Unravelling the hidden potential of two ORFans from Pseudomonas virus LUZ7

The gram-negative bacterium Pseudomonas aeruginosa is a highly versatile bacterium that is able to thrive in a variety of niches. Due to the selective pressure imposed by the widespread use of antibiotics, this highly adaptive bacterium has become resistant to many of the contemporary antibiotics. The occurrence of multi-drug resistant and even pan-resistant strains of this opportunistic human pathogen, makes it into one of the most dangerous bacterial pathogens today and results in an urgent need for novel antibacterial treatments. Bacterial viruses (bacteriophages) can greatly contribute to developing these new treatments. On one hand they can be used in their entirety as a self-replicating antibacterial agent. On the other hand they encode different proteins that can be used to combat infections. Moreover, phages are evolutionarily optimised to inhibit, activate or otherwise modulate crucial host processes. This is mainly achieved by phage proteins that are expressed early in infection. The study of these early proteins could thus indicate important targets for the development of antimicrobial compounds and provide insights in how to target them. In addition, the modulation of these important processes can have interesting applications in synthetic biology settings.

There is a vast diversity of early expressed proteins among phages, which can thus lead to a plethora of potentially interesting antibacterial applications and biotechnological tools. However, the majority of early genes are ‘ORFans’, which have no sequence similarity with any known gene. This makes their functional characterisation difficult and demands a tailored approach for each ORFan. The aim of this dissertation is to characterise two such ORFans from Pseudomonas virus LUZ7, which were previously identified to be toxic to the cell. The study of these ORFans will allow us to better understand the biology of the phage and to assess their potential in different phage-derived applications.

The first ORFan that is studied here, is gp8 from phage LUZ7. This protein is highly toxic upon expression in P. aeruginosa, causing a full growth stop and lysis of the cell. The protein has been extensively studied in the past with protein-protein interaction screens and functional assays. Despite these efforts, its true function or host target has not yet been identified. This dissertation builds further on that work. Initially, the interaction with the suggested interaction partners (based on yeast two-hybrid), DnaB and RibB was assessed as well as interaction with the toxicity complementing proteins PA0842 and PA3778. However, gp8 could not be confirmed to interact with any of these. Interestingly, an interaction was found with the histidine metabolism enzyme HisC1, which was a hit in a previous yeast two-hybrid screen but initially not confirmed as a true interaction partner. Although this interaction was identified in multiple independent assays, it cannot be directly linked to the toxicity of gp8 as neither HisC1 overexpression or knock-out can complement toxicity. Moreover, the HisC1 enzyme activity did not seem to be affected by gp8 in preliminary assays. Therefore, the true biological function of gp8 remains unclear, as well as its potential role in histidine metabolism. To better understand what effects gp8 may have, it was also investigated from a structural perspective. This allowed the identification of five residues that are important for its toxicity. The protein structure itself has not yet been elucidated, as instability of gp8, likely due to disulphide bridge formation, hindered NMR measurements over longer times. Addition of reducing agents to the protein buffer should enable future structural studies. Based on our results, we suggest that gp8 likely affects other processes aside from the histidine metabolism, or it affects HisC1 in a more subtle way. Possibly, it forms a switch that regulates the usage of energy for either the histidine pathway or the closely connected purine biosynthesis pathway, depending on the need for amino acids or purines. However, additional research will be required to evaluate this hypothesis.

The second studied ORFan is LUZ7 gp14, which is also toxic to the P. aeruginosa cell and causes it to grow in a filamentous way. This protein was previously found to negatively affect the host transcription, but no interaction with any host protein, including its RNA polymerase, was observed. Therefore, it was hypothesised to interact with nucleic acids. In this study, the protein is found to bind a variety of nucleic acids with a preference for ssDNA. The elucidation of the crystal structure at 2.20 Å reveals that it achieves this DNA binding as a dimer with a fold that is novel among single-strand DNA-binding proteins (SSBs). Amino acid substitution and truncation mutants of the protein provided first insights into the binding mechanism. This mechanism seems to be similar to that of other SSB proteins, using aromatic amino acids to form stacking interactions with the DNA bases and positively charged residues to interact with the DNA backbone. To understand the biological role of the SSB in the cell, the presence of SSB proteins in the related phage N4 was compared to that of LUZ7. Interestingly, a homolog  of the SSB (N4 gp2) required for transition from early to middle phase transcription was not present in LUZ7. Gp2 binds the ssDNA region of a melted promoter and recruits the phage-encoded RNA polymerase (RNAPII) to the promoters to activate their transcription. Gp14 (termed Drc ssDNA-binding RNA Polymerase Cofactor) was also found to interact with RNAPII indicating that it might be functionally equivalent to gp2, despite having no homology at the sequence level. However, the way that Drc enacts its function is likely different to gp2, as the latter needs a cofactor to melt promoter DNA, whereas Drc does not seem to require this. Initial experiments indicate that also the promoters themselves are different. A detailed analysis of the presence of Drc and gp2 across the different N4-related genera reveals a clear distinction between clades that encode  Drc and those that encode gp2. Some clades do not seem to encode either of these two proteins indicating that additional SSB proteins might exist. Thus, although the transcription scheme of N4 is generally conserved among the N4-related genera, this research indicates significant differences in regulation of middle gene transcription are present within the group.

The proteins studied here, elegantly demonstrate the hidden potential in ORFan proteins. The study of these two proteins has revealed novel protein functions, a novel protein fold and even insights into core biological processes like the transcriptional progression during phage infection. As these two proteins are merely a very small outtake of the large number of ORFans out there, it is clear that phage ORFans form a treasure trove of functions and tools waiting to be discovered. However, it also demonstrates the tailored approach required to study every ORFan separately, making functional assignment of ORFans one of the major hurdles to be overcome by molecular biologist. Potentially, the systematic determination of protein structure, in combination with the screening-based approaches used to gain initial functional insights will allow a faster functional determination. Moreover as more structures become available, functional predictions are bound to improve. Hence, the large gap between the discovery of massive amounts of novel ORFan sequences and their limited functional assignment might become a little smaller.

The two proteins characterized in this dissertation might find their use in a range of applications. On one hand, gp8 could form an interesting template for development of antibacterials. The targeting of the histidine pathway, with which gp8 interacts, has been proposed in the past to be an interesting target for antibiotics. However, the toxicity mechanism is not yet fully understood and might be caused by the targeting of a different process in the cell. Given the high toxicity of gp8, causing growth stop and lysis, it would be worthwhile to investigate this mechanism further. In a synthetic biology setting, gp8 can be interesting for modulation of the histidine pathway to improve the production of amino acids. Also for this application, additional insights into the exact effect of gp8 on the metabolic pathway will be needed. On the other hand, Drc will have little application for development of antibiotics as it interacts with the nucleic acids. However, it can be very useful in the design of synthetic biology circuits in combination with its cognate RNA polymerase (RNAPII). Due to its ability to bind promoter-containing DNA that undergoes torsional stress, it might serve as a sensor of processes that affect DNA torsion. In addition, the requirement of two RNA polymerase subunits and an SSB allow the construction of (bio)logic AND gates that depend on additional inputs.

The current study has laid bare a small piece of phage ingenuity encoded in their ORFans, with much more still to be discovered. It will be up to human ingenuity to transform these ORFans, originally used to benefit the phage, into something that benefits us all.