Host Takeover by Bacteriophage T4

National Institute Of Diabetes And Digestive And Kidney Diseases

Investigators

Linked publications & trials

Abstract

It is well-established that in the phage biosphere a large proportion of genes do not have homologs outside of phage. Consequently, these genes may represent a unique repository of genetic information of unknown functions and mechanisms. For bacteriophage T4, most of these genes encode non-essential, early gene products, which are expressed within the first 5 min of infection and are thought to involve phage-host interactions. As such, they may function as anti-bacterial strategies or as phage factors that establish a better environment for infection. One such T4 early gene is motB. We have previously shown that MotB is a DNA-binding protein that binds tightly and nonspecifically to both host and T4 modified DNA and is conserved in a wide range of phages from the Myoviridae family/Tevenvirinae subfamily of the genera Tequatrovirus, Gaprivervirus, Dhakavirus, or Mosigvirus. No MotB homologs have been observed outside of phage. Secondary structure analyses have predicted that MotB contains two functional motifs. The N-terminal half contains a domain related to the Kyprides-Onzonis-Woese (KOW) motif, while the C-terminal half is related to the DNA-binding domain of oligonucleotide/oligosaccharide (OB)-fold. Both motifs have been associated with protein-protein and protein-nucleic acid functions. Our previous work has also indicated that MotB is a nonessential, but highly abundant protein (40,000 to 50,000 molecules per cell during infection), which appears to function as a phage-encoded Nucleoid Associated Protein (NAP). The binding of MotB to DNA results in compacted protein/DNA complexes in vitro that resemble those formed by the bacterial NAP, DNA binding protein from starved cells (Dps), and by yeast cohesin. In vivo expression of plasmid-borne motB yields a condensed host nucleoid in the E. coli K12 strain TOP10F'. Importantly, our detailed studies using the K12 strain indicated that the nucleoid condensation is biologically relevant since it occurs when the level of MotB is similar to that present during T4 infection. Interestingly, in the presence of DNA, MotB co-purifies with the host NAP, H-NS, and in vivo, the nucleoid condensation is accompanied by the dysregulation of genes within the H-NS regulon. Despite their similar genomes, a comparison of E. coli K12 and B strains reveals that they have differences both in sequence and in phenotypes. Previous work has suggested that E. coli B has increased amino acid biosynthesis and decreased amino acid degradation, explaining why it is a good choice for protein expression. K12 appears to be more affected by stress, including exposure to changes in osmolarity and pH or to inhibitory compounds, such as beta-lactam antibiotics. For T4, which can adsorb to E. coli via attachment to either lipopolysaccharide (LPS) or the outer membrane porin OmpC, the lack of ompC in E. coli B makes a notable difference. Our work with MotB has also revealed differences between the E. coli K12 strain TOP10F' vs. the B strain BL21(DE3) after the expression of motB. In the absence of infection, motB expression in the K12 strain results in the arrest of cell growth, while in the B strain, motB expression results in cell death. In addition, our RNA-seq analyses indicated that the heterologous expression of motB in the K12 strain in the absence of infection results in the up-regulation of >500 host genes, of which 74% are within the H-NS regulon; a similar pattern was seen 5 or 10 min after T4 infection. In contrast, heterologous expression of motB in the B strain in the absence of infection results in the upregulation of 75 genes, of which 30% correlate with H-NS regulation. However, in BL21(DE3), many of the upregulated genes are found within the H-NS-repressed lambda lysogen (DE3), which is not present in TOP10F'. This suggests that the induction of the prophage genes perhaps results in cell death before more of the H-NS regulon can be derepressed. In contrast, the effect of MotB on the T4 transcriptome is much more subtle than what we observe for the host even though in vitro MotB condenses host and T4 DNA similarly. Now we have used AlphaFold 2 to obtain a MotB structural model, and we have asked how motB overexpression in the B strain BL21(DE3) affects host and T4 gene expression at 5 min after T4 infection. No T4 transcripts are affected at this time point. However, we observe an upregulation of 30% of the H-NS regulated genes, as we previously observed after expression of motB in the absence of infection in this strain. In addition, surprisingly, we observe the down-regulation of 21 of the 84 chargeable host tRNA genes. To our knowledge, this is the first evidence of a phage gene product being involved in the down-regulation of specific host tRNAs. Our analysis of the affected tRNAs predicts that in many cases, their down-regulation should generate a tRNA pool, which is better suited for translation of the A/U rich T4 transcripts than for translation of the more G/C rich host RNA. As other work has indicated that the RNA-seq mean reads for bacterial tRNAs accurately indicate tRNA abundance and correlate well with codon bias, it was reasonable then to assume that the MotB-induced change in relative tRNA levels might be related to the codon bias of the phage. Consequently, we investigated whether this down-regulation would reshape the tRNA pools toward a more favorable environment for T4 protein expression. Our analysis revealed an association. In several cases, the down-regulation of a host tRNA was associated with the presence of a T4-encoded tRNA that recognizes the same codon. Thus, it appears that the down-regulation of these specific host tRNAs reduces translational competition with the phage tRNA, which has a different sequence from the host isoacceptor and thus, is thought to be better suited for T4. In several other cases, the down-regulation of particular host tRNAs correlated with codons, which T4 uses at a lower frequency than the host, suggesting that the abundance of these tRNAs is lowered because there is less need. To ask if there were differences in T4 protein abundances, we performed mass spectrometry (MS) analyses of the T4 proteins present at 5 min post-infection. As predicted from these tRNA pool changes, we observed changes in the abundance of various T4 proteins, many of which have not been characterized. The work here together with our previous work indicates that MotB functions pleiotropically as a coordinator for infection, helping to set up the host in various ways that will aid the phage. MotB compacts the host chromosomal DNA and condenses the nucleoid. We speculate that it is this change in DNA condensation that then leads to the dysregulation of the H-NS regulon as well as other global changes in host gene expression. In BL21(DE3), derepression of the lambda DE3 operon and other prophage genes occurs, specific host tRNAs are down-regulated, and the levels of certain T4 proteins change, altering the overall infection. It should be noted that as this is occurring, T4 nucleases will also begin to degrade the host DNA, and the host nucleoid will be disrupted by the T4 protein Ndd. Although it is not yet known whether MotB condensation of the DNA affects this degradation/disruption, it seems likely that the DNA compaction would at the very least generate more room for phage replication and assembly. In addition, later in infection the level of MotB actually increases. Given that MotB also binds to T4 modified DNA, it seems likely that this binding will serve an as yet undetermined role related to the phage DNA at this time. Taken together, we speculate that all these various MotB-induced changes optimize the specific host, depending on the strain and perhaps also on the specific conditions, creating a better environment for the phage.