Haverford College Annual Progress Report: 2012 Formula Grant

Transcription

1 Haverford College Annual Progress Report: 2012 Formula Grant Reporting Period January 1, 2013 June 30, 2013 Formula Grant Overview Haverford College received $18,238 in formula funds for the grant award period January 1, 2013 through June 30, Accomplishments for the reporting period are described below. Research Project 1: Project Title and Purpose Pathoadaptivity of Cell Detaching Escherichia coli We have accumulated preliminary evidence pointing to cell detaching Escherichia coli (CDEC) as diarrheal pathogens. We hypothesize that E. coli strains that fit this description evolved greater virulence through loss of a common gene encoding the lactose repressor. We will determine whether CDEC represent a distinct diarrheagenic E. coli lineage and will identify potential virulence targets that could be repressed by laci. Anticipated Duration of Project 1/1/2013 6/30/15 Project Overview Cell-detaching E. coli (CDEC) detach cultured epithelial cells from solid supports. In 1993 CDEC were identified as unusual fecal isolates associated with infantile diarrhea. Cell detachment was later attributed to specific bacterial toxins, which with the adhesins also elaborated by CDEC, are produced by E. coli that cause urinary tract infections. Some researchers suspect that these organisms are merely potential uropathogens in stool. A handful of other studies however support the idea that CDEC may be diarrheal pathogens. CDEC isolates are predominantly unable to utilize lactose and the question as to whether or not they can cause diarrhea remains unanswered. The lactose repressor (LacI) is an E. coli protein that ensures that lactose-utilization genes (lacayz) are only expressed when that sugar is available. LacI is a recently uncovered antivirulence factor in Salmonella because it blocks virulence gene expression. The current evolutionary model proposes that negative selection led to loss of laci in Salmonella and then created selective pressure for subsequent loss of lacayz, accounting for lactose-negativity. Noting that pathoadaptive evolutionary paradigms have played out in multiple pathogens, we screened CDEC isolates from an earlier epidemiological study for laci and found that CDEC as a whole were not significantly associated with disease, but laci negative CDEC were (p<0.05). A CDEC strain expressing LacI is less virulent than the isogenic wildtype strain in a nematode model. We hypothesize that CDEC are a diarrheagenic lineage of Haverford College 2012 Formula Grant Page 1

2 E. coli that lost the laci gene, and subsequently other lac genes, as they evolved virulence because laci represses one or more CDEC virulence factors. We will test this hypothesis through the following specific aims: (1) Do CDEC represent a distinct lineage of diarrhea-causing E. coli? We will use multilocus sequence typing to determine whether CDEC strains from different individuals belong to the same or different E. coli lineages. (2) Is loss of laci by CDEC pathoadaptive? We will determine whether LacI binds to virulence genes and/or their promoters and thereby represses the transcription of genes that are important for virulence. Principal Investigator Iruka N. Okeke, PhD Associate Professor of Biology Haverford College 370 Lancaster Avenue Haverford, PA Other Participating Researchers Erin Remaly BSc, MSc employed by Haverford College Ann Wolski HC 13 student at Haverford College Expected Research Outcomes and Benefits The principal outcome of our study will be to determine whether CDEC are indeed pathogens, and therefore merit further study. This project will also contribute to the understanding of pathodaptivity through loss of house-keeping genes, as well as to the current knowledge on gene regulation by the E. coli lactose repressor. Strains that fit the description of CDEC have only been found in studies where under-nutrition was observed. It is therefore possible that CDEC are only virulent in a subset of individuals (for example the very young, the very old or the undernourished) that are at high risk of persistent diarrhea of unknown etiology and that epidemiologic studies that are not appropriately stratified will miss them. In the event that we are able to identify CDEC lineages and/or elucidate virulence mechanisms, these preliminary data will form the basis for collaborative research to define the epidemiology and risk groups for CDEC diarrhea as well as the potential for interventions available or in development for uropathogenic E. coli to ameliorate CDEC disease. Should our data support CDEC diarrheagenicity and the pathoadaptivity of LacI in these bacteria, we will institute collaborations to investigate their role in diarrheal disease. The experiments described in this project will be performed largely by mentored undergraduate students working in the laboratory of the PI. The students will receive training in molecular microbiology research and opportunities to present their work to a scientific audience. Findings from the study will be disseminated through publication in the peer-reviewed scientific literature and presentation and a national microbiology conference. Haverford College 2012 Formula Grant Page 2

3 Summary of Research Completed Specific aim 1: To determine whether Cell Detaching Escherichia coli (CDEC) represent a distinct lineage of diarrhea-causing E. coli. No progress to report at this time. Specific aim 2: To determine whether loss of laci by CDEC is pathoadaptive IBC approval for experiments requiring genetic modification Our application to genetically modify Escherichia coli strains for the purpose of this research was reviewed by Haverford College s Institutional Biosafety Committee at its spring meeting. The Institutional Biosafety Committee exempted this project from review based on Section III-F of the NIH Guidelines for Research Involving Recombinant DNA Molecules because the project will use recombinant DNA that is Entirely from a prokaryotic host including its indigenous plasmids or viruses when propagated only in that host or when transferred to another host by well established physiological means and Entirely segments from different species that exchange DNA by known physiological processes, though one or more may be a synthetic equivalent Identification of candidate lac operator sequences for further investigation In order to identify and prioritize potential operator sequences to which the lactose utilization (lac) repressor (LacI) might bind, we queried publicly available E. coli genome sequences with a short-sequence nucleotide Basic Local Alignment Search tool (BLAST) of the lac operator sequence defined by Dickson et al. in 1975 (Genetic regulation: the Lac control region. Science 187: 27-35). No cell-detaching E. coli (CDEC) strains have yet been genome-sequenced but CDEC share virulence factors with uropathogenic E. coli. We therefore used nine uropathogenic E. coli genomes as subjects for the BLAST and additionally queried nine E. coli commensal genomes. Our BLAST search identified over 200 initial hits and we prioritized those that were absent or less prevalent in commensal strains for further analysis. Genes containing or immediately downstream of these operator sites are listed in Figure 1. The next line of analysis comprised peptide and nucleotide BLAST searches of each gene sequence containing a putative lac operator motif in the 18 strains of interest. The purpose of this second search was to determine whether each gene was present in additional genomes without an intact or discernible lac operator site. Genes with altered operator sites would not have shown up on our original BLAST, but a comparison of operator site presence/absence within each individual gene could prove important in understanding virulence mechanisms and allow for better prioritization of the candidate genes. Indeed, gene-specific BLASTs revealed that a high number of strains contain the genes of interest without intact lac operator sites. We assessed the significance of association with uropathogenic or commensal genomes using the Fisher s Exact test. For five genes, there was a significant association of the presence of the gene containing the lac operator in uropathogenic E. coli compared to commensal E. coli (p < 0.05) (Figure 1). Among the five were potential virulence loci, encoding the Tia invasin and a Haverford College 2012 Formula Grant Page 3

4 capsular export protein. For two other loci that could play a role in regulation, an invasion protein regulator and a LexA-type repressor, an association was noted but no statistically significant differences were seen between uropathogenic and commensal strains Generation of a CDEC derivative containing laci in single-copy To accomplish specific aim 2, we will need to compare expression of candidate genes in a CDEC background in the presence and absence of LacI. Making such comparisons in a strain expressing LacI from a plasmid could lead to artifacts due to copy number effects. We therefore set out to construct a CDEC strain derivative that carries a single copy of laci. Although this construct will be needed later in the project, preparing the construct is one of our early tasks because we have no prior experience making such modifications in CDEC and therefore might have to try multiple approaches. A derivative of Mackenzie and Craig s transposon 7 (Tn7)-based transposition vector pgrg36 (Fast, easy and efficient: site-specific insertion of transgenes into enterobacterial chromosomes using Tn7 without need for selection of the insertion event. BMC Microbiology 6: 39, 2006) was used to site-specifically integrate laci cloned from E. coli K-12 and a kanamycin/neomycin resistance marker into the attachment site of Tn7 (atttn7) on the chromosome of laci-negative CDEC strain C06b. The temperature-sensitive plasmid pgrg36 and its derivatives were maintained at 30 C. A kanamycin and neomycin-resistant vector derivative, pgrg36k was prepared by cloning a 1,252 bp aminoglycoside phosphotransferase gene cassette (originally from Tn903) excised with HincII from puc4k into the SmaI site of pgrg36. A 1.2K b fragment containing laci and its natural promoter were amplified from a lac-operon-positive E. coli strains by polymerase chain reaction (PCR) and cloned into the NotI site of pgrg36k to produce a clone we have designated pink1351. Both the neomycin/kanamycin resistance cassette and the laci lie adjacent to one another and between the Tn7 recognition sites of pink1351. Integration of these genes at atttn7 (which lies between the genes psts and glms) on the chromosome of CDEC strain C06b was effected as described by Mackenzie and Craig. Briefly, plasmid pink1351 was electroporated into CDEC strain C06b and transformants were selected on LB containing neomycin and ampicillin at 30 C. Transformants were cultured overnight without selection in Luria broth at room temperature, transcription of the transposition genes, tnsabcd, was induced from an arabinose promoter for two hours by addition of 0.2% arabinose and dilutions of the induced culture were plated on LB with neomycin and incubated at 44 C overnight. Following one subculture at 44 C overnight, C06b::lacI candidates were identified by PCR. The background of positive candidates was verified by plating on sheep s blood agar, Macconkey agar and Luria agar containing ampicillin as well as Luria agar containing neomycin. Genotype was verified by PCR with four more primer pairs. We successfully produced the desired construct, which was lactose negative on Macconkey agar, hemolytic on blood agar, neomycin resistant and ampicillin sensitive. We additionally verified that it carried the laci gene, located between the glms and psts genes. For control purposes, an isogenic construct containing the neomycin/kanamycin resistance cassette without laci was also prepared. Haverford College 2012 Formula Grant Page 4

5 Figure 1: Schematic summary of the results of individual gene BLASTs for each prioritized gene. Genes containing, or immediately downstream of, putative lac operator sites identified by short-sequence BLAST are listed in descending order of p-value computed by Fisher s Exact Test. Orange shading indicates that both the gene and the operator site were present in the strain listed at the top of the chart; blue shading indicates that the gene was present but the operator site was either absent or altered; white cells indicate that the gene was not present in the strain. U stands for uropathogenic, C for commensal. Haverford College 2012 Formula Grant Page 5