Genetic Selection of an Errorprone Variant of Bacteriophage T7 RNA Polymerase

Studies with several RNA viruses have demonstrated their exceptionally high replication error rates that confer genetic flexibility, while resulting in substantial decreases of infectivity [35, 36]. Enhanced mutagenesis might even result in virus extinction, as predicted from virus entry into an ''error catastrophe'' [37, 38]. We are interested in studying and utilizing the link between error-prone polymerases and virus viability and started our search for an error-prone variant with the RNA polymerase of bacteriophage T7 (that is, the transcriptase of a DNA virus). We therefore developed a stringent positive selection scheme that rewarded inaccurate transcription by a T7 RNAP mutant with the survival of bacteria [39]. This genetic selection scheme involved a system of two compatible plasmids which coupled mutant polymerase genes to the essential, but inactivated, tetracycline resistance gene (Figure 4.3.2).

Feedback coupling was achieved by introducing a T7 promoter upstream of the tetracycline resistance gene (tet). This resulted in the exclusive dependence of re-

Bacteriophage

Fig. 4.3.2. Scheme for the genetic selection of an error-prone RNA polymerase variant based on the reversion of antibiotic resistance. The ''selection plasmid'' carries a mutant tetracycline resistance gene under the control of a T7 promoter. This construct renders the host cell sensitive to tetracycline (tet) unless an active and error-prone variant of T7 RNA polymerase is present. This enzyme would produce randomly altered transcripts among which one or more might exist that encode restored resistance activity (for example, as a consequence of random second-site mutation) [39].

Fig. 4.3.2. Scheme for the genetic selection of an error-prone RNA polymerase variant based on the reversion of antibiotic resistance. The ''selection plasmid'' carries a mutant tetracycline resistance gene under the control of a T7 promoter. This construct renders the host cell sensitive to tetracycline (tet) unless an active and error-prone variant of T7 RNA polymerase is present. This enzyme would produce randomly altered transcripts among which one or more might exist that encode restored resistance activity (for example, as a consequence of random second-site mutation) [39].

sistance expression on the presence of T7 RNA polymerase. Substitution of a single amino acid of the tet fully inactivated the resistance which is mediated by a hydrophobic, membrane-associated efflux pump [40]. The mutation (Y100 ! P) is located within a periplasmic region of the membrane-bound protein which previously had been identified to be critical for the performance of resistance [41]. Our positive selection scheme was based on the finding that functional interactions exists between the N- and C-terminal domains of the tetracycline efflux pump [42]. As a consequence of this interaction a deleterious mutation in one domain of the protein might be suppressed by some second-site mutation within the other domain. For selection, E. coli cells harboring the ''selection plasmid'', were first co-transformed with a library of ~106 plasmid-encoded, randomly mutated T7 RNAP gene variants, and then grown in liquid media containing 40% of the standard tetracycline dose. Surviving bacteria emerged immediately and were shown to express a single variant of T7 RNAP with three amino acid substitutions, F11 ! L, C515 ! Y, and T613 ! A. In contrast to the wild-type enzyme that catalyzes tran scription of the T7 genome with a nucleotide substitution error rate of <6 x 10~5 only, a 20-fold-increased error rate of 1.25 x 10~3 (no preference for a certain type of mutation, or a certain sequence context) was observed for the isolated variant. Beyond that we observed that bacteria which constitutively express the selected polymerase variant are less efficiently infected and lysed by bacteriophage T7 -suggesting that the infecting phage utilizes the host-encoded error-prone enzyme.

Regarding the efficiency of our selection scheme, we can conclude that our expectations have been completely fulfilled, because the theoretical error rate of ~10~3 (compensation of one missense mutation within ca. 1300 base pairs of tet) was reached. Attempts to obtain a further increase of the transcriptional error rate by compensating two missense mutations failed, however - surviving bacteria represented artifacts harboring either no T7 RNAP gene or non-functional polymerase fragments.

0 0

Post a comment