Gil Segal

Genome-Scale Identification of Legionella pneumophila Effectors Using a Machine Learning Approach

A large number of highly pathogenic bacteria utilize secretion systems to translocate effector proteins into host cells. Using these effectors, the bacteria subvert host cell processes during infection. Legionella pneumophila translocates effectors via the Icm/Dot type-IV secretion system and to date, approximately 100 effectors have been identified by various experimental and computational techniques. Effector identification is a critical first step towards the understanding of the pathogenesis system in L. pneumophila as well as in other bacterial pathogens. Here, we formulate the task of effector identification as a classification problem: each L. pneumophila open reading frame (ORF) was classified as either effector or not. We computationally defined a set of features that best distinguish effectors from non-effectors. These features cover a wide range of characteristics including taxonomical dispersion, regulatory data, genomic organization, similarity to eukaryotic proteomes and more. Machine learning algorithms utilizing these features were then applied to classify all the ORFs within the L. pneumophila genome. Using this approach we were able to predict and experimentally validate 40 new effectors, reaching a success rate of above 90%. Increasing the number of validated effectors to around 140, we were able to gain novel insights into their characteristics. Effectors were found to have low G+C content, supporting the hypothesis that a large number of effectors originate via horizontal gene transfer, probably from their protozoan host. In addition, effectors were found to cluster in specific genomic regions. Finally, we were able to provide a novel description of the C-terminal translocation signal required for effector translocation by the Icm/Dot secretion system. To conclude, we have discovered 40 novel L. pneumophila effectors, predicted over a hundred additional highly probable effectors, and shown the applicability of machine learning algorithms for the identification and characterization of bacterial pathogenesis determinants.

The response regulator PmrA is a major regulator of the icm/dot type IV secretion system in Legionella pneumophila and Coxiella burnetii.

Legionella pneumophila and Coxiella burnetii have been shown to utilize the icm/dot type IV secretion system for pathogenesis and recently a large number of icm/dot‐translocated substrates were identified in L. pneumophila. Bioinformatic analysis has revealed that 13 of the genes encoding for L. pneumophila‐translocated substrates and five of the C. burnetii icm/dot genes, contain a conserved regulatory element that resembles the target sequence of the PmrA response regulator. Experimental analysis which included the construction of a L. pneumophila pmrA deletion mutant, intracellular growth analysis, comparison of gene expression between L. pneumophila wild type and the pmrA mutant, construction of mutations in the PmrA conserved regulatory element, controlled expression studies as well as mobility shift assays, demonstrated the direct relation between the PmrA regulator and the expression of L. pneumophila icm/dot‐translocated substrates and several C. burnetii icm/dot genes. Furthermore, genomic analysis identified 35 L. pneumophila and 68 C. burnetii unique genes that contain the PmrA regulatory element and few of these genes from L. pneumophila were found to be new icm/dot‐translocated substrates. Our results establish the PmrA regulator as a fundamental regulator of the icm/dot type IV secretion system in these two bacteria.

Computational modeling and experimental validation of the Legionella and Coxiella virulence-related type-IVB secretion signal

Legionella and Coxiella are intracellular pathogens that use the virulence-related Icm/Dot type-IVB secretion system to translocate effector proteins into host cells during infection. These effectors were previously shown to contain a C-terminal secretion signal required for their translocation. In this research, we implemented a hidden semi-Markov model to characterize the amino acid composition of the signal, thus providing a comprehensive computational model for the secretion signal. This model accounts for dependencies among sites and captures spatial variation in amino acid composition along the secretion signal. To validate our model, we predicted and synthetically constructed an optimal secretion signal whose sequence is different from that of any known effector. We show that this signal efficiently translocates into host cells in an Icm/Dot-dependent manner. Additionally, we predicted in silico and experimentally examined the effects of mutations in the secretion signal, which provided innovative insights into its characteristics. Some effectors were found to lack a strong secretion signal according to our model. We demonstrated that these effectors were highly dependent on the IcmS-IcmW chaperons for their translocation, unlike effectors that harbor a strong secretion signal. Furthermore, our model is innovative because it enables searching ORFs for secretion signals on a genomic scale, which led to the identification and experimental validation of 20 effectors from Legionella pneumophila, Legionella longbeachae, and Coxiella burnetii. Our combined computational and experimental methodology is general and can be applied to the identification of a wide spectrum of protein features that lack sequence conservation but have similar amino acid characteristics.

Characterization of a Legionella pneumophila relA Insertion Mutant and Roles of RelA and RpoS in Virulence Gene Expression

To investigate the involvement of RelA in the regulation of Legionella pneumophila virulence, a deletion substitution was constructed in the relA gene. The relA knockout resulted in an undetectable level of ppGpp in the cells during the stationary phase, but the original level was restored when the relA gene product was supplied on a plasmid. The effect of the relA mutation was examined with two systems that are known to be expressed during the stationary phase in L. pneumophila. Pigment production was found to be dependent on the relA gene product, and only one-half as much pigment was produced by the relA mutant as by the wild-type strain. Flagellum gene expression was also found to be dependent on the relA gene product, as determined with a flaA::lacZ fusion. However, the relA gene product was found to be dispensable for intracellular growth both in HL-60-derived human macrophages and in the protozoan host Acanthamoeba castellanii. To determine the involvement of the relA gene product in expression of L. pneumophila genes required for intracellular growth (icm/dot genes), nine icm::lacZ fusions were constructed, and expression of these fusions in the wild-type strain was compared with their expression in relA mutant strains. Expression of only one of the icm::lacZ fusions was moderately reduced in the relA mutant strain. Expression of the nine icm::lacZ fusions was also examined in a strain containing an insertion in the gene that codes for the stationary-phase sigma factor RpoS, and similar results were obtained. We concluded that RelA is dispensable for intracellular growth of L. pneumophila in the two hosts examined and that both RelA and RpoS play minor roles in L. pneumophila icm/dot gene expression.

Regulation of Heat-Shock Response in Bacteria

Stress response in bacteria is essential for effective adaptation to changes in the environment, as well as to the changes in the physiological state of the bacterial culture itself. This response is mediated by global regulatory mechanisms affecting several pathways. It now appears that these regulatory mechanisms operate by transcriptional control, translational control, and proteolysis. One example to be discussed extensively is the heat-shock response. In Escherichia coli, where it has been studied initially and most extensively, the expression of the heat-shock operon is transcriptionally controlled by the employment of the heat-shock transcription factor sigma 32, that recognizes specific heat-shock promoters. Later studies indicated that in most bacteria the control of the major heat-shock genes is much more complicated, and involves additional--or alternative--control channels. These regulatory elements will be reviewed looking at the groE and dnaK operons. These operons, coding for the bacterial equivalent of Hsp10+60 and Hsp70, respectively, contain in many bacteria a conserved regulatory inverted repeat (IR = CIRCE), and are transcribed either by the vegetative sigma factor--sigma 70--or by a sigma 32-like factor. The IR functions at the DNA level as a repressor binding site and also controls the half life of the transcript. In addition, in Agrobacterium tumefaciens there also exists a system for mRNA processing that involves a temperature-controlled cleavage of the groE transcript.

The LetA-RsmYZ-CsrA regulatory cascade, together with RpoS and PmrA, post-transcriptionally regulates stationary phase activation of Legionella pneumophila Icm/Dot effectors

Legionella pneumophila utilize the Icm/Dot type‐IV secretion system to translocate effector proteins into host cells. Some of these effectors were shown before to be regulated at the transcriptional level by the PmrAB and CpxRA two‐component systems. In addition, the stationary phase‐related regulators LetA and CsrA, which are both members of the same post‐transcriptional regulatory cascade, were shown to be involved in L. pneumophila virulence. In this report, we identified two small non‐coding RNAs which are part of the LetA‐CsrA regulatory cascade and three effector‐encoding genes which are directly controlled by this regulatory system. We found that the small non‐coding RNAs RsmY and RsmZ, were upregulated by LetA at stationary phase, and relieve the repression of CsrA from its target genes. The three effector‐encoding genes were found to be post‐transcriptionally upregulated at stationary phase and to contain CsrA regulatory elements that were found to be essential for their stationary phase activation. In addition, rsmY and rsmZ were found to be regulated by the RpoS sigma‐factor and the csrA encoding gene was found to be regulated by PmrA. Our results demonstrate that L. pneumophila effectors are regulated at both the transcriptional and the post‐transcriptional levels by a complicated regulatory network.

Identification of CpxR as a Positive Regulator of icm and dot Virulence Genes of Legionella pneumophila

To date, 24 Legionella pneumophila genes (icm and dot genes) have been shown to be required for intercellular growth and host cell killing. A previous report indicated that the regulation of these genes is complicated and probably involves several regulatory proteins. In this study, a genetic screen performed in Escherichia coli identified the CpxR response regulator as an activator of the L. pneumophila icmR gene. Construction of an L. pneumophila cpxR insertion mutant showed that the expression of icmR is regulated by CpxR. In addition, a conserved CpxR binding site (GTAAA) was identified in the icmR regulatory region and L. pneumophila His-tagged CpxR protein was shown to bind to the icmR regulatory region using a mobility shift assay. Besides its dramatic effect on the icmR level of expression, the CpxR regulator was also found to affect the expression of the icmV-dotA and icmW-icmX operons, but to a lesser extent. The role of CpxA, the cognate sensor kinase of CpxR, was also examined and its effect on the icmR level of expression was found to be less pronounced than the effect of CpxR. The RpoE sigma factor, which was shown to coregulate genes together with CpxR, was examined as well, but it did not influence icm and dot gene expression. In addition, when the cpxR mutant strain, in which the expression of the icmR gene was dramatically reduced, and the cpxA and rpoE mutant strains were examined for their ability to grow inside Acanthamoeba castellanii and HL-60-derived human macrophages, no intracellular growth defect was observed. This study presents the first evidence for a direct regulator (CpxR) of an icm-dot virulence gene (icmR). The CpxR regulator together with other regulatory factors probably concerts with the expression of icm and dot genes to result in successful infection.

The Response Regulator CpxR Directly Regulates Expression of Several Legionella pneumophila icm/dot Components as Well as New Translocated Substrates

Legionella pneumophila has been shown to utilize the icm/dot type IV secretion system for pathogenesis. This system was shown to be composed of icm/dot complex components and accessory proteins, as well as a large number of translocated substrates. Bioinformatic analysis of the regulatory regions of all the genes revealed that several icm/dot genes, as well as two genes encoding icm/dot translocated substrates, contain the conserved CpxR regulatory element, a regulator that has been shown previously to control the expression of the icmR gene. An experimental analysis, which included a comparison of gene expression in a L. pneumophila wild-type strain and gene expression in a cpxR deletion mutant, construction of mutants with mutations in the CpxR conserved regulatory elements, controlled expression studies, and mobility shift assays, demonstrated the direct relationship between the CpxR regulator and the expression of the genes. Furthermore, genomic analysis identified nine additional genes that contain a putative CpxR regulatory element; five of these genes (two legA genes and three ceg genes) were suggested previously to be putative icm/dot translocated substrates. The three ceg genes identified, which were shown previously to contain a putative PmrA regulatory element, were found here to be regulated by both CpxR and PmrA. The other six genes (two legA genes and four new genes products were found to be regulated by CpxR. Moreover, using the CyaA translocation assay, these nine gene products were found to be translocated into host cells in an Icm/Dot-dependent manner. Our results establish that the CpxR regulator is a fundamental regulator of the icm/dot type IV secretion system in L. pneumophila.

A Legionella effector acquired from protozoa is involved in sphingolipids metabolism and is targeted to the host cell mitochondria

Legionella pneumophila infects alveolar macrophages and protozoa through establishment of an intracellular replication niche. This process is mediated by bacterial effectors translocated into the host cell via the Icm/Dot type IV secretion system. Most of the effectors identified so far are unique to L. pneumophila; however, some of the effectors are homologous to eukaryotic proteins. We performed a distribution analysis of many known L. pneumophila effectors and found that several of them, mostly eukaryotic homologous proteins, are present in different Legionella species. In‐depth analysis of LegS2, a L. pneumophila homologue of the highly conserved eukaryotic enzyme sphingosine‐1‐phosphate lyase (SPL), revealed that it was most likely acquired from a protozoan organism early during Legionella evolution. The LegS2 protein was found to translocate into host cells using a C‐terminal translocation domain absent in its eukaryotic homologues. LegS2 was found to complement the sphingosine‐sensitive phenotype of a Saccharomyces serevisia SPL‐null mutant and this complementation depended on evolutionary conserved residues in the LegS2 catalytic domain. Interestingly, unlike the eukaryotic SPL that localizes to the endoplasmic reticulum, LegS2 was found to be targeted mainly to host cell mitochondria. Collectively, our results demonstrate the remarkable adaptations of a eukaryotic protein to the L. pneumophila pathogenesis system.

Genomic analysis of 38 Legionella species identifies large and diverse effector repertoires

Infection by the human pathogen Legionella pneumophila relies on the translocation of ∼300 virulence proteins, termed effectors, which manipulate host cell processes. However, almost no information exists regarding effectors in other Legionella pathogens. Here we sequenced, assembled and characterized the genomes of 38 Legionella species and predicted their effector repertoires using a previously validated machine learning approach. This analysis identified 5,885 predicted effectors. The effector repertoires of different Legionella species were found to be largely non-overlapping, and only seven core effectors were shared by all species studied. Species-specific effectors had atypically low GC content, suggesting exogenous acquisition, possibly from the natural protozoan hosts of these species. Furthermore, we detected numerous new conserved effector domains and discovered new domain combinations, which allowed the inference of as yet undescribed effector functions. The effector collection and network of domain architectures described here can serve as a roadmap for future studies of effector function and evolution.