Evolution of pH-sensitive transcription termination during adaptation to repeated long-term starvation

Abstract

Fluctuating environments that consist of regular cycles of co-occurring stress are a common challenge faced by cellular populations. For a population to thrive in constantly changing conditions, an ability to coordinate a rapid cellular response is essential. Here, we identify a mutation conferring an arginine-to-histidine (Arg to His) substitution in the transcription terminator Rho. The rho R109H mutation frequently arose in E. coli populations experimentally evolved under repeated long-term starvation conditions, during which feast and famine result in drastic environmental pH fluctuations. Metagenomic sequencing revealed that populations containing the rho mutation also possess putative loss-of-function mutations in ydcI, which encodes a recently characterized transcription factor associated with pH homeostasis. Genetic reconstructions of these mutations show that the rho allele confers a plastic alkaline-induced reduction of Rho function that, when found in tandem with a ΔydcI allele, leads to intracellular alkalinization and genetic assimilation of Rho mutant function. We further identify Arg to His substitutions at analogous sites in rho alleles from species originating from fluctuating alkaline environments. Our results suggest that Arg to His substitutions in global regulators of gene expression can serve to rapidly coordinate complex responses through pH sensing and shed light on how cellular populations across the tree of life use environmental cues to coordinate rapid responses to complex, fluctuating environments.

Keywords: adaptation; alkaline stress; changing environment; gene regulation; phenotypic plasticity.

Fig 1:. Repeated long-term starvation selects for mutations in rho and ydcI.

(A) Number of populations containing polymorphic (dashed line) or fixed (solid line) mutations in the rho (red) or ydcI (blue) loci over 900 days of evolution to RLTS conditions. (B) Number of fixed mutations in rho (top half) and ydcI (bottom half) genes indicating timing at which each particular mutation arose over 900 days. Purple lines indicate alleles co-occurring within populations. The genic location of affected (C) rho (orange: R109H, purple: G63S, green: K105E) and (D) ydcI residues (red: R10H, blue: R30C, purple: F74Y, magenta: D83G, cyan: S91P, green: V181A, orange: L198Q) and (E) structural locations of affected rho and (F) ydcI residues identified in populations over 900 days of evolution to RLTS conditions. Only the A and B chains of the YdcI homo tetramer are visualized for simplicity. Structural locations colored in green indicate H-T-H domain residues. NTD: N-terminal domain, CTD: C-terminal domain, NHB: N-terminal helix bundle, CSD: cold shock domain, RBD: RNA binding domain, DBD: DNA binding domain, RD: regulatory domain, HTH: helix-turn-helix.

Fig 2:. In vitro Rho Assay Reveals Diminished Activity of Rho^R109H at pH 9.

The in vitro activities of Rho⁺ (WT) and Rho^R109H (R109H) were assessed at pH 7 and pH 9 in three ways: (A) efficiency of ATP-dependent unwinding of an RNA:DNA duplex containing a 5’-single-stranded rut overhang (graph: helicase activity after 20 min), (B) ability to terminate transcription at the λtR1 terminator, and (C) ability to bind to RNA. A schematic of the DNA template used in the transcription termination assay is shown above gel images in (B). Chart above plots in (C) contains the calculated average K_d for WT and R109H proteins. Additional information regarding binding parameters can be found in Table 1.

Fig 3:. rho R109H/ΔydcI cells have increased intracellular pH

The intracellular pH (pHi) of WT, rho R109H, ΔydcI, and rho R109H/ΔydcI single cells were measured in buffers with increasing pH values. These measurements were derived from a ladder control (Control) containing WT cells treated with nigericin, yielding the pHi and environmental pH equal. (A). Representative microscopic images of single cells from tested strains in each pH buffered solution with colors of cells corresponding to their calculated pHi. Each image is cropped to 5.5 × 5.5 μm; a 1 μm scale bar is included in the bottom right image. Remaining microscopic images can be found in supplementary files. (B). pHi measurements derived from these images across the environmental pH conditions. Lightly shaded data points represent single cell replicates with the color denoting the pHi. Error bars display 95% confidence intervals. (C). Plot displaying the pHi range of each strain in tested conditions using data from panel (B) illustrates an upward shift in pHi range in the rho R109H/ΔydcI cells.

Fig 4:. Benefits of rho R109H and ΔydcI alleles vary throughout the RLTS regime.

(A). Frequency of final genotypes identified in populations after 12 replicate cultures of WT, ΔydcI, and rho R109H (initial genotypes) were grown in RLTS conditions for 100 days. A subscript of p denotes the allele is polymorphic as indicated by the Sanger sequence. (B). Pairwise competitions between WT ancestor and mutant strains when co-cultured for 1 or 14 days in unbuffered LB and LB buffered at pH 9. Colored bars indicate the median selection rate (s) for each mutant strain relative to the WT with single points representing each of three biological replicates. (C). Median CFU/mL of WT and mutant strains over 14 days of growth. Lightly shaded data points represent each of three biological replicates. Shaded vertical bars indicate the day of the first initial growth rebound post-death phase. Error bars display 95% confidence intervals.

Fig 5:. Amino acid alignment of Rho N-terminal RNA binding domain in laboratory strains and natural isolates.

Alignments of Rho sequences from selected RefSeq genomes (top) and the mutated R109H E. coli clone with natural non-E. coli isolates (bottom) that encode a Histidine residue at the analogous position. Natural non-E. coli isolates were identified with PSI-BLAST.