Final Report Summary - PNEUMO-CELL (Regulation of pneumolysin in the human pathogen Streptococcus pneumoniae: A single cell approach)
Introduction of the research question
Streptococcus pneumoniae (the pneumococcus) is a major human pathogen and the last decades have seen the emergence and spread of pneumococcal strains with multiple antibiotic resistances posing a serious threat to human health (Kadioglu et al., 2008; Scott et al., 2008). The pneumococcus is carried asymptomatically in up to 60% of the (healthy) population (Mitchell, 2003). However, S. pneumoniae can also switch to a pathogenic state and can cause a number of respiratory tract and life-threatening invasive diseases such as otitis media, bacteraemia, meningitis and pneumonia (Kadioglu et al., 2008). In order for the pneumococcus to become virulent, it needs to breach the host tissue barriers to gain access to the submucosa and the blood. This invasiveness depends on the expression of so-called virulence factors. A major S. pneumoniae virulence factor is pneumolysin, which is a cytolysin encoded by the ply gene. Pneumolysin is a well characterised toxin produced in the cytoplasm of S. pneumoniae and its multiple biological activities towards eukaryotic host cell function and the immune system are well known (Hammerschmidt, 2007). ply codes for a 470 amino acid protein and genomic analysis showed that its sequence is highly conserved among clinical isolates (Hammerschmidt, 2007). Cryo-electron microscopy studies have revealed that pneumolysin forms pores in cholesterol containing lipid bilayers (Tilley et al., 2005). In high concentrations, pneumolysin inserts into the host membrane and causes lysis of the cell. At lower concentrations, pneumolysin directly stimulates the inflammatory response of host cells. Importantly, S. pneumoniae mutated for ply are less virulent in animal models (Paton, 1996).
Most S. pneumoniae virulence factors are deposited on the outside of the cell, such as the choline binding proteins and the capsule. Pneumolysin is exceptional in this sense, because it can only reach the host cell after lysis of the producer. This poses an interesting paradox because if all cells simultaneously die to release pneumolysin, then the population would fail to propagate. It was recently suggested that only part of a culture within clonal populations of bacterial pathogens undergo lysis as a form of self-destructive cooperation (Ackermann et al., 2008). Since most toxins induce inflammation in the host, it decreases competition by co-inhabitants of the same niche. The cells that do not lyse benefit from the sacrificial action of its siblings by making use of the empty niches cleared by the host immune response (Ackermann et al., 2008). The molecular mechanisms for self-destructive cooperation however are largely unknown.
Whether pneumolysin is only expressed in part of the clonal population, and/or whether pneumolysin is only released in part of the clonal population remains to be tested. Bulk studies indicate that cell-to-cell killing by competent cells towards non-competent cells is partly responsible for release of pneumolysin (Guiral et al., 2005), but whether all cells that are killed by competent cells express and release ply is unknown.
According to the latest genome annotation of the commonly used S. pneumoniae D39 strain, ply is the last gene of an operon containing three other genes of unknown function. Little information is available how and when ply is expressed (Hammerschmidt, 2007). It is also unknown where the promoter(s) are located that drive transcription of ply and whether expression is constitutive or regulated.
To address this research question, a single cell approach is essential. The few studies that employ (fluorescence) microscopy on Streptococci have mainly been performed by immunostaining and not using the green fluorescent protein (GFP) (Buist et al., 2006).
Development of single cell analytical tools for S. pneumoniae
Measuring gene expression by fluorescence microscopy in live cells is instrumental in studying phenotypic variation and gene expression at the single cell level. In particular, the use of the Green Fluorescent Protein (GFP) has allowed the study of gene-dynamics of model bacteria such as Bacillus subtilis and Escherichia coli (Ozbudak et al., 2002; Elowitz et al., 2002). Maturation of GFP requires post-translational oxidation and it is likely that this has hindered employment of GFP in microaerophiles, such as S. pneumoniae (Tsien, 1998; Acebo et al., 2000). Using a fast-folding variant of GFP we generated a GFP-reporter system that can be efficiently used to examine protein localization in S. pneumoniae (Eberhardt et al., 2009; Minnen et al., 2011). However, this GFP-variant is not well-suited to analyse single cell gene expression because of its limited intrinsic brightness compared to other, newly identified variants (see below). Furthermore, to be able to compare to a control gene, a second genetically-encoded fluorophore such as a Red Fluorescent Protein (RFP) is required. Therefore, we constructed a battery of different GFP and RFP variants, with different codon-optimization strategies. Out of this screen we identified bright superfolding GFP (GFPspn) and RFP (RFPspn) variants that efficiently express in S. pneumoniae. In addition to these new tools, we have successfully developed time-lapse microscopy for S. pneumoniae (de, I et al., 2011) and have setup flow cytometry assays. These new tools now open up the doors for examining the regulation of phenotypic variation in this organism and the regulation of pneumolysin in specific.
Preliminary data on pneumolysin gene expression and localization.
At least twenty bacterial species can produce cytolysins, however pneumolysin can be distinguished by the absence of a signal peptide for secretion. Using a C-terminal GFP fusion, we could show that all cells in the population express pneumolysin (Beilharz et al., unpublished). Part of pneumolysin is secreted and attached to the cell wall (Beilharz et al., unpublished). Preliminary time-lapse microscopy experiments revealed that cells produce more pneumolysin before lysis occurs but the molecular mechanism underlying this phenomenon is currently unknown. It is tempting to speculate that, when cells are going to lyse more pneumolysin is produced, to release it at the end.
How pneumolysin is regulated at the transcriptional level is poorly understood. Therefore we performed RNA-Seq analysis (Yuzenkova et al., under revision) which identified the transcriptional start site and showed that ply is part of a five gene operon. Random transposon mutagenesis identified five genes that are likely involved in regulating ply expression since they demonstrated a reduced activity of the promoter of ply (Beilharz et al., unpublished). Further research is required to characterize these mutants in more detail.
Publications and other impact
The research carried out during this project has been described in seven publications, including two in one of the leading microbiology journals Molecular Microbiology and one in the top journal PNAS. Moreover, the work has been presented at various international research meetings. The reintegration grant has also contributed to the establishment of an independent research position for the fellow at the University of Groningen where part of the research will be continued. It also allowed the fellow to acquire additional external funds and awards such as the prestigious Dutch Vidi grant, the ERC-starting grant and the election as EMBO Young Investigator.
Streptococcus pneumoniae (the pneumococcus) is a major human pathogen and the last decades have seen the emergence and spread of pneumococcal strains with multiple antibiotic resistances posing a serious threat to human health (Kadioglu et al., 2008; Scott et al., 2008). The pneumococcus is carried asymptomatically in up to 60% of the (healthy) population (Mitchell, 2003). However, S. pneumoniae can also switch to a pathogenic state and can cause a number of respiratory tract and life-threatening invasive diseases such as otitis media, bacteraemia, meningitis and pneumonia (Kadioglu et al., 2008). In order for the pneumococcus to become virulent, it needs to breach the host tissue barriers to gain access to the submucosa and the blood. This invasiveness depends on the expression of so-called virulence factors. A major S. pneumoniae virulence factor is pneumolysin, which is a cytolysin encoded by the ply gene. Pneumolysin is a well characterised toxin produced in the cytoplasm of S. pneumoniae and its multiple biological activities towards eukaryotic host cell function and the immune system are well known (Hammerschmidt, 2007). ply codes for a 470 amino acid protein and genomic analysis showed that its sequence is highly conserved among clinical isolates (Hammerschmidt, 2007). Cryo-electron microscopy studies have revealed that pneumolysin forms pores in cholesterol containing lipid bilayers (Tilley et al., 2005). In high concentrations, pneumolysin inserts into the host membrane and causes lysis of the cell. At lower concentrations, pneumolysin directly stimulates the inflammatory response of host cells. Importantly, S. pneumoniae mutated for ply are less virulent in animal models (Paton, 1996).
Most S. pneumoniae virulence factors are deposited on the outside of the cell, such as the choline binding proteins and the capsule. Pneumolysin is exceptional in this sense, because it can only reach the host cell after lysis of the producer. This poses an interesting paradox because if all cells simultaneously die to release pneumolysin, then the population would fail to propagate. It was recently suggested that only part of a culture within clonal populations of bacterial pathogens undergo lysis as a form of self-destructive cooperation (Ackermann et al., 2008). Since most toxins induce inflammation in the host, it decreases competition by co-inhabitants of the same niche. The cells that do not lyse benefit from the sacrificial action of its siblings by making use of the empty niches cleared by the host immune response (Ackermann et al., 2008). The molecular mechanisms for self-destructive cooperation however are largely unknown.
Whether pneumolysin is only expressed in part of the clonal population, and/or whether pneumolysin is only released in part of the clonal population remains to be tested. Bulk studies indicate that cell-to-cell killing by competent cells towards non-competent cells is partly responsible for release of pneumolysin (Guiral et al., 2005), but whether all cells that are killed by competent cells express and release ply is unknown.
According to the latest genome annotation of the commonly used S. pneumoniae D39 strain, ply is the last gene of an operon containing three other genes of unknown function. Little information is available how and when ply is expressed (Hammerschmidt, 2007). It is also unknown where the promoter(s) are located that drive transcription of ply and whether expression is constitutive or regulated.
To address this research question, a single cell approach is essential. The few studies that employ (fluorescence) microscopy on Streptococci have mainly been performed by immunostaining and not using the green fluorescent protein (GFP) (Buist et al., 2006).
Development of single cell analytical tools for S. pneumoniae
Measuring gene expression by fluorescence microscopy in live cells is instrumental in studying phenotypic variation and gene expression at the single cell level. In particular, the use of the Green Fluorescent Protein (GFP) has allowed the study of gene-dynamics of model bacteria such as Bacillus subtilis and Escherichia coli (Ozbudak et al., 2002; Elowitz et al., 2002). Maturation of GFP requires post-translational oxidation and it is likely that this has hindered employment of GFP in microaerophiles, such as S. pneumoniae (Tsien, 1998; Acebo et al., 2000). Using a fast-folding variant of GFP we generated a GFP-reporter system that can be efficiently used to examine protein localization in S. pneumoniae (Eberhardt et al., 2009; Minnen et al., 2011). However, this GFP-variant is not well-suited to analyse single cell gene expression because of its limited intrinsic brightness compared to other, newly identified variants (see below). Furthermore, to be able to compare to a control gene, a second genetically-encoded fluorophore such as a Red Fluorescent Protein (RFP) is required. Therefore, we constructed a battery of different GFP and RFP variants, with different codon-optimization strategies. Out of this screen we identified bright superfolding GFP (GFPspn) and RFP (RFPspn) variants that efficiently express in S. pneumoniae. In addition to these new tools, we have successfully developed time-lapse microscopy for S. pneumoniae (de, I et al., 2011) and have setup flow cytometry assays. These new tools now open up the doors for examining the regulation of phenotypic variation in this organism and the regulation of pneumolysin in specific.
Preliminary data on pneumolysin gene expression and localization.
At least twenty bacterial species can produce cytolysins, however pneumolysin can be distinguished by the absence of a signal peptide for secretion. Using a C-terminal GFP fusion, we could show that all cells in the population express pneumolysin (Beilharz et al., unpublished). Part of pneumolysin is secreted and attached to the cell wall (Beilharz et al., unpublished). Preliminary time-lapse microscopy experiments revealed that cells produce more pneumolysin before lysis occurs but the molecular mechanism underlying this phenomenon is currently unknown. It is tempting to speculate that, when cells are going to lyse more pneumolysin is produced, to release it at the end.
How pneumolysin is regulated at the transcriptional level is poorly understood. Therefore we performed RNA-Seq analysis (Yuzenkova et al., under revision) which identified the transcriptional start site and showed that ply is part of a five gene operon. Random transposon mutagenesis identified five genes that are likely involved in regulating ply expression since they demonstrated a reduced activity of the promoter of ply (Beilharz et al., unpublished). Further research is required to characterize these mutants in more detail.
Publications and other impact
The research carried out during this project has been described in seven publications, including two in one of the leading microbiology journals Molecular Microbiology and one in the top journal PNAS. Moreover, the work has been presented at various international research meetings. The reintegration grant has also contributed to the establishment of an independent research position for the fellow at the University of Groningen where part of the research will be continued. It also allowed the fellow to acquire additional external funds and awards such as the prestigious Dutch Vidi grant, the ERC-starting grant and the election as EMBO Young Investigator.