ and cpnA- cells were found to be less polarized with fewer plasma membrane projections compared to wild-type cells. Assays were performed to characterize any defects in chemotaxis. A droplet of cells was placed on agar containing either a cAMP or folate gradient and cell population movement away from the original drop was analyzed. These data showed that cpnA- cells moved in response to cAMP and folate, but did not move toward the chemotactic signal, while wild-type cells did. Dunn chambers and ImageJ were used to track individual cell trajectories within a cAMP gradient. cpnA- cells were unable to properly migrate towards the cAMP gradient. Wild-type cells sensed the cAMP gradient and migrated in the direction of the chemoattractant. These results suggest that CpnA plays a role in regulating actin filament dynamics and this function may be important to CpnA’s role in cAMP chemotaxis

A Type IV Pilus Plays Cupid: Bringing Caldicellulosiruptor bescii Closer to Polysaccharides
Khan Asma MAM*#, Bell Valerie J#, Blumer-Schuette Sara E##
Oakland University, Rochester, MI 48309.

At the upper thermal limits for biological crystalline cellulose hydrolysis, only one bacterial genus, the Caldicellulosiruptor thrive. From this genus, eight species are strongly cellulolytic, among which Caldicellulosiruptor bescii has been extensively studied with regards to its extraordinary cellulolytic ability. Although the cellulolytic ability of C. bescii is largely attributed to its modular, multi-functional carbohydrate acting enzymes, this thermophile is also likely to increase efficiency of cellulose hydrolysis through the use of non-catalytic mechanisms. One example would be attachment to lignocellulosic biomass and polysaccharides through the use of proteins like the t?pirins, S-layer located proteins or substrate binding proteins. Using genome annotation, we identified a complete T4P gene locus in C. bescii, arranged in a putative operon. Based on the proximity of the T4P operon to the glycoside hydrolases genes, we hypothesize that the T4P plays a role in attachment to plant biomass. Interestingly, the T4P from highly cellulolytic Caldicellulosiruptor are evolutionarily divergent from those of weakly cellulolytic species. Based on pilin-like protein domains, transcriptomics analysis and protein expression, we annotated the hypothetical protein Athe_1880 (PilA), as the major pilin and produced a soluble, recombinant form: Athe_1880T, by truncating the predicted hydrophobic regions. Immunoblots confirmed that xylan is the key inducer of Athe_1880 based on the highest presence of Athe_1880 on cell surfaces compared to other representative plant polysaccharides. Immunofluorescence microscopy confirmed that Athe_1880 is present at the outer surface of C. bescii cells. Furthermore, competitive cell binding assays indicate that Athe_1880T competes with C. bescii cells for binding sites on insoluble polysaccharides. From these observations we propose that the major C. bescii pilin Athe_1880, and by extension its T4P, plays a role in cell attachment to plant polysaccharides.

A Membrane Sensor Protein Alters Adhesion and Biofilm Formation in C. albicans
Bennett, Julia E.*, Cleary, Ian A. Grand Valley State University, Allendale, MI 49401

The opportunistic fungal pathogen Candida albicans grows in various human host environments ranging in factors such as pH, oxygen concentration, and osmotic strength. The cell responds to these stimuli by sensing changes at the cell membrane. The gene orf19.6705 is predicted to encode a cell membrane sensor protein and could therefore play a role in how C. albicans successfully adapts to its varied surroundings. When we deleted both copies of this gene we found that the adhesive properties of hyphae were altered with increased cell adhesion and biofilm formation in some media. In other media cell adhesion and biofilm formation was decreased, suggesting that the protein can activate or repress signaling pathways depending on the stimulus. We also saw changes in the appearance of embedded growth filamentation. Collectively these results reinforce the function of this protein as a sensor, and we are continuing our analysis examining changes in gene expression of surface adhesions.

A Candida-specific Membrane Protein Affects Cell Adhesion and Filamentation        Connor R McCarthy and Ian Cleary, Grand Valley State University

The co-repressor Nrg1p is a key regulator of cell morphology in the opportunistic fungal pathogen Candida albicans. Expression of many genes is activated during hyphal growth as a result of relief of Nrg1p-mediated repression, but the functions of some of these genes are not yet known. The gene orf19.2302 encodes a Candida-specific protein predicted to localize to the ER membrane. The only homologue of this protein is a hypothetical protein in the closely related species Candida dubliniensis. In a strain where we deleted both copies of this gene we found that cells showed increased adhesion to each other when grown in cell culture medium, although loss of this gene did not appear to influence biofilm formation. Interestingly, we found that over-expressing this gene resulted in increased filamentation in embedded growth conditions. These results suggest a stimulus-specific function for this gene, and we are continuing to examine growth under different environmental conditions.

Clostridium difficile Incorporation into the Gut Microbiota
Brewer, Thomas G.* Eastern Michigan University Ypsilanti, Michigan 48197

Clostridium difficile is a nosocomial pathogen that arises in patients post antibiotic treatment. However, Clostridium difficile can also appear as a non-pathogenic partner in the endogenous gut microbiota. Biofilms within the gut are composed primarily of Firmicute and Bacteroides species. Until now, Clostridium difficile's interaction with these bacteria has been uncharacterized. Assessing coaggregation of Clostridium difficile and members of Firmicute and Bacteroidetes genera identify the cell-cell interaction of these microbes and provide insight into gut biofilm construction and incorporation of this potential pathogen. Cell-cell interactions of Clostridium difficile with these microbes is then characterized through inhibitory assays to assign class of adhesin to the interaction, and which partner contributes an adhesin or receptor. Hydrophobicity is also assessed to ensure interactions are cell surface molecule based rather than hydrophobic interactions. A panel of inpatient and outpatient isolates were assayed with endogenous species of the gut microbiota, and it was found that interaction with the Bacteriodetes is null, where interaction with probiotic Firmicutes, Lactobacilli and Bifidobacterium, is consistent. This data suggests that Firmicutes of these genera are likely the entry point for Clostridium difficile to incorporate into endogenous gut biofilms.

In Search of New Antimicrobials: New Methods For Mining Antibiotic-Producing Bacteria Tanbour, Razan,* Hasan, Alaa,* Price, Paul, Eastern Michigan University, Ypsilanti, MI

Antibiotic resistance is quickly becoming one of the biggest healthcare issues facing humanity, as many of our existing antibiotics are less effective against a wide range of clinically relevant pathogens. There have been recent reports of bacterial pathogens that are completely resistant to all known antibiotics. Therefore there is an urgent need to develop or discover new classes of antibiotics. Many of our current classes of antibiotics were originally derived from soil microorganisms. However, soil mining for natural products that have antimicrobial activity, commonly referred to as antibiotics, has largely decreased over the past thirty years because the frequency of novel discoveries decreased substantially. Yet, new classes of antibiotics are still being discovered from soil dwelling bacteria. This project aimed to screen Michigan soil samples for bacteria that could potentially produce new/novel antibiotics. We used a modified Waksman method (competing soil microorganisms directly with bacterial pathogens on agar plates) to identify potential antibiotic-producing bacteria. Briefly, a soil sample was retrieved and serially diluted to obtain 200-300 bacteria per plate. Alternatively, diluted soil samples were combined with D-alanine auxotrophic pathogenic bacteria and plated on media containing D-alanine. Individual bacteria were and tested against a series of pathogenic bacteria and/or closely related pathogenic relatives. Bacteria that produced zones of clearing around the pathogenic bacteria were further characterized, noting the size of the zone of inhibition, and identified using 16S rRNA sequencing and BLAST searches against the 16S rRNA database. In total, we identified 24 unique bacterial isolates that produce antimicrobial compounds including 19 that were effective against methicillin-resistant Staphylococcus aureus. We plan to purify and study these antimicrobial compounds to determine their potential usefulness as clinically relevant antibiotics.

Using Insertional Mutagenesis to Study Legume/Rhizobia Symbiosis
Graham, Marion1*, Helliwell, Emily2, Porter, Stephanie2, Griffitts, Joel3, and Price, Paul1,
31Eastern Michigan University, Ypsilanti, MI 48197
2Washington State University-Vancouver, Vancouver, WA 986963 Brigham Young University, Provo, UT 84602

Legume plants rely on a symbiotic relationship with bacteria to obtain fixed nitrogen. Specific bacterial genes are known to affect the developmental processes of root nodules where nitrogen fixation occurs. On such gene, hrrP, has been observed to affect this development on a population-wide scale. The presence of the hrrP gene limits the ability of nodulating bacteria to fix nitrogen for certain ecotypes of plants, thus limiting the bacteria’s host range. We employed standard insertional mutagenesis to study the effects of hrrP on 24 wild bacterial isolates when interacting with their legume hosts.

Testing Functional Redundancy of Two Candida albicans Cell Surface Proteins
Martiradonna, Julia, R.*, Benke, Catherine, R., Cleary, Ian, A.
Grand Valley State University, Allendale MI 49401

Surface proteins are important contributors to virulence in the opportunistic fungal pathogen Candida albicans, primarily through their influence on adhesion to living and non-living surfaces and to biofilm formation. Two genes known to encode surface adhesins are ALS3 and DDR48. ALS3 is a member of a family of surface proteins expressed in hyphae and involved in adhesion. A strain lacking ALS3 forms normal hyphae, but has been reported to be defective in biofilm formation and in host cell damage. The expression of DDR48 is also highly upregulated in hyphae. To test whether DDR48 could compensate for a lack of ALS3, we introduced a DDR48 over-expression construct into the ALS3 deletion strain and tested biofilm formation using the 96-well microtiter plate model. We found that DDR48 over-expression did have some compensatory function in adhesion in some media and we are continuing to examine biofilm formation in additional conditions.

Impact of the SpaCBA Encoding Pilin of Lactobacillus rhamnosus GG (LGG) on Coaggregation with Various Gut Microbes
Dave Lall (GR); Daniel L. Clemans, PhD Eastern Michigan University, Ypsilanti, MI

Many bacteria inhabit the human gut and play a role in health and disease. In industry, some strains are evaluated and developed as probiotic supplements for the treatment of various gastrointestinal diseases. Various molecular mechanisms of the probiotic Lactobacillus rhamnosus GG (LGG) have been well-studied but others still need to be explored. LGG demonstrates a great degree of coaggregation with other gut microbes. The spaCBA gene encodes surface-pilins that mediate adherence to host cells and biofilm formation. The purpose of this study was to determine if SpaCBA mediates the coaggregation between LGG and Bacteroides/ Parabacteroides gut microbes. To accomplish this, a Gibson Assembly construct was made with pUC19, sortase (srtC) and spaC gene fragments from LGG and an erythromycin resistance gene (ermC) from plasmid pGK12. This new plasmid construct was cloned into Escherichia coli DH5a (E. coli DH5a), isolated, and transformed into LGG. Recombination resulted in spaCBA-deficient mutants of LGG that are erythromycin resistant. Biofilms between mutant LGG and other gut microbes still need to be done and will be compared to biofilms between the wild-type LGG and the same gut microbes. Though not complete, this study will bring additional insight to the function of the SpaCBA pilus allowing a greater understanding of LGG’s probiotic role in the human gut and improve human health.

Identification and Characterization of Proteins in Candida albicans for which Overexpression Interferes with Filamentation
Bigler, A.R*., Hoppe, A., and Thomas, D.P., Department of Biomedical Sciences
 

Candida albicans is a fungal commensal organism that is found on mucous membranes and in the digestive tract. Candida albicans is also an opportunistic pathogen that has the capability to invade human hosts that are immunocompromised. This organism can take many forms and is frequently found as either a yeast cell or one of two filamentous forms: pseudohyphae or true hyphae. The ability to form true hyphae in host tissues appears to be a virulence trait that contributes to the overall pathogenicity of the organism. We previously identified a subset of proteins that appeared to be involved in the transition from yeast cells to the filamentous form. The levels of these proteins change at a point of high protein flux and their depletion may be required for filamentation to proceed. To further investigate this observation, we overexpressed some of the proteins in question and characterized the impact on the ability to filament under various inducing conditions. We demonstrate a dramatic effect verifying that high level of these proteins may prevent filamentation. We then tested these strains in the Galleria mellonella model of infection to analyze the relevance of this effect on disease. We demonstrate attenuation of disease in those with dramatically altered filamentation. Overall, we demonstrate that high levels of certain Candida proteins prevent or reduce filamentation and alter the ability to cause disease.

An investigation into the role of Rat1 in splicing in budding yeast
Zuzer Dhoondia and Athar Ansari, Department of Biological Sciences, Wayne State University, Detroit, MI 48202

Transcription mediated by RNAPII yields mRNA that is processed co-transcriptionally by 5? end capping, splicing and 3? end cleavage/polyadenylation. Rat1 is a chromatin-bound factor and it is known to cross-link to the terminator and the promoter regions of a gene. We have discovered that Rat1 also crosslinks to the intronic region of the gene. Using RT-PCR and TRO analysis, we found accumulation of unspliced mRNA for the selected intron-containing genes in the absence of Rat1 in the nucleus. Accumulation of unspliced transcripts was not the consequence of defective termination in the absence of Rat1 activity, and did not require Rat1 termination complex. Upon complementing Rat1-depleted cells with wild type or exoribonuclease defective mutant (D235A) of Rat1, it was clear that exoribonuclease activity of Rat1 was important in regulating the levels of unspliced transcripts. These results suggest a role for Rat1 in either cotranscriptional splicing or cotranscriptional degradation of pre-mRNA. To confirm if the accumulation of unspliced transcript in the absence of Rat1 is due to a direct involvement of Rat1 in splicing or due to a role of Rat1 in degrading the unspliced transcripts, we are performing Rat1 crosslinking experiments to intronic regions by PAR-CLIP approach. Furthermore, we are planning to perform in vitro splicing assay in the presence and absence of Rat1 to further clarify the role of Rat1 in splicing. These experiments will help us in understanding the broader role of Rat1 in transcription and cotranscriptional RNA processing.

 
Click here to return to Spring 2018 conference page.
 
Questions or suggestions concerning website, contact [email protected]
Last updated: December 29, 2018