Investigation of type III effector proteins in enteric pathogens using Hidden Markov Models

Bachmann, Nathan Lloyd (2013). Investigation of type III effector proteins in enteric pathogens using Hidden Markov Models PhD Thesis, School of Chemistry & Molecular Biosciences, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s4098606_PhD_final.pdf Thesis full text application/pdf 21.09MB 0
Author Bachmann, Nathan Lloyd
Thesis Title Investigation of type III effector proteins in enteric pathogens using Hidden Markov Models
School, Centre or Institute School of Chemistry & Molecular Biosciences
Institution The University of Queensland
Publication date 2013
Thesis type PhD Thesis
Supervisor Scott Beatson
Nicola Petty
Total pages 140
Total colour pages 28
Total black and white pages 112
Language eng
Subjects 0605 Microbiology
Formatted abstract
The Type III Secretion System (T3SS) is a key mechanism of bacterial pathogenesis. The T3SS translocates effector proteins into eukaryote cells that then manipulate cellular functions to enhance the survival of the bacteria. Despite the rapid increase in available genomic data it is often surprisingly difficult to undertake comprehensive genome-scale analyses of effectors due to propagation of errors by automated annotation and the sheer number of different effector types. Here we describe the development of an annotation tool called EffectorFAM and use it to investigate the effector repertoire of E. coli and Salmonella using publicly available and newly sequenced genomes.

EffectorFAM is an online database of effector-specific profile Hidden Markov Models (HMMs): statistical models of sequence alignments that can be used to find distantly related sequences. We grouped and aligned 839 experimentally confirmed effector sequences into 146 families. Forty families were already represented by existing HMMs and a further 106 manually curated HMMs were constructed. EffectorFAM was used to catalog the effector protein repertoire of 381 complete and incomplete publicly available Escherichia coli genomes.

Five out of the 381 E. coli genomes encode E3 ubiquitin ligase (E3UL) mimics of the NEL subtype that have been characterized previously in Shigella species. E3UL mimics are capable of manipulating biological processes such as cell death and cell cycle progression. The five E. coli strains are dispersed on four lineages and each encode up to nine NEL-domain E3UL mimics, with the genes located either in prophage regions or on a Shigella-like plasmid. The NEL-domain E3UL mimics are 99 % identical to IpaH effectors secreted by Shigella to down-regulate the host inflammatory immune response. Notably, none of these E. coli strains appear to harbor a functional T3SS. However, one of the strains carries a plasmid that is 94-100 % identical across 26 % of the Shigella virulence plasmid pCP301. The Shigella plasmid encodes the Mxi-Spa T3SS, which is absent from the E. coli plasmid. Therefore it is possible that E. coli 53638 may have once encoded the Mxi-Spa T3SS and that it was recently lost.

EffectorFAM was used to search the genomes of 55 non-O157 Shigatoxigenic E. coli (STEC) genomes that were sequenced using Illumina HiSeq 2000 platform. Twenty-six of the STEC genomes encoded a T3SS on the Locus of Enterocyte Effacement (LEE) pathogenicity island and up to 36 effectors on exchangeable effector loci (EELs) that are distributed throughout the  chromosome. The other 29 STEC genomes are LEE-negative and are also missing the EEL encoded effectors. The findings suggest that the presence of the LEE is necessary for the bacteria to maintain the EEL encoded effectors. It is hypothesize that a functional T3SS provides a selective pressure for maintaining effectors and that the loss of the system will result in deletion of the effector genes. However, the presence of the E3UL-like effectors encoded in T3SS-negative E. coli genomes suggest that they may have recently lost a T3SS and that the effectors will eventually be deleted or that the effectors are being secreted by another mechanism such the flagella export apparatus.

EffectorFAM was applied to the genome of Salmonella Virchow SVQ-1 that was sequenced using Roche 454 GS-FLX. In this genome we identified a novel prophage that encodes the effector SopE, a key Salmonella virulence factor. This prophage shares very little similarity to the SopE prophage found in other Salmonella serovars. It is suggested that there is a small degree of parallel evolution in Salmonella serovars as it appears that S. Virchow acquired the SopE gene from a different source compared to other Salmonella. The studies describe in this thesis demonstrate that EffectorFAM is capable of making de novo observations and streamlining the investigation of effectors in large data sets. EffectorFAM will be available as a persistent online tool that will assist with annotation and research on Type III effectors in whole bacterial genomes.
Keyword Type III secretion system
Effector Protein
Hidden Markov model
Escherichia coli (E. coli)
Salmonella Virchow

Citation counts: Google Scholar Search Google Scholar
Created: Fri, 10 Jan 2014, 14:14:52 EST by Nathan Bachmann on behalf of Scholarly Communication and Digitisation Service