The Pathogen-annotated Tracking Resource Network (PATRN) system: a web-based resource to aid food safety, regulatory science, and investigations of foodborne pathogens and disease.

Investigation of foodborne diseases requires the capture and analysis of time-sensitive information on microbial pathogens that is derived from multiple analytical methods and sources. The web-based Pathogen-annotated Tracking Resource Network (PATRN) system (www.patrn.net) was developed to address the data aggregation, analysis, and communication needs important to the global food safety community for the investigation of foodborne disease. PATRN incorporates a standard vocabulary for describing isolate metadata and provides a representational schema for a prototypic data exchange standard using a novel data loading wizard for aggregation of assay and attribution information. PATRN currently houses expert-curated, high-quality "foundational datasets" consisting of published experimental results from conventional assays and next generation analysis platforms for isolates of Escherichia coli, Listeria monocytogenes, and Salmonella, Shigella, Vibrio and Cronobacter species. A suite of computational tools for data mining, clustering, and graphical representation is available. Within PATRN, the public curated data repository is complemented by a secure private workspace for user-driven analyses, and for sharing data among collaborators. To demonstrate the data curation, loading wizard features, and analytical capabilities of PATRN, three use-case scenarios are presented. Use-case scenario one is a comparison of the distribution and prevalence of plasmid-encoded virulence factor genes among 249 Cronobacter strains with similar attributes to that of nine Cronobacter isolates from recent cases obtained between March and October, 2010-2011. To highlight PATRN's data management and trend finding tools, analysis of datasets, stored in PATRN as part of an ongoing surveillance project to identify the predominant molecular serogroups among Cronobacter sakazakii isolates observed in the USA is shown. Use-case scenario two demonstrates the secure workspace available for private users to upload and analyze sensitive data, and for collating cross-platform datasets to identify and validate congruent datapoints. SNP datasets from WGS assemblies and pan-genome microarrays are analyzed in a combinatorial fashion to determine relatedness of 33 Salmonella enterica strains to six strains collected as part of an outbreak investigation. Use-case scenario three utilizes published surveillance results that describe the incidence and sources of O157:H7 E. coli isolates associated with a produce pre-harvest surveillance study that occurred during 2002-2006. In summary, PATRN is a web-based integrated platform containing tools for the management, analysis and visualization of data about foodborne pathogens.

Genomic paradigms for food-borne enteric pathogen analysis at the USFDA: case studies highlighting method utility, integration and resolution.

Modern risk control and food safety practices involving food-borne bacterial pathogens are benefiting from new genomic technologies for rapid, yet highly specific, strain characterisations. Within the United States Food and Drug Administration (USFDA) Center for Food Safety and Applied Nutrition (CFSAN), optical genome mapping and DNA microarray genotyping have been used for several years to quickly assess genomic architecture and gene content, respectively, for outbreak strain subtyping and to enhance retrospective trace-back analyses. The application and relative utility of each method varies with outbreak scenario and the suspect pathogen, with comparative analytical power enhanced by database scale and depth. Integration of these two technologies allows high-resolution scrutiny of the genomic landscapes of enteric food-borne pathogens with notable examples including Shiga toxin-producing Escherichia coli (STEC) and Salmonella enterica serovars from a variety of food commodities. Moreover, the recent application of whole genome sequencing technologies to food-borne pathogen outbreaks and surveillance has enhanced resolution to the single nucleotide scale. This new wealth of sequence data will support more refined next-generation custom microarray designs, targeted re-sequencing and "genomic signature recognition" approaches involving a combination of genes and single nucleotide polymorphism detection to distil strain-specific fingerprinting to a minimised scale. This paper examines the utility of microarrays and optical mapping in analysing outbreaks, reviews best practices and the limits of these technologies for pathogen differentiation, and it considers future integration with whole genome sequencing efforts.

Three R's of bacterial evolution: how replication, repair, and recombination frame the origin of species.

The genetic diversity of bacteria results not only from errors in DNA replication and repair but from horizontal exchange and recombination of DNA sequences from similar and disparate species as well. New individuals carrying adaptive changes are thus being spawned constantly among the population at large. When new selection pressures appear, these are the individuals that survive, at the expense of the general population, to forge new populations. Depending on the severity and uniqueness of the selection pressure, this could lead to new speciation. It is becoming more and more evident that, as nucleotide sequences of numerous loci from many bacterial strains continue to amass, horizontal transfer has played a key role in configuring the Escherichia coli chromosome. Here, we examine views, both old and new, for the role of recombination in the evolution of bacterial chromosomes. We present novel phylogenetic evidence for horizontal transfer of three genes involved in DNA replication and repair (mutS, uvrD, and polA). These data reveal a prominent role for horizontal transfer in the evolution of genes known to play a key role in the fidelity of DNA replication and, thus, ultimate survival of the organism. Our data underscore that recombination plays both a diversifying and a homogenizing role in defining the structure of the E. coli genome.

Genetic Bit Analysis: a solid phase method for typing single nucleotide polymorphisms.

A new method for typing single nucleotide polymorphisms in DNA is described. In this method, specific fragments of genomic DNA containing the polymorphic site(s) are first amplified by the polymerase chain reaction (PCR) using one regular and one phosphorothioate-modified primer. The double-stranded PCR product is rendered single-stranded by treatment with the enzyme T7 gene 6 exonuclease, and captured onto individual wells of a 96 well polystyrene plate by hybridization to an immobilized oligonucleotide primer. This primer is designed to hybridize to the single-stranded target DNA immediately adjacent from the polymorphic site of interest. Using the Klenow fragment of E. coli DNA polymerase I or the modified T7 DNA polymerase (Sequenase), the 3' end of the capture oligonucleotide is extended by one base using a mixture of one biotin-labeled, one fluorescein-labeled, and two unlabeled dideoxynucleoside triphosphates. Antibody conjugates of alkaline phosphatase and horseradish peroxidase are then used to determine the nature of the extended base in an ELISA format. This paper describes biochemical features of this method in detail. A semi-automated version of the method, which we call Genetic Bit Analysis (GBA), is being used on a large scale for the parentage verification of thoroughbred horses using a predetermined set of 26 diallelic polymorphisms in the equine genome.

The use of phosphorothioate primers and exonuclease hydrolysis for the preparation of single-stranded PCR products and their detection by solid-phase hybridization.

The effect of phosphorothioate bonds on the hydrolytic activity of the 5'-->3' double-strand-specific T7 gene 6 exonuclease was studied. Double-stranded DNA substrates containing one phosphorothioate residue at the 5' end were found to be hydrolyzed by this enzyme as efficiently as unmodified ones. The enzyme activity was, however, completely inhibited by the presence of four phosphorothioates. On the basis of these results, a method for the conversion of double-stranded PCR products into full-length, single-stranded DNA fragments was developed. In this method, one of the PCR primers contains four phosphorothioates at its 5' end, and the opposite strand primer is unmodified. Following the amplification, the double-stranded product is treated with T7 gene 6 exonuclease. The phosphorothioated strand is protected from the action of this enzyme, whereas the opposite strand is hydrolyzed. When the phosphorothioated PCR primer is 5' biotinylated, the single-stranded PCR product can be easily detected colorimetrically after hybridization to an oligonucleotide probe immobilized on a microtiter plate. We also describe a simple and efficient method for the immobilization of relatively short oligonucleotides to microtiter plates with a hydrophilic surface in the presence of salt.

Isolation of cloned Moloney murine leukemia virus reverse transcriptase lacking ribonuclease H activity.

Retroviral reverse transcriptase possesses DNA polymerase and ribonuclease H (RNase H) activity within a single polypeptide. Chemical or proteolytic treatment of reverse transcriptase has been used in the past to produce enzyme that is missing DNA polymerase activity and retains RNase H activity. It has not been possible to obtain reverse transcriptase that lacks RNase H but retains DNA polymerase activity. We have constructed a novel deletion derivative of the cloned Moloney murine leukemia virus (M-MLV) reverse transcriptase gene, expressed the gene in E. coli, and purified the protein to near homogeneity. The purified enzyme has a fully active DNA polymerase, but has no detectable RNase H activity. These results are consistent with, but do not prove, the conclusion that the DNA polymerase and RNase H activities of M-MLV reverse transcriptase reside within separate structural domains.

Influence on stability in Escherichia coli of the carboxy-terminal structure of cloned Moloney murine leukemia virus reverse transcriptase.

We have cloned and expressed in Escherichia coli a section of the Moloney murine leukemia virus (Mo-MLV) pol gene which includes the entire coding region of mature reverse transcriptase (RT) plus 284 additional base pairs 3' to the coding region (Kotewicz et al., 1985). To prepare cloned Mo-MLV RT as close as possible to authentic RT in structure and activity, the universal terminator sequence GC(TTAA)3GC was introduced at a number of positions inside and outside the RT coding region within 200 nucleotides of its 3' end. The level of RT activity expressed from these constructs varied sevenfold. This variation was found to be directly related to the stability of the RT protein products in the E. coli K-12 strain K802; half-lives varied from 2 to 35 min. The stability of most of the RT proteins was not increased in E. coli K802 lon- cells, with the exception of two, whose half-lives were increased by a factor of two.

Cloning and overexpression of Moloney murine leukemia virus reverse transcriptase in Escherichia coli.

A pBR322-derived expression vector, plasmid pKD1, was constructed containing the strong leftward promoter (pL) of bacteriophage lambda, the ribosome-binding site (RBS) of the cII gene of lambda, and a unique downstream NdeI restriction site for construction of an ATG initiation codon. The section of the pol gene of Moloney murine leukemia virus (M-MLV) that codes for reverse transcriptase (RT) was cloned into the NdeI site of this vector generating the plasmid pRT103. Upon thermal induction, enzymatically active RT was expressed in Escherichia coli [pRT103]. The identity of this activity was confirmed by its template specificity and its sensitivity to inhibition by immunoglobulin G (IgG) prepared against authentic murine RT. RT represented 20% of the newly synthesized protein in these cells 20 min after induction.

Studies on the binding of lambda Int protein to attachment site DNA: identification of a tight-binding site in the P' region.

We have used three approaches to studying the interaction of lambda Int protein with bacteriophage attachment site DNA, POP': location of binding sites by retention of DNA fragments in a filter binding assay, reconstruction of a binding site by DNA synthesis and protection of a binding site from an exonuclease. Retention of restriction fragments on nitrocellulose filters in the presence of Int protein was used to locate binding sites. A high affinity binding site lies in P' between base pairs -6 and +173 from the center of the common core sequence, and low affinity sites are found in the 200 base pair region left of position -6. Reconstruction of the high affinity binding site region from the right using primed DNA synthesis and testing for filter binding in the presence of Int protein shows that sequences sufficient for tight binding of Int protein lie to the right of position +66. When attachment site DNA is protected by bound Int protein against digestion by exonuclease III, four Int dependent protection bands are seen in positions +58, +68, +79 and +88. This can be interpreted either as showing that four Int protein monomers bind to the high affinity region in series, or as evidence for wrapping of the DNA around Int protein, leading to structural changes resembling those occurring to DNA in nucleosomes.