Besides a resurrection tale that I'll relate to in the second part, this post will be about anti-defense systems of plasmids as an intrinsic part of the seemingly endless arms race between conjugative elements and their hosts. Still, I am not inclined to act as war correspondent, and certainly not at a time when the war in Ukraine has been going on unabated for one year, with no realistic prospect of an end in the near future. What to do? I trust you will accept that I spare you here a reasoning of why I consider the war metaphor that is prevalent in discussions of the conflicts between bacteriophages and plasmids and their bacterial hosts to be inappropriate – and save this for a later occasion. So I will, grudgingly, stick to the common parlance and turn to the "anti‑defense" systems of plasmids.
To explain what these anti‑defense systems of conjugative plasmids are about, I need to dive into how they propagate (no deep dive, just snorkeling). In a donor cell, a conjugative plasmid is usually present at a copy number close to 1 and replicates from its "vegetative origin," oriV, in synchrony with the cell's chromosome. For propagation, it uses a relaxase, TraM, that "nicks" its dsDNA at a site termed oriT (origin of transfer), binds covalently to the 5'-end of the nick, and directs the peeling single strand to the membrane-bound Type IV secretion system (T4SS). The remaining successively single-stranded plasmid is returned to its double-stranded form. I addressed in a previous post, Pictures Considered #55: Conjugation, and some pairing issues, the questions that still exist in understanding the actual conjugation process, the transfer of the single strand into the recipient cell.
What is known, however, is that after being threaded through the cell envelope of the recipient cell (outer membrane, peptidoglycan, inner membrane) the single-stranded plasmid reaches the cytoplasm with the 5'-end first, TraM attached. It serves as a template for lagging-strand synthesis by the replication system of the host cell, analogous to rolling circle replication of single-stranded phage lambda DNA. The restoration of the circular form of the plasmid is carried out by TraM still bound at the 5'-end as a reverse reaction of the "nicking" reaction.
During the temporary single-strand stage, but even more so after "replicative repair" to the double strand, the plasmid DNA is, like "incoming" phage DNA, exposed to all the mechanisms that the recipient cell musters to maintain the integrity of its own genetic material. Among these mechanisms are, and in varying degrees of repertoire completeness in different bacteria, restriction-modification systems (RM), SOS response for RecA activation, and CRISPR-Cas, to mention just the most notorious. In their extensive and expansive travels through Domain Bacteria, plasmids (and bacteriophages) have picked up, tested, and, if useful, kept a variety of genes that allow them to maintain the integrity of their own genome when challenged with the "genome integrity maintenance" of host cells. These are the 'anti-defense systems of plasmids' I wanted to explain, and examples are (mostly) small proteins that block the activity of restriction enzymes, methyltransferases, inhibitors of SOS response induction, and anti-CRISPR proteins (acr).
Researchers consistently found a number of anti-defense genes preferentially located in the "leading regions" (~10–15 kb) – the single-stranded DNA with TraM relaxase covalently bound to its 5'-end, which first reaches the cytoplasm of the recipient cell during conjugation – of several well-studied conjugative plasmids, for example the ardA gene encoding an antirestriction protein in E. coli plasmid ColIb-P9 and its homolog ardA in pKM101. In a recent study, published now as a preprint, Samuel & Burstein (2023) explored whether an enrichment of anti-defense genes is also found in the leading regions of other plasmids.
The authors screened the relevant databases (NCBI, EBI) for the genomes/sequences of (1.) conjugative plasmids, (2.) integrative conjugative elements (ICEs), the plasmid equivalent of prophages (chromosomally integrated phage genomes), and (3.) so called "mobilizable plasmids." Such mobilizable plasmids can be transferred by hitchhiking the conjugation machinery of a co‑residing conjugative element and are considered decay products of ICEs that retain the origin of transfer (oriT ) adjacent to a relaxase gene (traM ). They collected altogether 1,554 plasmid sequences, and 17,505 sequences of ICEs and mobilizable plasmids (see here for the distribution of 13,882 non-redundant sequences across the bacterial phyla, 4,613 of which could not be mapped to the tree). They note that their collection is biased because genomes/sequences of human pathogens and their plasmids are vastly over-represented in the databases.
Next they screened their sequence collection by software-aided homology searches for known anti-defense systems, that is, systems encoding anti-CRISPR, anti-restriction, anti-SOS, and other known or presumed anti-defense proteins. They measured the relative abundance of anti-defense genes at each position of the plasmid sequences with respect to the location of the oriT sequence. The frequency of anti-defense genes at each position revealed that the leading region of these elements is highly enriched for anti-defense genes. Specifically, the first 30 ORFs of the leading region were significantly enriched with anti-defense genes of the categories anti-CRISPRs, anti-restriction, and SOS-inhibition (see Figure 1).
The authors suggest a model of the protection provided to plasmids by diverse anti-defense genes encoded in the leading region of conjugative plasmids. I take it from their legend to Figure 6 of their paper (here Figure 2):
Owing to ssDNA promoters, the anti-defense genes can be expressed at the very early stages of transfer to a recipient cell. During this phase, the bacterial immune response recruits its defense systems to prevent the entry of the transferred foreign DNA. Anti-CRISPRs encoded on the plasmid can inhibit CRISPR-Cas systems; SOS-inhibitors, such as PsiB protein, can repress the cell SOS-response by preventing the activation of RecA thus inhibiting the cleavage of LexA, an SOS response transcriptional repressor; Single-stranded binding (SSB) protein are known to be involved in the SOS response inhibition mechanism, and may protect the transferred ssDNA from host nucleases. Methyltransferases (MTase), methylating the ssDNA can prevent recognition by the host restriction-modification (RM) systems; Anti-restriction proteins can prevent DNA cleavage by RM systems; and Antitoxins can neutralize host TA systems. In the top-right panel, a schematic genetic organization of a conjugative plasmid. The four main functional gene groups are colored: propagation (blue), adaptation (purple), replication (green), and the anti-defense genes (red) within the stability/establishment region (orange).
Note that the tiny hairpins on top of the line representing ssDNA in Figure 2 indicate Frpo-type promoters (ssDNA promoters as explained in part two). Note also that the line representing the ssDNA is the template strand, so that the 5'→3' transcription triggered by the Frpo-type promoters is "leftward" for most genes in the leading region (see here or here for textbook illustrations showing the conventional "rightward" transcription with the template strand being the lower strand of the dsDNA. I go into this in such detail here because I know from lectures that many students are initially confused by 5'→3' strand orientations if they do not correspond exactly to the western reading direction, from left to right).
Samuel & Burstein (2023) are very clear about how their work could be taken forward in a purposeful way:
The robustness of the anti-defense gene location and Frpo promoters demonstrated in this study could be used to develop new approaches to identify previously unknown oriT sequences based on the location of the relaxase gene and the anti-defense islands. Similarly, the combination of oriT and anti-defense islands can be used to determine the leading region in small plasmids lacking a relaxase gene. These will allow studying additional plasmids and specifically investigating early expressed genes in small plasmids encoding only an oriT. Such small plasmids are important to study, as they have been shown to carry numerous antimicrobial resistance genes (ARGs) and make up at least half of all plasmids.
The statistically significant accumulation of known anti-defense genes in the leading regions of the studied plasmid/ICE sequences also suggests that other genes in this region that could not yet be functionally characterized are also anti-defense genes, the closer investigation of which could lead to an expansion of the repertoire (a notion I borrow from Pinilla-Redondo et al. (2020)).