A thermostable DNA primase-polymerase from a mobile genetic element involved in defence against environmental DNA

Abstract

Primase-polymerases (Ppol) are one of the few enzymes able to start DNA synthesis on ssDNA templates. The role of Thermus thermophilus HB27 Ppol, encoded along a putative helicase (Hel) within a mobile genetic element (ICETh2), has been studied. A mutant lacking Ppol showed no effects on the replication of the element. Also, no apparent differences in the sensitivity to DNA damaging agents and other stressors or morphological changes in the mutant cells were detected. However, the mutants lacking Ppol showed an increase in two to three orders of magnitude in their transformation efficiency with plasmids and genomic DNA acquired from the environment (eDNA), independently of its origin and G + C content. In contrast, no significant differences with the wild type were detected when the cells received the DNA from other T. thermophilus partners in conjugation-like mating experiments. The similarities of this behaviour with that shown by mutants lacking the Argonaute (ThAgo) protein suggests a putative partnership Ppol-ThAgo in the DNA-DNA interference mechanism of defence, although other eDNA defence mechanisms independent of ThAgo cannot be discarded.

Fig 1

Genetic context of ppol within MGE elements of Thermales. Selected genome maps of homologues to the ICETh2 structure defined in T. thermophilus HB27 strain in (Baquedano et al., 2020), employing available data in the KEGG and NCBI databases. ICETh2 is an MGE mobilized by a set of highly conserved integrase (in pale pink) and an excisionase (in red) followed by a tRNA‐Val at the 3′ termini. This 11.3 Kbp ICETh2 harbours a putative replication module containing several hypothetical proteins, Ppol (in green) and a putative helicase gene (in dark green); a regulation module including a peptidase (in yellow), putative transposases (in blue navy) and other ORFs coding for hypothetical proteins including helicase‐like motifs (in light green) which may contribute to the physiology of this MGE. Flanking this ICE element there is a 47 nt direct repeat (att sites, in orange) identical to the 3′ end of the tRNA, which show high homology among all strains. A type II toxin (purple)‐ antitoxin (pink) (HicA‐B type) is detected in certain strains. Loci of the Ppol and the helicase ‐like genes as well as the integrase are annotated in each of the ICEs identified. Sequence identity percentages to ICEth2 genes are given in the corresponding arrows. Highly similar structures to that of ICETh2 are found in T.aquaticus (Taq), T.oshimai (Theos) and T. thermophilus SG0.5 JP17‐16 (SG0.5) and with a lower synteny in T. thermophilus TMY (T_TMY), HC11 (Tth_HC11) and AA2 strains, the later showing two identical ICEs (AA2‐20 and AA2‐29). In these strains, the helicase is replaced by an hypothetical protein containing a P‐loop‐NTPase on its N‐termini (in darker green). Besides, several strains belonging to the Meiohermus genus: M. silvanus (Mesil), M. taiwanensis (Mtai) and two strains of M. ruber, DSM1279 (Mrub) and K649 (Mre) also harbour Ppol homologues within an ICE‐like structure. Search was done on fully assembled genomes available on NCBI databases, selecting those that fulfilled the identity and coverage established (40% identity, 60% coverage, within a 30 kbp distance).

Fig 2

An insertional deletion of ppol increases the excision of ICETh2. Relative quantities detected by qPCR of the integrated ICETh2 attL2 form (Chr), the circular excised attI2 form (Circ), and the ICE‐free genomic locus attB2 (Scar). The ratios are shown for the wild type and the Δppol::kat mutant subjected to UV treatment respect to untreated controls. First replicate of wild‐type cultures have been normalized to 1 on each form. Asterisks indicate significant statistical differences (*P‐value < 0.05 and ***P‐value < 0.01) between wild type and mutant ratios under the same conditions for the detection of the same ICE form. Samples have been normalized relative to the RNA polymerase alpha subunit gene (n = 5).

Fig 3

An insertional deletion of Hel increases the proportion of ICE‐free form. Relative quantities detected by qPCR of Chr, Circ, and Scar forms. First replicate of wild type cultures has been normalized to 1 on each form. Asterisks indicate significant statistical differences (P‐value < 0.05) between wild type and mutant ratios under the same conditions for the detection of the same ICE form. Samples have been normalized relative to RNA polymerase alpha subunit, DNA Polymerase III and 16S genes (n = 3).

Fig 4

Absence of both Ppol and ThAgo sharply increases ICETh2 excision. Relative quantities detected by qPCR of Chr, Circ, and Scar forms. First replicate of wild‐type cultures has been normalized to 1 on each form. Asterisks indicate significant statistical differences (*P‐value < 0.05 and **P‐value < 0.01) between wild type and mutant ratios for the detection of the same ICE form. Samples have been normalized relative to RNA polymerase alpha subunit, DNA Polymerase III and 16S genes (n = 3).

Fig 5

Transformation efficiency of Ppol mutants. The 100 ng of replicative plasmids conferring Kn^R (pMoTK) or Hyg^R (pMoTH), or the suicide plasmid pyrEK (pyrEK) conferring Kn^R, were used to transform 1.5 × 10⁸ cells of T. thermophilus HB27 and its derivative mutants lacking TthAgo (Δago), Ppol (Δppol) or both (Δago‐Δppol). Transformation is represented as the number of colonies on selection plates at 60°C in three independent experiments. Asterisks indicate significant statistical differences (*P‐value < 0.05 and **P‐value < 0.002).

Fig 6

Ppol functions as a barrier against isogenic chromosomal DNA in transformation, but not in transjugation. In transjugation, two wild type strains (black bars), one labelled in the genome at the pyrE locus with Kn^R and the other labelled at locus TTC0313 (coding for putative ferredoxin‐nitrite reductase) with Hyg^R marker, as well as two ∆ppol derivatives labelled identically with Kn^R and Hyg^R (lined bar) were used in mating experiments. For transformation experiments, a wild type strain labelled in the genome at the pyrE locus with a Kn^R marker (black bar) and a ∆ppol mutant identically labelled (lined bar) were transformed with 20 ng of genomic DNA isolated from ∆ppol mutant labelled with a Hyg^R marker at the TTC0313 locus. Transjugation and transformation frequencies were calculated as the ratio between the number of double resistant colonies (Kn^R plus Hyg^R) and Kn^R colonies. Frequencies are the average of four independent experiments, and error bars correspond to standard deviation. The asterisk indicates significant statistical differences (P‐value < 0.05).